Informacja

Drogi użytkowniku, aplikacja do prawidłowego działania wymaga obsługi JavaScript. Proszę włącz obsługę JavaScript w Twojej przeglądarce.

English (EN)·Polski (PL)·Yкраїнський (UK)
Przejdź do strony domowej biblioteki Centralna Biblioteka Wojskowa Link otwiera się w nowym oknie Logo biblioteki Prolib Integro - strona głównaCentralna Biblioteka Wojskowa

Wyszukujesz frazę "feature extraction" wg kryterium: Temat

  Lista filtrów

Dostawca treści

Wyrównaj
    Wyświetlanie 1-69 z 69
    Tytuł:
    A new approach to image-based recommender systems with the application of heatmaps maps
    Autorzy:
    Woldan, Piotr
    Duda, Piotr
    Cader, Andrzej
    Laktionov, Ivan
    Powiązania:
    https://bibliotekanauki.pl/articles/2201330.pdf
    Data publikacji:
    2023
    Wydawca:
    Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych
    Tematy:
    feature extraction
    recommender system
    heatmap
    Opis:
    One of the fundamental issues of modern society is access to interesting and useful content. As the amount of available content increases, this task becomes more and more challenging. Our needs are not always formulated in words; sometimes we have to use complex data types like images. In this paper, we consider the three approaches to creating recommender systems based on image data. The proposed systems are evaluated on a real-world dataset. Two case studies are presented. The first one presents the case of an item with many similar objects in a database, and the second one with only a few similar items
    Źródło:
    Journal of Artificial Intelligence and Soft Computing Research; 2023, 13, 2; 63--72
    2083-2567
    2449-6499
    Pojawia się w:
    Journal of Artificial Intelligence and Soft Computing Research
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Klasyfikacja stanów przedkrytycznych
    Classification of pre-critical states
    Autorzy:
    Topczewska, M.
    Frischmuth, K.
    Powiązania:
    https://bibliotekanauki.pl/articles/154431.pdf
    Data publikacji:
    2012
    Wydawca:
    Stowarzyszenie Inżynierów i Techników Mechaników Polskich
    Tematy:
    klasyfikacja
    ekstrakcja cech
    classification
    feature extraction
    Opis:
    Praca zawiera przykład klasyfikacji danych rzeczywistych opisujących sygnały niekrytyczne, przedkrytyczne i krytyczne. Celem jest rozpoznanie stanów niebezpiecznych tak wcześnie jak to możliwe. Ze względu na brak separowalności liniowej danych w celu separacji klas użyto klasyfikacji hierarchicznej z cięciami za pomocą klasyfikatorów liniowych oraz z podejściem one-versus-rest z wyróżnioną klasą sygnałów bezpiecznych. W wyniku ośmiu cięć uzyskano ostateczny podział przestrzeni skutkujący odseparowaniem klasy sygnałów bezpiecznych od podejrzanych, tj. przedkrytycznych i krytycznych oraz dający najmniejszą liczbę błędnie sklasyfikowanych obiektów z klasy sygnałów niekrytycznych.
    The paper presents an application of classification methods to time-continuous signals (1). Signals with values that exceed a certain critical maximum are called dangerous or critical, otherwise we speak about normal or routine operation of the system under consideration, Fig. 1. The problem is to recognize pre-critical states, i.e. states preceding the actual dangerous ones, and that as early as possible. False negative classifications may have very serious consequences, while false positive verdicts cause expensive but unnecessary counter-measures. As pre-processing, the input signals are characterized by a number of features, which form sequences of vector data, indexed by the cycle number (2). In a first stage, suspicious feature vectors are selected, from which in a second sweep unlikely candidates are removed. The focus of the present paper is this second stage, i.e. the distinction between actual pre-critical and the harmless routine states among the suspicious states, indicated in the first stage by a certain preliminary test. The choice of features and the logic behind the preliminary test are beyond our present scope. Let it suffice to say that the first step is a combination of Principal Component Analysis and some statistical test, and that it is very effective but unspecific in the application at hand.For the real-world data we used to develop the method, it turned out that the obtained feature vectors were linearly non-separable. For that reason a hierarchical approach was applied, where in several steps linear cuts (4,5) of the one-versus-rest type were performed in order to single out the true pre-critical states. For the example under consideration, in eight iterations separation between pre-critical and non-pre-critical ones was achieved. We succeeded to keep the number of wrong negatives at zero while reducing the number of wrong positives to a fraction of the starting value, established by the preliminary test, Fig. 3, 4, 5. The final sensitivity, for the given data set, is 100%, and the achieved specificity is at 93.15%. Numerical experiments, using nonlinear classifiers on much larger data sets, are under way. The present aim is to find an optimal set of features and a one-step criterion which further improves the quality of the classification.
    Źródło:
    Pomiary Automatyka Kontrola; 2012, R. 58, nr 10, 10; 872-875
    0032-4140
    Pojawia się w:
    Pomiary Automatyka Kontrola
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Human Detection in Thermal Images Using Low-level Features
    Autorzy:
    Budzan, S.
    Powiązania:
    https://bibliotekanauki.pl/articles/114333.pdf
    Data publikacji:
    2015
    Wydawca:
    Stowarzyszenie Inżynierów i Techników Mechaników Polskich
    Tematy:
    human detection
    infrared
    feature extraction
    HOG
    Opis:
    In this work the human detection method in infrared has been presented. The proposed solution focuses on the use low-level features and detecting parts of the human body. Low–level processing is based on modified HOG (Histogram of Oriented Gradients) algorithm. First, the only squared cells have been used, also calculation of the gradient has been improved. Next, the model of the head from the dataset IR (Infra Red) images has been created, also the model of the human body. Finally, the probability matrix has been examined using minimal distance classifier. The novelty of the proposed solution focuses on the combination of the pixel-gradient and body parts processing, also three stage classification process (head modelling, human modelling and classifier), which has been proposed to reduce the false detection. The experiments were performed on self-created IR images database, which contains images with most of the possible difficult situations such as overlapped people, different pose, small and high resolution of the people. The performance of the proposed algorithm was evaluated using Precision and Recall quality measure.
    Źródło:
    Measurement Automation Monitoring; 2015, 61, 6; 191-194
    2450-2855
    Pojawia się w:
    Measurement Automation Monitoring
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Phase Autocorrelation Bark Wavelet Transform (PACWT) Features for Robust Speech Recognition
    Autorzy:
    Majeed, S. A.
    Husain, H.
    Samad, S. A.
    Powiązania:
    https://bibliotekanauki.pl/articles/177326.pdf
    Data publikacji:
    2015
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    speech recognition
    feature extraction
    phase autocorrelation
    wavelet transform
    Opis:
    In this paper, a new feature-extraction method is proposed to achieve robustness of speech recognition systems. This method combines the benefits of phase autocorrelation (PAC) with bark wavelet transform. PAC uses the angle to measure correlation instead of the traditional autocorrelation measure, whereas the bark wavelet transform is a special type of wavelet transform that is particularly designed for speech signals. The extracted features from this combined method are called phase autocorrelation bark wavelet transform (PACWT) features. The speech recognition performance of the PACWT features is evaluated and compared to the conventional feature extraction method mel frequency cepstrum coefficients (MFCC) using TI-Digits database under different types of noise and noise levels. This database has been divided into male and female data. The result shows that the word recognition rate using the PACWT features for noisy male data (white noise at 0 dB SNR) is 60%, whereas it is 41.35% for the MFCC features under identical conditions.
    Źródło:
    Archives of Acoustics; 2015, 40, 1; 25-31
    0137-5075
    Pojawia się w:
    Archives of Acoustics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A survey of methods for 3D model feature extraction
    Autorzy:
    Hlavaty, T.
    Skala, V.
    Powiązania:
    https://bibliotekanauki.pl/articles/119225.pdf
    Data publikacji:
    2003
    Wydawca:
    Polskie Towarzystwo Geometrii i Grafiki Inżynierskiej
    Tematy:
    feature extraction
    retrieval systems
    ekstrakcja cech
    systemy wyszukiwania
    Opis:
    This paper deals with problems that are related to a feature extraction from 3D objects. The main aim of the feature extraction is to describe a shape of 3D object by a feature vector. Then the elements of this feature vector characterize the shape of the own 3D objects and they can serve as a key in searching for similar models. In this paper are introduced current methods for the feature extraction of 3D models and their classification. These methods are based on different mathematical background and according to that they are separated into several groups.
    Źródło:
    Journal Biuletyn of Polish Society for Geometry and Engineering Graphics; 2003, 13; 5-8
    1644-9363
    Pojawia się w:
    Journal Biuletyn of Polish Society for Geometry and Engineering Graphics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Overcoming Overfitting Challenges with HOG Feature Extraction and XGBoost-Based Classification for Concrete Crack Monitoring
    Autorzy:
    Barkiah, Ida
    Sari, Yuslena
    Powiązania:
    https://bibliotekanauki.pl/articles/27311909.pdf
    Data publikacji:
    2023
    Wydawca:
    Polska Akademia Nauk. Czasopisma i Monografie PAN
    Tematy:
    HOG
    XGBoost
    classification
    feature extraction
    concrete crack monitoring
    Opis:
    This study proposes a method that combines Histogram of Oriented Gradients (HOG) feature extraction and Extreme Gradient Boosting (XGBoost) classification to resolve the challenges of concrete crack monitoring. The purpose of the study is to address the common issue of overfitting in machine learning models. The research uses a dataset of 40,000 images of concrete cracks and HOG feature extraction to identify relevant patterns. Classification is performed using the ensemble method XGBoost, with a focus on optimizing its hyperparameters. This study evaluates the efficacy of XGBoost in comparison to other ensemble methods, such as Random Forest and AdaBoost. XGBoost outperforms the other algorithms in terms of accuracy, precision, recall, and F1-score, as demonstrated by the results. The proposed method obtains an accuracy of 96.95% with optimized hyperparameters, a recall of 96.10%, a precision of 97.90%, and an F1-score of 97%. By optimizing the number of trees hyperparameter, 1200 trees yield the greatest performance. The results demonstrate the efficacy of HOG-based feature extraction and XGBoost for accurate and dependable classification of concrete fractures, overcoming the overfitting issues that are typically encountered in such tasks.
    Źródło:
    International Journal of Electronics and Telecommunications; 2023, 69, 3; 571--577
    2300-1933
    Pojawia się w:
    International Journal of Electronics and Telecommunications
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A comparative study of CNN, LSTM, BiLSTM, AND GRU architectures for tool wear prediction in milling processes
    Autorzy:
    Zegarra, Fabio C.
    Vargas-Machuca, Juan
    Coronado, Alberto M.
    Powiązania:
    https://bibliotekanauki.pl/articles/28407329.pdf
    Data publikacji:
    2023
    Wydawca:
    Wrocławska Rada Federacji Stowarzyszeń Naukowo-Technicznych
    Tematy:
    tool wear
    feature extraction
    preprocessing
    recurrent neural network
    Opis:
    Accurately predicting machine tool wear requires models capable of capturing complex, nonlinear interactions in multivariate time series inputs. Recurrent neural networks (RNNs) are well-suited to this task, owing to their memory mechanisms and capacity to construct highly complex models. In particular, LSTM, BiLSTM, and GRU architectures have shown promise in wear prediction. This study demonstrates that RNNs can automatically extract relevant information from time series data, resulting in highly precise wear models with minimal feature engineering. Notably, this approach avoids the need for excessively large window sizes of data points during model training, which would increase model complexity and processing time. Instead, this study proposes a procedure that achieves low prediction errors with window sizes as small as 100 data points. By employing Bayesian hyperparameter optimization and two preprocessing techniques (detrend and offset), RMSE errors consistently fall below 10. A key difference in this study is the use of boxplots to provide a better representation of result variability, as opposed to solely reporting the best values. The proposed approach matches more complex state of-the-art. methods and offers a powerful tool for wear prediction in engineering applications.
    Źródło:
    Journal of Machine Engineering; 2023, 23, 4; 122--136
    1895-7595
    2391-8071
    Pojawia się w:
    Journal of Machine Engineering
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Brain-computer interface as measurement and control system The review paper
    Autorzy:
    Rak, R. J.
    Kołodziej, M.
    Majkowski, A.
    Powiązania:
    https://bibliotekanauki.pl/articles/221747.pdf
    Data publikacji:
    2012
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    EEG
    brain-computer interface
    feature extraction
    feature selection
    measurement and control
    Opis:
    In the last decade of the XX-th century, several academic centers have launched intensive research programs on the brain-computer interface (BCI). The current state of research allows to use certain properties of electromagnetic waves (brain activity) produced by brain neurons, measured using electroencephalographic techniques (EEG recording involves reading from electrodes attached to the scalp - the non-invasive method - or with electrodes implanted directly into the cerebral cortex - the invasive method). A BCI system reads the user's "intentions" by decoding certain features of the EEG signal. Those features are then classified and "translated" (on-line) into commands used to control a computer, prosthesis, wheelchair or other device. In this article, the authors try to show that the BCI is a typical example of a measurement and control unit.
    Źródło:
    Metrology and Measurement Systems; 2012, 19, 3; 427-444
    0860-8229
    Pojawia się w:
    Metrology and Measurement Systems
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    An improved algorithm for feature extraction from a fingerprint fuzzy image
    Autorzy:
    Surmacz, K.
    Saeed, K.
    Rapta, P.
    Powiązania:
    https://bibliotekanauki.pl/articles/174451.pdf
    Data publikacji:
    2013
    Wydawca:
    Politechnika Wrocławska. Oficyna Wydawnicza Politechniki Wrocławskiej
    Tematy:
    biometrics
    feature extraction
    fingerprints
    minutiae
    Gabor filter
    feature vector
    image processing
    Opis:
    Proper fingerprint feature extraction is crucial in fingerprint-matching algorithms. For good results, different pieces of information about a fingerprint image, such as ridge orientation and frequency, must be considered. It is often necessary to improve the quality of a fingerprint image in order for the feature extraction process to work correctly. In this paper we present a complete (fully implemented) improved algorithm for fingerprint feature extraction, based on numerous papers on this topic. The paper describes a fingerprint recognition system consisting of image preprocessing, filtration, feature extraction and matching for recognition. The image preprocessing includes normalization based on mean value and variation. The orientation field is extracted and Gabor filter is used to prepare the fingerprint image for further processing. For singular point detection, the Poincaré index with a partitioning method is used. The ridgeline thinning is presented and so is the minutia extraction by CN algorithm. The paper contains the comparison of obtained results to the other algorithms.
    Źródło:
    Optica Applicata; 2013, 43, 3; 515-527
    0078-5466
    1899-7015
    Pojawia się w:
    Optica Applicata
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    An application of machine learning methods to cutting tool path clustering and rul estimation in machining
    Autorzy:
    Zegarra, Fabio C.
    Vargas-Machuca, Juan
    Roman-Gonzalez, Avid
    Coronado, Alberto M.
    Powiązania:
    https://bibliotekanauki.pl/articles/28407324.pdf
    Data publikacji:
    2023
    Wydawca:
    Wrocławska Rada Federacji Stowarzyszeń Naukowo-Technicznych
    Tematy:
    feature extraction
    k-means clustering
    time series
    unsupervised learning
    Opis:
    Machine learning has been widely used in manufacturing, leading to significant advances in diverse problems, including the prediction of wear and remaining useful life (RUL) of machine tools. However, the data used in many cases correspond to simple and stable processes that differ from practical applications. In this work, a novel dataset consisting of eight cutting tools with complex tool paths is used. The time series of the tool paths, corresponding to the three-dimensional position of the cutting tool, are grouped according to their shape. Three unsupervised clustering techniques are applied, resulting in the identification of DBA-k-means as the most appropriate technique for this case. The clustering process helps to identify training and testing data with similar tool paths, which is then applied to build a simple two-feature prediction model with the same level of precision for RUL prediction as a more complex four-feature prediction model. This work demonstrates that by properly selecting the methodology and number of clusters, tool paths can be effectively classified, which can later be used in prediction problems in more complex settings.
    Źródło:
    Journal of Machine Engineering; 2023, 23, 4; 5--17
    1895-7595
    2391-8071
    Pojawia się w:
    Journal of Machine Engineering
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Determination of the Optimal Threshold Value and Number of Keypoints in Scale Invariant Feature Transform-based Copy-Move Forgery Detection
    Autorzy:
    Isnanto, R. Rizal
    Zahra, Ajub Ajulian
    Santoso, Imam
    Lubis, Muhammad Salman
    Powiązania:
    https://bibliotekanauki.pl/articles/227299.pdf
    Data publikacji:
    2020
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    forgery
    Gaussian noise
    feature extraction
    pattern matching
    Euclidean distance
    Opis:
    The copy-move forgery detection (CMFD) begins with the preprocessing until the image is ready to process. Then, the image features are extracted using a feature-transform-based extraction called the scale-invariant feature transform (SIFT). The last step is features matching using Generalized 2 Nearest-Neighbor (G2NN) method with threshold values variation. The problem is what is the optimal threshold value and number of keypoints so that copy-move detection has the highest accuracy. The optimal threshold value and number of keypoints had determined so that the detection n has the highest accuracy. The research was carried out on images without noise and with Gaussian noise.
    Źródło:
    International Journal of Electronics and Telecommunications; 2020, 66, 3; 561-569
    2300-1933
    Pojawia się w:
    International Journal of Electronics and Telecommunications
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A Classification Method Related to Respiratory Disorder Events Based on Acoustical Analysis of Snoring
    Autorzy:
    Wang, Can
    Peng, Jianxin
    Zhang, Xiaowen
    Powiązania:
    https://bibliotekanauki.pl/articles/176601.pdf
    Data publikacji:
    2020
    Wydawca:
    Polska Akademia Nauk. Czasopisma i Monografie PAN
    Tematy:
    acoustical analysis
    feature extraction
    support vector machine
    snoring sound
    Opis:
    Acoustical analysis of snoring provides a new approach for the diagnosis of obstructive sleep apnea hypopnea syndrome (OSAHS). A classification method is presented based on respiratory disorder events to predict the apnea-hypopnea index (AHI) of OSAHS patients. The acoustical features of snoring were extracted from a full night’s recording of 6 OSAHS patients, and regular snoring sounds and snoring sounds related to respiratory disorder events were classified using a support vector machine (SVM) method. The mean recognition rate for simple snoring sounds and snoring sounds related to respiratory disorder events is more than 91.14% by using the grid search, a genetic algorithm and particle swarm optimization methods. The predicted AHI from the present study has a high correlation with the AHI from polysomnography and the correlation coefficient is 0.976. These results demonstrate that the proposed method can classify the snoring sounds of OSAHS patients and can be used to provide guidance for diagnosis of OSAHS.
    Źródło:
    Archives of Acoustics; 2020, 45, 1; 141-151
    0137-5075
    Pojawia się w:
    Archives of Acoustics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    On the predictive power of meta-features in OpenML
    Autorzy:
    Bilalli, B.
    Abelló, A.
    Aluja-Banet, T.
    Powiązania:
    https://bibliotekanauki.pl/articles/331086.pdf
    Data publikacji:
    2017
    Wydawca:
    Uniwersytet Zielonogórski. Oficyna Wydawnicza
    Tematy:
    feature extraction
    feature selection
    meta learning
    ekstrakcja danych
    selekcja danych
    uczenie maszynowe
    Opis:
    The demand for performing data analysis is steadily rising. As a consequence, people of different profiles (i.e., nonexperienced users) have started to analyze their data. However, this is challenging for them. A key step that poses difficulties and determines the success of the analysis is data mining (model/algorithm selection problem). Meta-learning is a technique used for assisting non-expert users in this step. The effectiveness of meta-learning is, however, largely dependent on the description/characterization of datasets (i.e., meta-features used for meta-learning). There is a need for improving the effectiveness of meta-learning by identifying and designing more predictive meta-features. In this work, we use a method from exploratory factor analysis to study the predictive power of different meta-features collected in OpenML, which is a collaborative machine learning platform that is designed to store and organize meta-data about datasets, data mining algorithms, models and their evaluations. We first use the method to extract latent features, which are abstract concepts that group together meta-features with common characteristics. Then, we study and visualize the relationship of the latent features with three different performance measures of four classification algorithms on hundreds of datasets available in OpenML, and we select the latent features with the highest predictive power. Finally, we use the selected latent features to perform meta-learning and we show that our method improves the meta-learning process. Furthermore, we design an easy to use application for retrieving different meta-data from OpenML as the biggest source of data in this domain.
    Źródło:
    International Journal of Applied Mathematics and Computer Science; 2017, 27, 4; 697-712
    1641-876X
    2083-8492
    Pojawia się w:
    International Journal of Applied Mathematics and Computer Science
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Hybrid methodology of degradation feature extraction for bearing prognostics
    Metodyka hybrydowa ekstrakcji cech degradacji do zastosowań w prognozowaniu czasu życia łożysk
    Autorzy:
    Gu, H.
    Zhao, J.
    Zhang, X.
    Powiązania:
    https://bibliotekanauki.pl/articles/302055.pdf
    Data publikacji:
    2013
    Wydawca:
    Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN
    Tematy:
    ekstrakcja cech
    degradacja
    sygnał
    łożysko
    feature extraction
    degradation
    signal
    bearing
    Opis:
    Hybrid methodology of degradation feature extraction was presented which may enable prediction of remaining useful life of a product. In this methodology, firstly, the signal was de-noised by wavelet analysis. Then the autoregressive model was used to remove the discrete frequencies from de-noised signal. Further, the residual signal which mainly contained impulsive fault signal was enhanced by minimum entropy deconvolution filter. The kurtosis was extracted which was taken as the feature for prognostics. At last, the empirical mode decomposition was used to reduce fluctuation of feature value and to extract the trend content. A case study was presented to verify the effectiveness of the proposed method.
    Przedstawiono hybrydową metodę ekstrakcji cech degradacji, która umożliwia przewidywanie pozostałego okresu użytkowania produktu. W tej metodyce, sygnał został najpierw odfiltrowany z wykorzystaniem analizy falkowej. Następnie, za pomocą modelu autoregresyjnego usunięto z pozbawionego szumów sygnału częstotliwości dyskretne. W dalszej kolejności, sygnał resztkowy, który zawierał głównie impulsowy sygnał uszkodzenia został wzmocniony z zastosowaniem filtru dekonwolucji minimum entropii. Obliczono kurtozę, którą przyjęto jako cechę w procesie prognozowania. Na koniec, zastosowano empiryczną dekompozycję sygnału (EMD) w celu zmniejszenia wahań wartości cechy oraz w celu ekstrakcji trendu. Studium przypadku demonstruje efektywność proponowanej metody.
    Źródło:
    Eksploatacja i Niezawodność; 2013, 15, 2; 195-201
    1507-2711
    Pojawia się w:
    Eksploatacja i Niezawodność
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A Fast Method of Feature Extraction for Lowering Vehicle Pass-By Noise Based on Nonnegative Tucker3 Decomposition
    Autorzy:
    Wang, H.
    Cheng, G.
    Deng, G.
    Li, X.
    Li, H.
    Huang, Y.
    Powiązania:
    https://bibliotekanauki.pl/articles/177883.pdf
    Data publikacji:
    2017
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    vehicle pass-by noise
    NTD
    feature extraction
    sound pressure level
    Opis:
    Usually, the judgement of one type fault of vehicle pass-by noise is difficult for engineers, especially when some significant features are disturbed by other interference noise, such as the squealing noise is almost simultaneous with the whistle in the exhaust system. In order to cope with this problem, a new method, with the antinoise ability of the algorithm on the condition by which the features are entangled, is developed to extract clear features for the fault analysis. In the proposed method, the nonnegative Tucker3 decomposition (NTD) with fast updating algorithm, signed as NTD_FUP, can find out the natural frequency of the parts/components from the exhaust system. Not only does the NTD_FUP extract clear features from the confused noise, but also it is superior to the traditional methods in practice. Then, an aluminium-foil alloy material, which is used for the heat shield for its lower noise radiation, replaces the aluminium alloy alone. Extensive experiments show that the sound pressure level of the vehicle pass-by noise is reduced 0.9 dB(A) by the improved heat shield, which is also considered as a more lightweight design for the exhaust system of an automobile.
    Źródło:
    Archives of Acoustics; 2017, 42, 4; 619-629
    0137-5075
    Pojawia się w:
    Archives of Acoustics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    An improved feature extraction method for rolling bearing fault diagnosis based on MEMD and PE
    Autorzy:
    Zhang, H.
    Zhao, L.
    Liu, Q.
    Luo, J.
    Wei, Q.
    Zhou, Z.
    Qu, Y.
    Powiązania:
    https://bibliotekanauki.pl/articles/259770.pdf
    Data publikacji:
    2018
    Wydawca:
    Politechnika Gdańska. Wydział Inżynierii Mechanicznej i Okrętownictwa
    Tematy:
    improved feature extraction method
    rolling bearing fault diagnosis
    MEMD
    PE
    Opis:
    The health condition of rolling bearing can directly influence to the efficiency and lifecycle of rotating machinery, thus monitoring and diagnosing the faults of rolling bearing is of great importance. Unfortunately, vibration signals of rolling bearing are usually overwhelmed by external noise, so the fault frequencies of rolling bearing cannot be readily obtained. In this paper, an improved feature extraction method called IMFs_PE, which combines the multivariate empirical mode decomposition with the permutation entropy, is proposed to extract fault frequencies from the noisy bearing vibration signals. First, the raw bearing vibration signals are filtered by an optimal band-pass filter determined by SK to remove the irrelative noise which is not in the same frequency band of fault frequencies. Then the filtered signals are processed by the IMFs_PE to get rid of the relative noise which is in the same frequency band of fault frequencies. Finally, a frequency domain condition indicator FFR(Fault Frequency Ratio), which measures the magnitude of fault frequencies in frequency domain, is calculated to compare the effectiveness of the feature extraction methods. The feature extraction method proposed in this paper has advantages of removing both irrelative noise and relative noise over other feature extraction methods. The effectiveness of the proposed method is validated by simulated and experimental bearing signals. And the results are shown that the proposed method outperforms other state of the art algorithms with regards to fault feature extraction of rolling bearing.
    Źródło:
    Polish Maritime Research; 2018, S 2; 98-106
    1233-2585
    Pojawia się w:
    Polish Maritime Research
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A Multistage Procedure of Mobile Vehicle Acoustic Identification for Single-Sensor Embedded Device
    Autorzy:
    Astapov, S.
    Riid, A.
    Powiązania:
    https://bibliotekanauki.pl/articles/227146.pdf
    Data publikacji:
    2013
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    vehicle identification
    acoustic signal analysis
    feature extraction
    classification
    fuzzy logic
    Opis:
    Mobile vehicle identification has a wide application field for both civilian and military uses. Vehicle identification may be achieved by incorporating single or multiple sensor solutions and through data fusion. This paper considers a single-sensor multistage hierarchical algorithm of acoustic signal analysis and pattern recognition for the identification of mobile vehicles in an open environment. The algorithm applies several standalone techniques to enable complex decision-making during event identification. Computationally inexpensive procedures are specifically chosen in order to provide real-time operation capability. The algorithm is tested on pre-recorded audio signals of civilian vehicles passing the measurement point and shows promising classification accuracy. Implementation on a specific embedded device is also presented and the capability of real-time operation on this device is demonstrated.
    Źródło:
    International Journal of Electronics and Telecommunications; 2013, 59, 2; 151-160
    2300-1933
    Pojawia się w:
    International Journal of Electronics and Telecommunications
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Vein Biometric Recognition Methods and Systems: A Review
    Autorzy:
    Al-Khafaji, Ruaa S.S.
    Al-Tamimi, Mohammed S.H.
    Powiązania:
    https://bibliotekanauki.pl/articles/2022496.pdf
    Data publikacji:
    2022
    Wydawca:
    Stowarzyszenie Inżynierów i Techników Mechaników Polskich
    Tematy:
    biometric technology
    finger vein recognition
    pre-processing
    feature extraction
    matching
    Opis:
    The Finger-vein recognition (FVR) method has received increasing attention in recent years. It is a new method of personal identification and biometric technology that identifies individuals using unique finger-vein patterns, which is the first reliable and suitable area to be recognized. It was discovered for the first time with a home imaging system; it is characterized by high accuracy and high processing speed. Also, the presence of patterns of veins inside one’s body makes it almost difficult to repeat and difficult to steal. Based on the increased focus on protecting privacy, that also produces vein biometrics safer alternatives without forgery, damage, or alteration over time. Fingerprint recognition is beneficial because it includes the use of low-cost, small devices which are difficult to counterfeit. This paper discusses preceding finger-vein recognition approaches systems with the methodologies taken from other researchers’ work about image acquisition, pretreatment, vein extraction, and matching. It is reviewing the latest algorithms; continues to critically review the strengths and weaknesses of these methods, and it states the modern results following a key comparative analysis of methods.
    Źródło:
    Advances in Science and Technology. Research Journal; 2022, 16, 1; 36-46
    2299-8624
    Pojawia się w:
    Advances in Science and Technology. Research Journal
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Learning-free deep features for multispectral palm-print classification
    Autorzy:
    Aounallah, Asma
    Meraoumia, Abdallah
    Bendjenna, Hakim
    Powiązania:
    https://bibliotekanauki.pl/articles/27312870.pdf
    Data publikacji:
    2023
    Wydawca:
    Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie. Wydawnictwo AGH
    Tematy:
    feature extraction
    biometrics
    multispectral imaging
    deep learning
    DCTNet
    data fusion
    Opis:
    The feature-extraction step is a major and crucial step in analyzing and understanding raw data, as it has a considerable impact on system accuracy. Despite the very acceptable results that have been obtained by many handcrafted methods, these can unfortunately have difficulty representing features in the cases of large databases or with strongly correlated samples. In this context, we attempt to examine the discriminability of texture features by proposing a novel, simple, and lightweight method for deep feature extraction to characterize the discriminative power of different textures. We evaluated the performance of our method by using a palm print-based biometric system, and the experimental results (using the CASIA multispectral palm--print database) demonstrate the superiority of the proposed method over the latest handcrafted and deep methods.
    Źródło:
    Computer Science; 2023, 24 (2); 243--271
    1508-2806
    2300-7036
    Pojawia się w:
    Computer Science
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Feature selection for EEG-based discrimination between imagination of left and right hand movements
    Autorzy:
    Binias, B.
    Palus, H.
    Powiązania:
    https://bibliotekanauki.pl/articles/114144.pdf
    Data publikacji:
    2015
    Wydawca:
    Stowarzyszenie Inżynierów i Techników Mechaników Polskich
    Tematy:
    EEG signal
    brain-computer interfaces
    feature extraction
    classification
    lateralized brain activity
    Opis:
    : In this article was analyzed an influence of selected features on the accuracy of discrimination between imagination of right and left hand movements based on recorded EEG waveforms. The study showed a significant advantage that individual selection of features and a classification algorithm for analyzed data holds over the more general approach. The results were compared with the results obtained by the participants of the "BCI competition IV" and placed in the top three.
    Źródło:
    Measurement Automation Monitoring; 2015, 61, 4; 94-97
    2450-2855
    Pojawia się w:
    Measurement Automation Monitoring
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Classification of the indoor environment of a mobile robot using principal component analysis
    Autorzy:
    Yaqub, T.
    Katupitya, J.
    Powiązania:
    https://bibliotekanauki.pl/articles/384275.pdf
    Data publikacji:
    2008
    Wydawca:
    Sieć Badawcza Łukasiewicz - Przemysłowy Instytut Automatyki i Pomiarów
    Tematy:
    mobile robot environment
    PCA
    classification
    feature extraction
    training data
    bootstrap method
    Opis:
    Large indoor environments of a mobile robot usually consist of different types of areas connected together. The structure of a corridor differs from a room, a main hall or laboratory. A method for online classification of these areas using a laser scanner is presented in this paper. This classification can reduce the search space of localization module to a great extent making the navigation system efficient. The intention was to make the classification of a sensor observation in a fast and real-time fashion and immediately on its arrival in the sensor frame. Our approach combines both the feature based and statistical approaches. We extract some vital features of lines and corners with attributes such as average length of lines and distance between corners from the raw laser data and classify the observation based on these features. Bootstrap method is used to get a robust correlation of features from training data and finally Principal Component Analysis (PCA) is used to model the environment. In PCA, the underlying assumption is that data is coming from a multivariate normal distribution. The use of bootstrap method makes it possible to use the observations data set which set, which is not necessarily normally distributed. This technique lifts up the normality assumption and reduces the computational cost further as compared to the PCA techniques based on raw sensor data and can be easily implemented in moderately complex indoor environment. The knowledge of the environment can also be up-dated in an adaptive fashion. Results of experimentation in a simulated hospital building under varying environmental conditions using a real-time robotic software Player/Stage are shown.
    Źródło:
    Journal of Automation Mobile Robotics and Intelligent Systems; 2008, 2, 2; 44-53
    1897-8649
    2080-2145
    Pojawia się w:
    Journal of Automation Mobile Robotics and Intelligent Systems
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Domain specific key feature extraction using knowledge graph mining
    Autorzy:
    Barai, Mohit Kumar
    Sanyal, Subhasis
    Powiązania:
    https://bibliotekanauki.pl/articles/2027771.pdf
    Data publikacji:
    2020
    Wydawca:
    Uniwersytet Ekonomiczny w Katowicach
    Tematy:
    Feature extraction
    Knowledge graph
    Natural language processing
    Product review
    Text processing
    Opis:
    In the field of text mining, many novel feature extraction approaches have been propounded. The following research paper is based on a novel feature extraction algorithm. In this paper, to formulate this approach, a weighted graph mining has been used to ensure the effectiveness of the feature extraction and computational efficiency; only the most effective graphs representing the maximum number of triangles based on a predefined relational criterion have been considered. The proposed novel technique is an amalgamation of the relation between words surrounding an aspect of the product and the lexicon-based connection among those words, which creates a relational triangle. A maximum number of a triangle covering an element has been accounted as a prime feature. The proposed algorithm performs more than three times better than TF-IDF within a limited set of data in analysis based on domain-specific data.
    Źródło:
    Multiple Criteria Decision Making; 2020, 15; 1-22
    2084-1531
    Pojawia się w:
    Multiple Criteria Decision Making
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    k-centroids clustering for asymmetric dissimilarities
    Autorzy:
    Olszewski, D.
    Powiązania:
    https://bibliotekanauki.pl/articles/206422.pdf
    Data publikacji:
    2011
    Wydawca:
    Polska Akademia Nauk. Instytut Badań Systemowych PAN
    Tematy:
    k-centroids clustering
    asymmetric dissimilarity
    sound recognition
    heart rhythm recognition
    feature extraction
    Opis:
    In this paper, an asymmetric version of the kcentroids clustering algorithm is proposed. The asymmetry arises from the use of the asymmetric dissimilarities in the k-centroids algorithm. Application of the asymmetric measures of dissimilarity is motivated by the basic nature of the k-centroids algorithm, which uses dissimilarities in the asymmetric manner. It finds the minimal dissimilarity between an object being currently allocated, and one of the clusters centroids. Clusters centroids are treated as the dominant points governing the asymmetric relationships in the entire cluster analysis. The results of the experimental study on real and simulated data have shown the superiority of the asymmetric dissimilarities employed for the k-centroids method over their symmetric counterparts.
    Źródło:
    Control and Cybernetics; 2011, 40, 2; 554-574
    0324-8569
    Pojawia się w:
    Control and Cybernetics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Accurate identification on individual similar communication emitters by using HVG-NTE feature
    Autorzy:
    Li, Ke
    Ge, Wei
    Yang, Xiaoya
    Xu, Zhengrong
    Powiązania:
    https://bibliotekanauki.pl/articles/2128146.pdf
    Data publikacji:
    2021
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    communication emitter
    identification
    feature extraction
    HVG
    NTE
    emiter komunikacji
    identyfikacja
    wyodrębnianie cech
    Opis:
    Individual identification of similar communication emitters in the complex electromagnetic environment has great research value and significance in both military and civilian fields. In this paper, a feature extraction method called HVG-NTE is proposed based on the idea of system nonlinearity. The shape of the degree distribution, based on the extraction of HVG degree distribution, is quantified with NTE to improve the anti-noise performance. Then XGBoost is used to build a classifier for communication emitter identification. Our method achieves better recognition performance than the state-of-the-art technology of the transient signal data set of radio stations with the same plant, batch, and model, and is suitable for a small sample size.
    Źródło:
    Bulletin of the Polish Academy of Sciences. Technical Sciences; 2021, 69, 2; e136741, 1--6
    0239-7528
    Pojawia się w:
    Bulletin of the Polish Academy of Sciences. Technical Sciences
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Accurate identification on individual similar communication emitters by using HVG-NTE feature
    Autorzy:
    Li, Ke
    Ge, Wei
    Yang, Xiaoya
    Xu, Zhengrong
    Powiązania:
    https://bibliotekanauki.pl/articles/2173613.pdf
    Data publikacji:
    2021
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    communication emitter
    identification
    feature extraction
    HVG
    NTE
    emiter komunikacji
    identyfikacja
    wyodrębnianie cech
    Opis:
    Individual identification of similar communication emitters in the complex electromagnetic environment has great research value and significance in both military and civilian fields. In this paper, a feature extraction method called HVG-NTE is proposed based on the idea of system nonlinearity. The shape of the degree distribution, based on the extraction of HVG degree distribution, is quantified with NTE to improve the anti-noise performance. Then XGBoost is used to build a classifier for communication emitter identification. Our method achieves better recognition performance than the state-of-the-art technology of the transient signal data set of radio stations with the same plant, batch, and model, and is suitable for a small sample size.
    Źródło:
    Bulletin of the Polish Academy of Sciences. Technical Sciences; 2021, 69, 2; art. no. e136741
    0239-7528
    Pojawia się w:
    Bulletin of the Polish Academy of Sciences. Technical Sciences
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Enhancement in Bearing Fault Classification Parameters Using Gaussian Mixture Models and Mel Frequency Cepstral Coefficients Features
    Autorzy:
    Atmani, Youcef
    Rechak, Said
    Mesloub, Ammar
    Hemmouche, Larbi
    Powiązania:
    https://bibliotekanauki.pl/articles/177335.pdf
    Data publikacji:
    2020
    Wydawca:
    Polska Akademia Nauk. Czasopisma i Monografie PAN
    Tematy:
    bearing faults
    Gaussian mixture models
    Mel frequency cepstral coefficients
    feature extraction
    diagnosis
    Opis:
    Last decades, rolling bearing faults assessment and their evolution with time have been receiving much interest due to their crucial role as part of the Conditional Based Maintenance (CBM) of rotating machinery. This paper investigates bearing faults diagnosis based on classification approach using Gaussian Mixture Model (GMM) and the Mel Frequency Cepstral Coefficients (MFCC) features. Throughout, only one criterion is defined for the evaluation of the performance during all the cycle of the classification process. This is the Average Classification Rate (ACR) obtained from the confusion matrix. In every test performed, the generated features vectors are considered along to discriminate between four fault conditions as normal bearings, bearings with inner and outer race faults and ball faults. Many configurations were tested in order to determinate the optimal values of input parameters, as the frame analysis length, the order of model, and others. The experimental application of the proposed method was based on vibration signals taken from the bearing datacenter website of Case Western Reserve University (CWRU). Results show that proposed method can reliably classify different fault conditions and have a highest classification performance under some conditions.
    Źródło:
    Archives of Acoustics; 2020, 45, 2; 283-295
    0137-5075
    Pojawia się w:
    Archives of Acoustics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Przydatność algorytmów podpikselowej detekcji cech w wybranych zagadnieniach fotogrametrycznych
    The usefulness of sub-pixel feature extraction algorithms in selected photogrammetric cases
    Autorzy:
    Mikrut, S.
    Powiązania:
    https://bibliotekanauki.pl/articles/130666.pdf
    Data publikacji:
    2009
    Wydawca:
    Stowarzyszenie Geodetów Polskich
    Tematy:
    fotogrametria cyfrowa
    dokładność podpikselowa
    detekcja cech
    digital photogrammetry
    subpixel precision
    feature extraction
    Opis:
    Celem artykułu było zaprezentowanie wyników badań nad przydatnością algorytmów podpikselowej dokładności detekcji cech w wybranych zagadnieniach fotogrametrycznych. Pojęcie detekcji cech obejmuje wszystkie metody mające na celu wydobycie z obrazu cyfrowego właściwości takich jak: pojedyncze obiekty, krawędzie, geometryczne środki obiektów czy inne. W niniejszej publikacji autor skupił się na zagadnieniu detekcji cech prowadzących do zdefiniowania krawędzi wybranych obiektów celem ich lokalizacji z bardzo wysoką precyzją. Do tego celu wykorzystano autorskie algorytmy bazujące na wykrywaniu krawędzi z podpikselową dokładnością w oparciu o badanie pierwszej i drugiej pochodnej obrazu cyfrowego. Znajdowane z taką precyzją punkty na odpowiednim przekroju krawędzi (wiersz lub kolumna obrazu cyfrowego) są następnie aproksymowane wybraną funkcją matematyczną (w najprostszym przypadku będzie to linia prosta) i w efekcie końcowym dochodzimy do postaci wektorowej krawędzi. Algorytmy te zostały przetestowane w wielu pracach fotogrametrycznych. Niniejsza publikacja przybliża ich wykorzystanie w jednym z zadań fotogrametrii inżynierskiej jakim jest badanie imperfekcji kształtu wysmukłych budowli. Opracowane procedury pozwalające na wykonanie automatycznych pomiarów na obrazie cyfrowym, dały podstawy na zaprojektowania systemu, który docelowo miał być zintegrowany z notebookiem, aby umożliwić prace w czasie rzeczywistym. Algorytmy do badania imperfekcji kształtu na obrazie cyfrowym bazują na detekcji pojedynczych punktów przekroi krawędziowych, a następnie na wpasowaniu linii prostej reprezentującej badany kształt obiektu. Algorytmy zostały oprogramowanie i zaimplementowane do systemu powstającego w ramach projektu badawczego.
    The purpose of the article was to present the results of research on the usefulness of algorithms of sub-pixel accuracy of feature extraction in selected photogrammetric cases. The concept of feature extraction includes all methods aiming at the extraction from a digital image of such selected properties as single objects, edges, geometric centres of objects, or other properties. In this publication, the author focuses on the problem of feature extraction, leading to the definition of edges of selected objects in order to be able to locate them with a very high precision. For this purpose, own algorithms based on edge detection with a sub-pixel accuracy based on the examination of the first and the second derivatives of the digital image were applied. Points, which have been found with such a precision at the appropriate section of the edge (row or column of the digital image) are later approximated, using a selected mathematical function (in the simplest case this will be a straight line), and finally, a vector form of the edge is obtained. Those algorithms have been tested in many photogrammetric works. This publication brings closer their utilization in one of the photogrammetric engineering tasks, namely the investigation of the imperfection of shape of lofty buildings and structures. The developed procedures for the implementation of automated measurements on a digital image have helped to design a system that eventually was to be integrated with a notebook, allowing for work to be performed in a real time. Algorithms for testing the shape imperfection on a digital image are based on detecting single points of edge sections, and then fitting a straight line, which represents the shape of the object under investigation. The software and algorithms have been implemented to the system being created under a research project. The whole is written in C + +.
    Źródło:
    Archiwum Fotogrametrii, Kartografii i Teledetekcji; 2009, 19; 299-308
    2083-2214
    2391-9477
    Pojawia się w:
    Archiwum Fotogrametrii, Kartografii i Teledetekcji
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Texture and gene expression analysis of the MRI brain in detection of Alzheimer’s disease
    Autorzy:
    Bustamam, A.
    Sarwinda, D.
    Ardenaswari, G.
    Powiązania:
    https://bibliotekanauki.pl/articles/91834.pdf
    Data publikacji:
    2018
    Wydawca:
    Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych
    Tematy:
    Alzheimer’s disease
    MRI
    Feature Extraction
    Bi-Clustering
    Local Binary Pattern
    LBP
    Opis:
    Alzheimer’s disease is a type of dementia that can cause problems with human memory, thinking and behavior. This disease causes cell death and nerve tissue damage in the brain. The brain damage can be detected using brain volume, whole brain form, and genetic testing. In this research, we propose texture analysis of the brain and genomic analysis to detect Alzheimer’s disease. 3D MRI images were chosen to analyze the texture of the brain, and microarray data were chosen to analyze gene expression. We classified Alzheimer’s disease into three types: Alzheimer’s, Mild Cognitive Impairment (MCI), and Normal. In this study, texture analysis was carried out by using the Advanced Local Binary Pattern (ALBP) and the Gray Level Co-occurrence Matrix (GLCM). We also propose the bi-clustering method to analyze microarray data. The experimental results from texture analysis show that ALBP had better performance than GLCM in classification of Alzheimer’s disease. The ALBP method achieved an average value of accuracy of between 75% - 100% for binary classification of the whole brain data. Furthermore, Biclustering method with microarray data shows good performance gene expression, where this information show influence Alzheimer’s disease with total of bi-cluster is 6.
    Źródło:
    Journal of Artificial Intelligence and Soft Computing Research; 2018, 8, 2; 111-120
    2083-2567
    2449-6499
    Pojawia się w:
    Journal of Artificial Intelligence and Soft Computing Research
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Hydrogen Detection With a Gas Sensor Array – Processing and Recognition of Dynamic Responses Using Neural Networks
    Autorzy:
    Gwiżdż, P.
    Brudnik, A.
    Zakrzewska, K.
    Powiązania:
    https://bibliotekanauki.pl/articles/221723.pdf
    Data publikacji:
    2015
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    gas sensor
    sensor array
    temperature modulation
    dynamic response
    feature extraction
    neural networks
    Opis:
    An array consisting of four commercial gas sensors with target specifications for hydrocarbons, ammonia, alcohol, explosive gases has been constructed and tested. The sensors in the array operate in the dynamic mode upon the temperature modulation from 350°C to 500°C. Changes in the sensor operating temperature lead to distinct resistance responses affected by the gas type, its concentration and the humidity level. The measurements are performed upon various hydrogen (17-3000 ppm), methane (167-3000 ppm) and propane (167-3000 ppm) concentrations at relative humidity levels of 0-75%RH. The measured dynamic response signals are further processed with the Discrete Fourier Transform. Absolute values of the dc component and the first five harmonics of each sensor are analysed by a feed-forward back-propagation neural network. The ultimate aim of this research is to achieve a reliable hydrogen detection despite an interference of the humidity and residual gases.
    Źródło:
    Metrology and Measurement Systems; 2015, 22, 1; 3-12
    0860-8229
    Pojawia się w:
    Metrology and Measurement Systems
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Fault diagnosis method for an aerospace generator rotating rectifier based on dynamic FFT technology
    Autorzy:
    Feng, Sai
    Cui, Jiang
    Zhang, Zhuoran
    Powiązania:
    https://bibliotekanauki.pl/articles/1849142.pdf
    Data publikacji:
    2021
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    aerospace generator
    rotating rectifier
    fault diagnosis
    dynamic Fast Fourier Transform
    feature extraction
    Opis:
    A fault diagnosis method for the rotating rectifier of a brushless three-phase synchronous aerospace generator is proposed in this article. The proposed diagnostic system includes three steps: data acquisition, feature extraction and fault diagnosis. Based on a dynamic Fast Fourier Transform (FFT), this method processes the output voltages of aerospace generator continuously and monitors the continuous change trend of the main frequency in the spectrum before and after the fault. The trend can be used to perform fault diagnosis task. The fault features of the rotating rectifier proposed in this paper can quickly and effectively distinguish single and double faulty diodes. In order to verify the proposed diagnosis system, simulation and practical experiments are carried out in this paper, and good results can be achieved.
    Źródło:
    Metrology and Measurement Systems; 2021, 28, 2; 269-288
    0860-8229
    Pojawia się w:
    Metrology and Measurement Systems
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Fusing laser and vision data for a perceptually rich environment description
    Opis otoczenia na podstawie danych z sensorów laserowych i wizyjnych
    Autorzy:
    Skrzypczyński, P.
    Powiązania:
    https://bibliotekanauki.pl/articles/257108.pdf
    Data publikacji:
    2008
    Wydawca:
    Sieć Badawcza Łukasiewicz - Instytut Technologii Eksploatacji - Państwowy Instytut Badawczy
    Tematy:
    robot mobilny
    nawigacja
    system wizyjny
    ekstrakcja cech
    mobile robot
    navigation
    computer vision
    feature extraction
    Opis:
    In this paper we, discuss methods to increase the discriminative properties of the laser-based geometric landmarks used in simultaneous localisation and mapping by employing monocular vision data. Vertical edges extracted from images help to estimate the length of the line segments, which are only partially observed. Salient visual features, which defy simple geometric interpretation, are handled by the scale invariant feature transform method. These different types of photometric features are aggregated together with the basic 2D line segments extracted from the laser scanner data into the Perceptually Rich Segments.
    W pracy przedstawiono metody poprawiające rozróżnialność obiektów geometrycznych wyodrębnionych z danych uzyskanych ze skanera laserowego i wykorzystywanych w systemie jednoczesnej samolokalizacji i budowy mapy otoczenia robota. Założono, że robot porusza się w środowisku zbudowanym przez człowieka, w którym dominują pionowe płaszczyzny (ściany). Poprawę rozróżnialności obiektów uzyskano dzięki wykorzystaniu danych z monookularowego systemu wizyjnego robota. Krawędzie pionowe wyodrębnione z obrazów umożliwiają prawidłową estymację długości odcinków 2D odtworzonych uprzednio na podstawie danych ze skanera laserowego. Fotometryczne cechy znaczące wyodrębniane są z obrazów i opisywane za pomocą metody Scale Invariant Feature Transform (SIFT). Uzyskane wektory parametrów osadzane są następnie w "ramach" tworzonych przez odcinki poziome oraz krawędzie pionowe. Powstają w ten sposób obiekty nowego typu - PRS (ang. Perceptually Rich Segment). Zaprezentowano wyniki eksperymentów dotyczących wyodrębniania i dopasowywania do siebie wektorów SIFT oraz wstępne wyniki dotyczące budowy modelu otoczenia z użyciem obiektów PRS.
    Źródło:
    Problemy Eksploatacji; 2008, 3; 57-67
    1232-9312
    Pojawia się w:
    Problemy Eksploatacji
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Intelligent prediction of milling strategy using neural networks
    Autorzy:
    Klancnik, S.
    Balic, J.
    Cus, F.
    Powiązania:
    https://bibliotekanauki.pl/articles/971013.pdf
    Data publikacji:
    2010
    Wydawca:
    Polska Akademia Nauk. Instytut Badań Systemowych PAN
    Tematy:
    SOM neural networks
    CAD/CAM system
    feature extraction
    milling strategy
    CAD segmentation
    STL model
    Opis:
    This paper presents the prediction of milling tool-path strategy using Artificial Neural Network (ANN), by taking the predefined technological objectives into account. In the presented case, the best possible surface quality of a machined surface was taken as the primary technological aim. This paper shows how feature extraction from a 3D CAD model, and classification using a self-organizing neural network, are done. The experimental results presented in this paper suggest that the prediction of milling strategy using the self-organizing neural network (SOM) is effective.
    Źródło:
    Control and Cybernetics; 2010, 39, 1; 9-24
    0324-8569
    Pojawia się w:
    Control and Cybernetics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Recognition of acoustic signals of induction motor using fft, smofs-10 and ISVM
    Rozpoznawanie sygnałów akustycznich silnika indukcyjnego z zastosowaniem fft, smofs-10 i ISVM
    Autorzy:
    Głowacz, A.
    Powiązania:
    https://bibliotekanauki.pl/articles/1365918.pdf
    Data publikacji:
    2015
    Wydawca:
    Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN
    Tematy:
    acoustic signal
    induction motor
    feature extraction
    classification
    sygnał akustyczny
    silnik indukcyjny
    ekstrakcja cech
    klasyfikacja
    Opis:
    A correct diagnosis of electrical circuits is very essential in industrial plants. An article deals with a recognition method of early fault detection of induction motor. The described approach is based on patterns recognition. Acoustic signals of specific induction motor are analyzed patterns. Acoustic signals include information about motor state. The analysis of the patterns was conducted for three states of induction motor using Fast Fourier Transform (FFT), shortened method of frequencies selection (SMoFS-10) and Linear Support Vector Machine (LSVM). The results of calculations suggest that the method is efficient and can be also used for diagnostic purposes.
    Prawidłowa diagnostyka obwodów elektrycznych jest bardzo istotna w zakładach przemysłowych. Artykuł zajmuje się metodą rozpoznawania stanów przedawaryjnych silnika indukcyjnego. Opisane podejście jest oparte na rozpoznawaniu wzorców. Sygnały akustyczne określonego silnika indukcyjnego są badanymi wzorcami. Sygnały akustyczne zawierają informację o stanie silnika. Analiza wzorców została przeprowadzona dla trzech stanów silnika indukcyjnego używając FFT, skróconej metody wyboru częstotliwości (SMoFS-10) i liniowej maszyny wektorów wspierających (LSVM). Wyniki obliczeń sugerują, że metoda jest skuteczna i może być również zastosowana dla celów diagnostycznych.
    Źródło:
    Eksploatacja i Niezawodność; 2015, 17, 4; 569-574
    1507-2711
    Pojawia się w:
    Eksploatacja i Niezawodność
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Critical exponent analysis applied to surface EMG signals for multifunction myoelectric control
    Autorzy:
    Phinyomark, A.
    Phothisonothai, M.
    Phukpattaranont, P.
    Limsakul, C.
    Powiązania:
    https://bibliotekanauki.pl/articles/220544.pdf
    Data publikacji:
    2011
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    biomedical signal processing
    electromyography signal
    feature extraction
    fractal analysis
    human machine interface
    pattern classification
    Opis:
    Based on recent advances in non-linear analysis, the surface electromyography (sEMG) signal has been studied from the viewpoints of self-affinity and complexity. In this study, we examine usage of critical exponent analysis (CE) method, a fractal dimension (FD) estimator, to study properties of the sEMG signal and to deploy these properties to characterize different movements for gesture recognition. SEMG signals were recorded from thirty subjects with seven hand movements and eight muscle channels. Mean values and coefficient of variations of the CE from all experiments show that there are larger variations between hand movement types but there is small variation within the same type. It also shows that the CE feature related to the self-affine property for the sEMG signal extracted from different activities is in the range of 1.855∼2.754. These results have also been evaluated by analysis-of-variance (p-value). Results show that the CE feature is more suitable to use as a learning parameter for a classifier compared with other representative features including root mean square, median frequency and Higuchi's method. Most p-values of the CE feature were less than 0.0001. Thus the FD that is computed by the CE method can be applied to be used as a feature for a wide variety of sEMG applications.
    Źródło:
    Metrology and Measurement Systems; 2011, 18, 4; 645-658
    0860-8229
    Pojawia się w:
    Metrology and Measurement Systems
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Semantic Segmentation of Diseases in Mushrooms using Enhanced Random Forest
    Autorzy:
    Yacharam, Rakesh Kumar
    Sekhar, Dr. V. Chandra
    Powiązania:
    https://bibliotekanauki.pl/articles/31339414.pdf
    Data publikacji:
    2023
    Wydawca:
    Szkoła Główna Gospodarstwa Wiejskiego w Warszawie. Instytut Informatyki Technicznej
    Tematy:
    mushroom diseases
    semantic segmentation
    computer aided
    Machine Learning
    significant feature extraction
    Random Forest classifier
    Opis:
    Mushrooms are a rich source of antioxidants and nutritional values. Edible mushrooms, however, are susceptible to various diseases such as dry bubble, wet bubble, cobweb, bacterial blotches, and mites. Farmers face significant production losses due to these diseases affecting mushrooms. The manual detection of these diseases relies on expertise, knowledge of diseases, and human effort. Therefore, there is a need for computer-aided methods, which serve as optimal substitutes for detecting and segmenting diseases. In this paper, we propose a semantic segmentation approach based on the Random Forest machine learning technique for the detection and segmentation of mushroom diseases. Our focus lies in extracting a combination of different features, including Gabor, Bouda, Kayyali, Gaussian, Canny edge, Roberts, Sobel, Scharr, Prewitt, Median, and Variance. We employ constant mean-variance thresholding and the Pearson correlation coefficient to extract significant features, aiming to enhance computational speed and reduce complexity in training the Random Forest classifier. Our results indicate that semantic segmentation based on Random Forest outperforms other methods such as Support Vector Machine (SVM), Naïve Bayes, K-means, and Region of Interest in terms of accuracy. Additionally, it exhibits superior precision, recall, and F1 score compared to SVM. It is worth noting that deep learning-based semantic segmentation methods were not considered due to the limited availability of diseased mushroom images.
    Źródło:
    Machine Graphics & Vision; 2023, 32, 2; 129-146
    1230-0535
    2720-250X
    Pojawia się w:
    Machine Graphics & Vision
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Personal identification using retina
    Autorzy:
    Choraś, R.S.
    Powiązania:
    https://bibliotekanauki.pl/articles/333033.pdf
    Data publikacji:
    2009
    Wydawca:
    Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
    Tematy:
    siatkówka
    biometria siatkówki
    wybór cech obrazu
    retina
    retina biometrics
    vessel pattern
    feature extraction
    Gabor transform
    feature vector
    matching
    Opis:
    This paper proposes a biometric system for authentication that uses the retina blood vessel pattern. The retina biometric analyzes the layer of blood vessels located at the back of the eye. The blood vessels at the back of the eye have a unique pattern, from eye to eye and person to person. The retina, a layer of blood vessels located at the back of the eye, forms an identity card for the individual under investigation. In particular retinal recognition creates an ”eye signature” from its vascular configuration and its artificial duplication is thought to be virtually impossible.
    Źródło:
    Journal of Medical Informatics & Technologies; 2009, 13; 53-58
    1642-6037
    Pojawia się w:
    Journal of Medical Informatics & Technologies
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A diagnostic algorithm diagnosing the failure of railway signal equipment
    Autorzy:
    Wu, Yongcheng
    Cao, Dejin
    Powiązania:
    https://bibliotekanauki.pl/articles/1955227.pdf
    Data publikacji:
    2021
    Wydawca:
    Polska Akademia Nauk. Polskie Towarzystwo Diagnostyki Technicznej PAN
    Tematy:
    failure diagnosis
    railway signal equipment
    denoising
    feature extraction
    diagnostyka uszkodzeń
    sygnalizacja kolejowa
    odszumianie
    ekstrakcja cech
    Opis:
    Failure of railway signal equipment can cause an impact on its normal operation, and it is necessary to make a timely diagnosis of the failure. In this study, the data of a railway bureau from 2016 to 2020 were studied as an example. Firstly, denoising and feature extraction were performed on the data; then the Adaptive Comprehensive Oversampling (ADASYN) method was used to synthesize minority class samples; finally, three algorithms, back-propagation neural network (BPNN), support vector machine (SVM) and C4.5 algorithms, were used for failure diagnosis. It was found that the three algorithms performed poorly in diagnosing the original data but performed significantly better in diagnosing the synthesized samples, among which the BPNN algorithm had the best performance. The average precision, recall rate and F1 score of the BPNN algorithm were 0.94, 0.92 and 0.93, respectively. The results verify the effectiveness of the BPNN algorithm for failure diagnosis, and the algorithm can be further promoted and applied in practice.
    Źródło:
    Diagnostyka; 2021, 22, 4; 33-38
    1641-6414
    2449-5220
    Pojawia się w:
    Diagnostyka
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Research on bispectrum analysis of secondary feature for vehicle exterior noise based on nonnegative tucker3 decomposition
    Badania nad analizą bispektrum cech drugorzędnych hałasu zewnętrznego pojazdów w oparciu o nieujemną dekompozycję Tuckera3
    Autorzy:
    Wang, H.
    Deng, G.
    Li, Q.
    Kang, Q.
    Powiązania:
    https://bibliotekanauki.pl/articles/301586.pdf
    Data publikacji:
    2016
    Wydawca:
    Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN
    Tematy:
    feature extraction
    vehicle exterior noise
    NTD
    updating algorithm
    ekstrakcja cech
    hałas zewnętrzny pojazdu
    algorytm aktualizacyjny
    Opis:
    Nowadays, analysis of external vehicle noise has become more and more difficult for NVH (noise vibration and harshness) engineer to find out the fault among the exhaust system when some significant features are masked by the jamming signals, especially in the case of the vibration noise associating to the bodywork. New method is necessary to be explored and applied to decompose a high-order tensor and extract the useful features (also known as secondary features in this paper). Nonnegative Tucker3 decomposition (NTD) is proposed and applied into secondary feature extraction for its high efficiency of decomposition and well property of physical architecture, which serves as fault diagnosis of exhaust system for an automobile car. Furthermore, updating algorithm conjugating with Newton-Gaussian gradient decent is utilized to solve the problem of overfitting, which occurs abnormally on traditional iterative method of NTD. Extensive experimen results show the bispectrum of secondary features can not only exceedingly interpret the state of vehicle exterior noise, but also be benefit to observe the abnormal frequency of some important features masked before. Meanwhile, the overwhelming performance of NTD algorithm is verified more effective under the same condition, comparing with other traditional methods both at the deviation of successive relative error and the computation time.
    Obecnie inżynierowie NVH (zajmujący się problematyką hałasu, drgań i uciążliwości akustycznych) napotykają na coraz większe trudności przy analizie hałasu zewnętrznego pojazdów wynikające z faktu, że istotne cechy związane z nieprawidłowościami układu wydechowego są maskowane przez sygnały zakłócające, szczególnie hałas wibracyjny związany z pracą nadwozia. Niezbędna jest zatem nowa metoda, która pozwoli rozkładać tensory wysokiego rzędu i wyodrębniać przydatne cechy (zwane w tym artykule także cechami drugorzędnymi). Do ekstrakcji cech drugorzędnych wykorzystano w prezentowanej pracy metodę nieujemnej faktoryzacji tensorów znaną także jako nieujemna dekompozycja Tuckera 3 (NTD) , która cechuje się wysoką efektywnością dekompozycji i może być wykorzystywana w diagnostyce uszkodzeń układu wydechowego samochodów. Problem nadmiernego dopasowania, który występuje w tradycyjnej metodzie iteracyjnej NTD rozwiązano przy pomocy algorytmu aktualizacyjnego sprzężonego z gradientem prostym Newtona-Gaussa. Wyniki doświadczeń pokazują, że bispektrum cech drugorzędnych nie tylko pozwala doskonale interpretować stan hałasu zewnętrznego pojazdu, ale również umożliwia wykrywanie wcześniej maskowanych nieprawidłowych częstotliwości odpowiadających niektórym ważnym cechom. Badania potwierdzają, że algorytmu NTD jest bardziej efektywny, w tych samych warunkach, w porównaniu z innymi tradycyjnymi metodami zarówno w zakresie odchyleń błędu względnego jak i czasu obliczeń.
    Źródło:
    Eksploatacja i Niezawodność; 2016, 18, 2; 291-298
    1507-2711
    Pojawia się w:
    Eksploatacja i Niezawodność
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Hybridisation of Mel Frequency Cepstral Coefficient and Higher Order Spectral Features for Musical Instruments Classification
    Autorzy:
    Bhalke, D. G.
    Rama Rao, C. B.
    Bormane, D.
    Powiązania:
    https://bibliotekanauki.pl/articles/176497.pdf
    Data publikacji:
    2016
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    feature extraction
    MFCC
    HOS
    bispectrum
    bicoherence
    non-linearity
    non-Gaussianity
    CPNN
    zero crossing rate (ZCR)
    Opis:
    This paper presents the classification of musical instruments using Mel Frequency Cepstral Coefficients (MFCC) and Higher Order Spectral features. MFCC, cepstral, temporal, spectral, and timbral features have been widely used in the task of musical instrument classification. As music sound signal is generated using non-linear dynamics, non-linearity and non-Gaussianity of the musical instruments are important features which have not been considered in the past. In this paper, hybridisation of MFCC and Higher Order Spectral (HOS) based features have been used in the task of musical instrument classification. HOS-based features have been used to provide instrument specific information such as non-Gaussianity and non-linearity of the musical instruments. The extracted features have been presented to Counter Propagation Neural Network (CPNN) to identify the instruments and their family. For experimentation, isolated sounds of 19 musical instruments have been used from McGill University Master Sample (MUMS) sound database. The proposed features show the significant improvement in the classification accuracy of the system.
    Źródło:
    Archives of Acoustics; 2016, 41, 3; 427-436
    0137-5075
    Pojawia się w:
    Archives of Acoustics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Dimensionality Reduction for Probabilistic Neural Network in Medical Data Classification Problems
    Autorzy:
    Kusy, M.
    Powiązania:
    https://bibliotekanauki.pl/articles/226697.pdf
    Data publikacji:
    2015
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    probabilistic neural network
    dimensionality reduction
    feature selection
    feature extraction
    single decision tree
    random forest
    principal component analysis
    prediction ability
    Opis:
    This article presents the study regarding the problem of dimensionality reduction in training data sets used for classification tasks performed by the probabilistic neural network (PNN). Two methods for this purpose are proposed. The first solution is based on the feature selection approach where a single decision tree and a random forest algorithm are adopted to select data features. The second solution relies on applying the feature extraction procedure which utilizes the principal component analysis algorithm. Depending on the form of the smoothing parameter, different types of PNN models are explored. The prediction ability of PNNs trained on original and reduced data sets is determined with the use of a 10-fold cross validation procedure.
    Źródło:
    International Journal of Electronics and Telecommunications; 2015, 61, 3; 289-300
    2300-1933
    Pojawia się w:
    International Journal of Electronics and Telecommunications
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Research on ship trajectory extraction based on multiattribute DBSCAN optimisation algorithm
    Autorzy:
    Xu, Xiaofeng
    Cui, Deqaing
    Li, Yun
    Xiao, Yingjie
    Powiązania:
    https://bibliotekanauki.pl/articles/1551877.pdf
    Data publikacji:
    2021
    Wydawca:
    Politechnika Gdańska. Wydział Inżynierii Mechanicznej i Okrętownictwa
    Tematy:
    clustering algorithm
    abnormal route
    DBSCAN
    feature trajectory extraction
    fitting analysis
    Opis:
    With the vigorous development of maritime traffic, the importance of maritime navigation safety is increasing day by day. Ship trajectory extraction and analysis play an important role in ensuring navigation safety. At present, the DBSCAN (density-based spatial clustering of applications with noise) algorithm is the most common method in the research of ship trajectory extraction, but it has shortcomings such as missing ship trajectories in the process of trajectory division. The improved multi-attribute DBSCAN algorithm avoids trajectory division and greatly reduces the probability of missing sub-trajectories. By introducing the position, speed and heading of the ship track point, dividing the complex water area and vectorising the ship track, the function of guaranteeing the track integrity can be achieved and the ship clustering effect can be better realised. The result shows that the cluster fitting effect reaches up to 99.83%, which proves that the multi-attribute DBSCAN algorithm and cluster analysis algorithm have higher reliability and provide better theoretical guidance for the analysis of ship abnormal behaviour.
    Źródło:
    Polish Maritime Research; 2021, 1; 136-148
    1233-2585
    Pojawia się w:
    Polish Maritime Research
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Enhanced algorithm for energy optimization and improvised synchronization in knee exoskeleton system
    Autorzy:
    Arunamithra, J.
    Saravanan, R.
    Venkatesh Babu, S.
    Powiązania:
    https://bibliotekanauki.pl/articles/24200592.pdf
    Data publikacji:
    2022
    Wydawca:
    Stowarzyszenie Komputerowej Nauki o Materiałach i Inżynierii Powierzchni w Gliwicach
    Tematy:
    knee exoskeleton
    feature extraction
    data classification
    ANN algorithm
    egzoszkielet kolana
    ekstrakcja cech
    klasyfikacja danych
    algorytm ANN
    Opis:
    Purpose: The purpose of the study is to develop an augmented algorithm with optimised energy and improvised synchronisation to assist the knee exoskeleton design. This enhanced algorithm is used to estimate the accurate left and right movement signals from the brain and accordingly moves the lower-limb exoskeleton with the help of motors. Design/methodology/approach: An optimised deep learning algorithm is developed to differentiate the right and left leg movements from the acquired brain signals. The obtained test signals are then compared with the signals obtained from the conventional algorithm to find the accuracy of the algorithm. Findings: The obtained average accuracy rate of about 63% illustrates the improvised differentiation in identifying the right and left leg movement. Research limitations/implications: The future work involves the comparative study of the proposed algorithm with other classification technologies to extract more reliable results. A comparative analysis of the replaceable and rechargeable battery will be done in the future study to exhibit the effectiveness of the proposed model. Originality/value: This study involves the extended study of five frequency regions namely alpha, beta, gamma, delta and theta, to handle the real-time EEG signal processing exoskeleton, model.
    Źródło:
    Archives of Materials Science and Engineering; 2022, 117, 2; 79--85
    1897-2764
    Pojawia się w:
    Archives of Materials Science and Engineering
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Analysis of suspicious lesions in digital mammograms
    Autorzy:
    Choraś, R. S.
    Powiązania:
    https://bibliotekanauki.pl/articles/333369.pdf
    Data publikacji:
    2011
    Wydawca:
    Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
    Tematy:
    sterowane filtry
    mammografie
    wychwytywanie cech
    klasyfikacja
    steerable filters
    mammograms
    moment and texture features
    feature extraction
    classification
    Opis:
    The system using steerable filters for analysis suspicious lesions in mammograms is proposed. This system is based on moments and texture features. The set of well defined and classified suspicious lesions regions from mammograms database are used as a reference pattern. The similarity measure for reference pattern image and patient mammogram is found by computing the distance between their corresponding feature vectors. The Euclidean distance metric is used to finding the nearest class to patient feature vector what in result mark the automatically classify this mammograms.
    Źródło:
    Journal of Medical Informatics & Technologies; 2011, 17; 151-158
    1642-6037
    Pojawia się w:
    Journal of Medical Informatics & Technologies
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Optical character recognition using artifical inelligence technologies
    Optyczne rozpoznawanie znaków z użyciem sztucznej inteligencji
    Autorzy:
    Musiał, A.
    Szczepaniak, P.
    Powiązania:
    https://bibliotekanauki.pl/articles/408862.pdf
    Data publikacji:
    2014
    Wydawca:
    Politechnika Lubelska. Wydawnictwo Politechniki Lubelskiej
    Tematy:
    character recognition
    artificial intelligence
    feature extraction
    clustering algorithms
    rozpoznawanie znaków
    sztuczna inteligencja
    ekstrakcja cech
    algorytmy klastrowania
    Opis:
    The article represents results of the research of an Optical Character Recognition system. Proposed OCR system is able to convert a raster image into the text string, which represents the text shown on the input image. The main innovation is the fact that the system was created without following any strict rules. It was more an innovative research rather than simple programming using ready guidelines.
    Celem projektu opisywanego w artykule było przygotowanie działającego systemu do optycznego rozpoznawania znaków, tj. zdolnego przekształcić rastrowy obraz wejściowy w łańcuch znaków odpowiadający zapisanemu tekstowi na obrazie. Nowością jest m.in. fakt wykonania tego systemu bez podążania za z góry znaną architekturą aplikacji, a przygotowanie go w sposób bardziej doświadczalny, czyli wykorzystując podejście nowatorskie.
    Źródło:
    Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska; 2014, 2; 41-44
    2083-0157
    2391-6761
    Pojawia się w:
    Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    An AI & ML based detection & identification in remote imagery: state-of-the-art
    Autorzy:
    Hashmi, Hina
    Dwivedi, Rakesh
    Kumar, Anil
    Powiązania:
    https://bibliotekanauki.pl/articles/2141786.pdf
    Data publikacji:
    2021
    Wydawca:
    Sieć Badawcza Łukasiewicz - Przemysłowy Instytut Automatyki i Pomiarów
    Tematy:
    convolutional neural network
    remote sensed imagery
    object detection
    artificial intelligence
    feature extraction
    deep learning
    machine learning
    Opis:
    Remotely sensed images and their allied areas of application have been the charm for a long time among researchers. Remote imagery has a vast area in which it is serving and achieving milestones. From the past, after the advent of AL, ML, and DL-based computing, remote imagery is related techniques for processing and analyzing are continuously growing and offering countless services like traffic surveillance, earth observation, land surveying, and other agricultural areas. As Artificial intelligence has become the charm of researchers, machine learning and deep learning have been proven as the most commonly used and highly effective techniques for object detection. AI & ML-based object segmentation & detection makes this area hot and fond to the researchers again with the opportunities of enhanced accuracy in the same. Several researchers have been proposed their works in the form of research papers to highlight the effectiveness of using remotely sensed imagery for commercial purposes. In this article, we have discussed the concept of remote imagery with some preprocessing techniques to extract hidden and fruitful information from them. Deep learning techniques applied by various researchers along with object detection, object recognition are also discussed here. This literature survey is also included a chronological review of work done related to detection and recognition using deep learning techniques.
    Źródło:
    Journal of Automation Mobile Robotics and Intelligent Systems; 2021, 15, 4; 3-17
    1897-8649
    2080-2145
    Pojawia się w:
    Journal of Automation Mobile Robotics and Intelligent Systems
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Marine Mammals Classification using Acoustic Binary Patterns
    Autorzy:
    Nadir, Maheen
    Adnan, Syed M.
    Aziz, Sumair
    Khan, Muhammad Umar
    Powiązania:
    https://bibliotekanauki.pl/articles/1953520.pdf
    Data publikacji:
    2020
    Wydawca:
    Polska Akademia Nauk. Czasopisma i Monografie PAN
    Tematy:
    marine mammals
    1D Local Binary Patterns
    Mel frequency cepstral coefficients
    feature extraction
    passive acoustic monitoring
    Opis:
    Marine mammal identification and classification for passive acoustic monitoring remain a challenging task. Mainly the interspecific and intraspecific variations in calls within species and among different individuals of single species make it more challenging. Varieties of species along with geographical diversity induce more complications towards an accurate analysis of marine mammal classification using acoustic signatures. Prior methods for classification focused on spectral features which result in increasing bias for contour base classifiers in automatic detection algorithms. In this study, acoustic marine mammal classification is performed through the fusion of 1D Local Binary Pattern (1D-LBP) and Mel Frequency Cepstral Coefficient (MFCC) based features. Multi-class Support Vector Machines (SVM) classifier is employed to identify different classes of mammal sounds. Classification of six species named Tursiops truncatus, Delphinus delphis, Peponocephala electra, Grampus griseus, Stenella longirostris, and Stenella attenuate are targeted in this research. The proposed model achieved 90.4% accuracy on 70-30% training testing and 89.6% on 5-fold cross-validation experiments.
    Źródło:
    Archives of Acoustics; 2020, 45, 4; 721-731
    0137-5075
    Pojawia się w:
    Archives of Acoustics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A simultaneous localization and tracking method for a worm tracking system
    Autorzy:
    Kowalski, M.
    Kaczmarek, P.
    Kabaciński, R.
    Matuszczak, M.
    Tranbowicz, K.
    Sobkowiak, R.
    Powiązania:
    https://bibliotekanauki.pl/articles/330526.pdf
    Data publikacji:
    2014
    Wydawca:
    Uniwersytet Zielonogórski. Oficyna Wydawnicza
    Tematy:
    Caenorhabditis elegans behavior
    worm tracking
    computer vision
    image processing
    feature extraction
    wizja komputerowa
    przetwarzanie obrazu
    ekstrakcja cech
    Opis:
    The idea of worm tracking refers to the path analysis of Caenorhabditis elegans nematodes and is an important tool in neurobiology which helps to describe their behavior. Knowledge about nematode behavior can be applied as a model to study the physiological addiction process or other nervous system processes in animals and humans. Tracking is performed by using a special manipulator positioning a microscope with a camera over a dish with an observed individual. In the paper, the accuracy of a nematode’s trajectory reconstruction is investigated. Special attention is paid to analyzing errors that occurred during the microscope displacements. Two sources of errors in the trajectory reconstruction are shown. One is due to the difficulty in accurately measuring the microscope shift, the other is due to a nematode displacement during the microscope movement. A new method that increases path reconstruction accuracy based only on the registered sequence of images is proposed. The method Simultaneously Localizes And Tracks (SLAT) the nematodes, and is robust to the positioning system displacement errors. The proposed method predicts the nematode position by using NonParametric Regression (NPR). In addition, two other methods of the SLAT problem are implemented to evaluate the NPR method. The first consists in ignoring the nematode displacement during microscope movement, and the second is based on a Kalman filter. The results suggest that the SLAT method based on nonparametric regression gives the most promising results and decreases the error of trajectory reconstruction by 25% compared with reconstruction based on data from the positioning system.
    Źródło:
    International Journal of Applied Mathematics and Computer Science; 2014, 24, 3; 599-609
    1641-876X
    2083-8492
    Pojawia się w:
    International Journal of Applied Mathematics and Computer Science
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Image retrieval based on hierarchical Gabor filters
    Autorzy:
    Andrysiak, T.
    Choraś, M.
    Powiązania:
    https://bibliotekanauki.pl/articles/908448.pdf
    Data publikacji:
    2005
    Wydawca:
    Uniwersytet Zielonogórski. Oficyna Wydawnicza
    Tematy:
    filtr Gabora
    pobieranie obrazu
    ekstrakcja cech
    reprezentacja hierarchiczna
    Gabor filters
    image retrieval
    texture feature extraction
    hierarchical representation
    Opis:
    Content Based Image Retrieval (CBIR) is now a widely investigated issue that aims at allowing users of multimedia information systems to automatically retrieve images coherent with a sample image. A way to achieve this goal is the computation of image features such as the color, texture, shape, and position of objects within images, and the use of those features as query terms. We propose to use Gabor filtration properties in order to find such appropriate features. The article presents multichannel Gabor filtering and a hierarchical image representation. Then a salient (characteristic) point detection algorithm is presented so that texture parameters are computed only in a neighborhood of salient points. We use Gabor texture features as image content descriptors and efficiently emply them to retrieve images.
    Źródło:
    International Journal of Applied Mathematics and Computer Science; 2005, 15, 4; 471-480
    1641-876X
    2083-8492
    Pojawia się w:
    International Journal of Applied Mathematics and Computer Science
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Hardware acceleration of data classifiers for multimedia processing tasks
    Sprzętowe przyspieszenie klasyfikacji danych multimedialnych
    Autorzy:
    Dziurzański, P.
    Mąka, T.
    Forczmański, P.
    Powiązania:
    https://bibliotekanauki.pl/articles/153826.pdf
    Data publikacji:
    2014
    Wydawca:
    Stowarzyszenie Inżynierów i Techników Mechaników Polskich
    Tematy:
    feature extraction
    multimedia data classification
    Network on Chip (NoC)
    ImpulseC
    ekstrakcja cech
    klasyfikacja danych multimedialnych
    sieci wewnątrzukładowe
    Opis:
    In this paper, experimental results of a proposed hardware acceleration of feature extraction and data classifiers for multimedia are presented. This hardware is based on multi-core architecture connected with a mesh Network on Chip (NoC). The cores in the system execute both data classifiers and feature extraction for audio and image data. Using various meta heuristics the system is optimized with regards to different data communication criteria. The system was implemented on an FPGA platform with use of ImpulseC hardware description language.
    W artykule zostały zeprezentowane wyniki eksperymentalne dotyczące sprzętowego przyspieszania ekstrakcji cech i klasyfikacji danych multimedialnych. Opracowane rozwiązanie sprzętowe bazuje na architekturze wielordzeniowej, w której każdy blok realizuje przypisaną mu statycznie funkcjonalność. Rdzenie połączone są ze sobą za pomocą sieci wewnątrzukładowej (ang. Network on Chip, NoC) o architekturze siatki. W artykule opisano pokrótce autorskie oprogramowanie służące do generowania kodu sieci wewnątrzukładowej. Graficzny interfejs użytkownika został zaprezentowany na rys. 1. Narzędzie ma za zadanie dokonywać odwzorowania wybranych funkcjonalności do poszczególnych rdzeni z wykorzystaniem takich meta-heurystyk jak algorytmy genetyczne, symulowane wyżarzanie, poszukiwanie losowe czy algorytmu gradientowego. Jako kryterium optymalizacji można wybrać minimalizację całkowitego przesyłu danych, minimalizację maksymalnej liczby danych transmitowanych przez pojedyncze łącze, a także minimalizację odchylenia standardowego rozmiaru strumieni transmitowanych przez poszczególne łącza. Przykładowe wyniki optymalizacji losowej dla sieci wewnątrzukładowej zostały przedstawione w tab. 1, natomiast wyniki optymalizacji dla sieci wewnątrzukładowej wykorzystującej inne podejścia - w tab. 2. Dla systemu zoptymalizowanego w ten sposób został wygnerowany opisujący go kod w języku ImpulseC, który następnie posłużył do syntezy sprzętowej na układzie FPGA z rodziny Xilinx Virtex 5. Zajętość układu XC5VSX50T dla trzech wykorzystanych klasyfikatorów została przedstawiona na rys. 3. Z kolei tab. 3 przedstawia liczbę zasobów wykorzystanych przez narzędzie syntezy wysokiego poziomu dla tych klasyfikatorów. Technika przedstawiona w publikacji umożliwia określenie warunków i ograniczeń implementacji sprzętowej systemu służącego klasyfikacji danych multimedialnych.
    Źródło:
    Pomiary Automatyka Kontrola; 2014, R. 60, nr 6, 6; 382-384
    0032-4140
    Pojawia się w:
    Pomiary Automatyka Kontrola
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Grain size determination and classification using adaptive image segmentation with grain shape information for milling quality evaluation
    Określenie rozmiaru ziarna i klasyfikacja z użyciem adaptacyjnej segmentacji obrazu i informacji o kształcie dla oceny jakości mielenia
    Autorzy:
    Budzan, S.
    Pawełczyk, M.
    Powiązania:
    https://bibliotekanauki.pl/articles/328384.pdf
    Data publikacji:
    2018
    Wydawca:
    Polska Akademia Nauk. Polskie Towarzystwo Diagnostyki Technicznej PAN
    Tematy:
    grain classification
    particle analysis
    image segmentation
    feature extraction
    klasyfikacja ziaren
    analiza
    wielkość ziaren
    segmentacja obrazów
    ekstrakcja cech
    Opis:
    In this paper, authors described methods of material granularity evaluation and a novel method for grain size determination with inline electromagnetic mill device diagnostics. The milling process quality evaluation can be carried out with vibration measurements, analysis of the milling material images or well-known screening machines. The method proposed in this paper is developed to the online examination of the milled product during the milling process using real-time digital images. In this paper, authors concentrated their work on copper ore milling process. Determination of the total number of the grain, the size of each grain, also the classification of the grains are the main goal of the developed method. In the proposed method the vision camera with lightning mounted at two assumed angles has been used. The detection of the grains has been based on an adaptive segmentation algorithm, improved with distance transform to enhance grains detection. The information about particles shape and context is used to optimize the grain classification process in the next step. The final classification is based on the rule-based method with defined particle shape and size parameters.
    W pracy autorzy opisali metody oceny uziarnienia materiału i nową metodę określania wielkości ziaren z jednoczesną diagnostyką pracy młyna elektromagnetycznego. Ocena jakości mielenia może być realizowana na kilka sposobów, tj. poprzez pomiar drgań, analizę obrazów materiału zmielonego, lub wykorzystanie matryc przesiewowych. Proces mielenia jest procesem obciążonym znacznym zużyciem energii, dlatego proces diagnostyki powinien być wykonywany z dużą efektywnością. Metoda zaproponowana w niniejszym artykule opiera się na badaniu mielonego produktu podczas procesu mielenia przy użyciu analizy obrazów cyfrowych w czasie rzeczywistym. Głównym celem opracowanej metody jest określenie całkowitej liczby ziaren, wielkości ziaren, jak i klasyfikacja ziaren. W zaproponowanej metodzie wykorzystano akwizycję obrazów z kamery przy oświetlaniu badanych próbek pod kątem, co pozwala zwiększyć liczbę wykrywanych ziaren. Detekcja ziaren bazuje na metodzie segmentacji adaptacyjnej rozszerzonej o analizę map odległościowych w celu poprawienia jakości i liczby wykrytych ziaren. Informacje na temat kształtu ziaren są wykorzystywane w celu optymalizacji procesu klasyfikacji ziaren. Ostateczna klasyfikacja opiera się na metodzie bazującej na regułach, w których określono zależności dla różnych parametrów kształtu i rozmiaru ziaren.
    Źródło:
    Diagnostyka; 2018, 19, 1; 41-48
    1641-6414
    2449-5220
    Pojawia się w:
    Diagnostyka
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Automatic Genre Classification Using Fractional Fourier Transform Based Mel Frequency Cepstral Coefficient and Timbral Features
    Autorzy:
    Bhalke, D. G.
    Rajesh, B.
    Bormane, D. S.
    Powiązania:
    https://bibliotekanauki.pl/articles/177599.pdf
    Data publikacji:
    2017
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    feature extraction
    Timbral features
    MFCC
    Mel Frequency Cepstral Coefficient
    FrFT
    fractional Fourier transform
    Fractional MFCC
    Tamil Carnatic music
    Opis:
    This paper presents the Automatic Genre Classification of Indian Tamil Music and Western Music using Timbral and Fractional Fourier Transform (FrFT) based Mel Frequency Cepstral Coefficient (MFCC) features. The classifier model for the proposed system has been built using K-NN (K-Nearest Neighbours) and Support Vector Machine (SVM). In this work, the performance of various features extracted from music excerpts has been analysed, to identify the appropriate feature descriptors for the two major genres of Indian Tamil music, namely Classical music (Carnatic based devotional hymn compositions) & Folk music and for western genres of Rock and Classical music from the GTZAN dataset. The results for Tamil music have shown that the feature combination of Spectral Roll off, Spectral Flux, Spectral Skewness and Spectral Kurtosis, combined with Fractional MFCC features, outperforms all other feature combinations, to yield a higher classification accuracy of 96.05%, as compared to the accuracy of 84.21% with conventional MFCC. It has also been observed that the FrFT based MFCC effieciently classifies the two western genres of Rock and Classical music from the GTZAN dataset with a higher classification accuracy of 96.25% as compared to the classification accuracy of 80% with MFCC.
    Źródło:
    Archives of Acoustics; 2017, 42, 2; 213-222
    0137-5075
    Pojawia się w:
    Archives of Acoustics
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A proposal of biologically inspired hierarchical approach to object recognition
    Autorzy:
    Kozik, R.
    Powiązania:
    https://bibliotekanauki.pl/articles/333962.pdf
    Data publikacji:
    2013
    Wydawca:
    Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
    Tematy:
    object recognition
    hierarchical model
    feature extraction
    visual cortex
    computer vision
    rozpoznawanie obiektów
    model hierarchiczny
    kora wzrokowa
    wizja komputerowa
    Opis:
    In this article a biologically-inspired algorithm for object recognition is presented. The approach is based on a hierarchical HMAX cortex model that was initially proposed by Riesenhuber and Poggio [12] and later extended by Serre et al [13]. The results show that despite the modification that were undertaken to simplify the HMAX model (in order to make it feasible for a real-time solutions) it is possible to achieve high effectiveness for a one-class detection problems. Moreover, it is also demonstrated how the proposed algorithm can be successfully deployed on a low-cost Android smartphone.
    Źródło:
    Journal of Medical Informatics & Technologies; 2013, 22; 169-176
    1642-6037
    Pojawia się w:
    Journal of Medical Informatics & Technologies
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Classification of EEG signal by methods of machine learning
    Autorzy:
    Alyamani, Amina
    Yasniy, Oleh
    Powiązania:
    https://bibliotekanauki.pl/articles/1837774.pdf
    Data publikacji:
    2020
    Wydawca:
    Polskie Towarzystwo Promocji Wiedzy
    Tematy:
    machine learning
    EEG signal
    classification
    data balancing
    feature extraction
    uczenie maszynowe
    sygnał EEG
    klasyfikacja
    równoważenie danych
    ekstrakcja cech
    Opis:
    Electroencephalogram (EEG) signal of two healthy subjects that was available from literature, was studied using the methods of machine learning, namely, decision trees (DT), multilayer perceptron (MLP), K-nearest neighbours (kNN), and support vector machines (SVM). Since the data were imbalanced, the appropriate balancing was performed by Kmeans clustering algorithm. The original and balanced data were classified by means of the mentioned above 4 methods. It was found, that SVM showed the best result for the both datasets in terms of accuracy. MLP and kNN produce the comparable results which are almost the same. DT accuracies are the lowest for the given dataset, with 83.82% for the original data and 61.48% for the balanced data.
    Źródło:
    Applied Computer Science; 2020, 16, 4; 56-63
    1895-3735
    Pojawia się w:
    Applied Computer Science
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Music Playlist Generation using Facial Expression Analysis and Task Extraction
    Autorzy:
    Sen, A.
    Popat, D.
    Shah, H.
    Kuwor, P.
    Johri, E.
    Powiązania:
    https://bibliotekanauki.pl/articles/908868.pdf
    Data publikacji:
    2016
    Wydawca:
    Uniwersytet Marii Curie-Skłodowskiej. Wydawnictwo Uniwersytetu Marii Curie-Skłodowskiej
    Tematy:
    facial expression analysis
    emotion recognition
    feature extraction
    viola jones face detection
    gabor filter
    adaboost
    k-NN algorithm
    task extraction
    music classification
    playlist generation
    Opis:
    In day to day stressful environment of IT Industry, there is a truancy for the appropriate relaxation time for all working professionals. To keep a person stress free, various technical or non-technical stress releasing methods are now being adopted. We can categorize the people working on computers as administrators, programmers, etc. each of whom require varied ways in order to ease themselves. The work pressure and the vexation of any kind for a person can be depicted by their emotions. Facial expressions are the key to analyze the current psychology of the person. In this paper, we discuss a user intuitive smart music player. This player will capture the facial expressions of a person working on the computer and identify the current emotion. Intuitively the music will be played for the user to relax them. The music player will take into account the foreground processes which the person is executing on the computer. Since various sort of music is available to boost one's enthusiasm, taking into consideration the tasks executed on the system by the user and the current emotions they carry, an ideal playlist of songs will be created and played for the person. The person can browse the playlist and modify it to make the system more flexible. This music player will thus allow the working professionals to stay relaxed in spite of their workloads.
    Źródło:
    Annales Universitatis Mariae Curie-Skłodowska. Sectio AI, Informatica; 2016, 16, 2; 1-6
    1732-1360
    2083-3628
    Pojawia się w:
    Annales Universitatis Mariae Curie-Skłodowska. Sectio AI, Informatica
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Comparative analysis of selected classifiers in posterior cruciate ligaments computer aided diagnosis
    Autorzy:
    Zarychta, P.
    Badura, P.
    Pietka, E.
    Powiązania:
    https://bibliotekanauki.pl/articles/200544.pdf
    Data publikacji:
    2017
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    posterior cruciate ligament
    computer aided diagnosis
    feature extraction
    classification
    soft computing
    więzadło krzyżowe tylne
    diagnostyka wspierana komputerowo
    klasyfikacja
    obliczenia miękkie
    Opis:
    A study on computer aided diagnosis of posterior cruciate ligaments is presented in this paper. The diagnosis relies on T1-weighted magnetic resonance imaging. During the image analysis stage, the ligament region is automatically detected, localized, and extracted using fuzzy segmentation methods. Eight geometric features are defined for the ligament object. With a clinical reference database containing 107 cases of both healthy and pathological cases, a Fisher linear discriminant is used to select 4 most distinctive features. At the classification stage we employ five different soft computing classifiers to evaluate the feature vector suitability for the computerized ligament diagnosis. Among the classifiers we introduce and specify the particle swarm optimization based Sugeno-type fuzzy inference system and compare its performance to other established classification systems. The classification accuracy metrics: sensitivity, specificity, and Dice index all exceed 90% for each classifier under consideration, indicating high level of the proposed feature vector relevance in the computer aided ligaments diagnosis.
    Źródło:
    Bulletin of the Polish Academy of Sciences. Technical Sciences; 2017, 65, 1; 63-70
    0239-7528
    Pojawia się w:
    Bulletin of the Polish Academy of Sciences. Technical Sciences
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Decision trees and the effects of feature extraction parameters for robust sensor network design
    Wykorzystanie drzew decyzyjnych oraz wpływu parametrów ekstrakcji cech do projektowania odpornych sieci czujników
    Autorzy:
    Gerdes, M.
    Galar, D.
    Scholz, D.
    Powiązania:
    https://bibliotekanauki.pl/articles/301345.pdf
    Data publikacji:
    2017
    Wydawca:
    Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN
    Tematy:
    decision trees
    feature extraction
    sensor optimization
    sensor fusion
    sensor selection
    drzewa decyzyjne
    ekstrakcja cech
    optymalizacja czujników
    fuzja czujników
    dobór czujników
    Opis:
    Reliable sensors and information are required for reliable condition monitoring. Complex systems are commonly monitored by many sensors for health assessment and operation purposes. When one of the sensors fails, the current state of the system cannot be calculated in same reliable way or the information about the current state will not be complete. Condition monitoring can still be used with an incomplete state, but the results may not represent the true condition of the system. This is especially true if the failed sensor monitors an important system parameter. There are two possibilities to handle sensor failure. One is to make the monitoring more complex by enabling it to work better with incomplete data; the other is to introduce hard or software redundancy. Sensor reliability is a critical part of a system. Not all sensors can be made redundant because of space, cost or environmental constraints. Sensors delivering significant information about the system state need to be redundant, but an error of less important sensors is acceptable. This paper shows how to calculate the significance of the information that a sensor gives about a system by using signal processing and decision trees. It also shows how signal processing parameters influence the classification rate of a decision tree and, thus, the information. Decision trees are used to calculate and order the features based on the information gain of each feature. During the method validation, they are used for failure classification to show the influence of different features on the classification performance. The paper concludes by analysing the results of experiments showing how the method can classify different errors with a 75% probability and how different feature extraction options influence the information gain.
    Niezawodne monitorowanie stanu wymaga niezawodności czujników i pochodzących z nich informacji. Systemy złożone są zazwyczaj monitorowane przez wiele czujników, co pozwala na ocenę stanu technicznego oraz aspektów eksploatacyjnych. Gdy jeden z czujników ulega uszkodzeniu, uniemożliwia to obliczenie bieżącego stanu systemu z dotychczasową niezawodnością lub uzyskanie kompletnych informacji o bieżącym stanie. Stan można co prawda monitorować nawet przy niekompletnych danych, ale wyniki takiego monitorowania mogą nie odpowiadać rzeczywistemu stanowi systemu. Sytuacja taka ma miejsce w szczególności, gdy uszkodzony czujnik jest odpowiedzialny za monitorowanie istotnego parametru systemu. Problem uszkodzenia czujnika można rozwiązywać na dwa sposoby. Pierwszy polega na zwiększeniu złożoności systemu, co umożliwia jego sprawniejsze działanie w sytuacji, gdy dane są niekompletne. Drugim sposobem jest wprowadzenie nadmiarowego sprzętu (hardware'u) lub oprogramowania. Niezawodność czujników stanowi krytyczny aspekt systemu. Oczywiście, ze względu na ograniczenia przestrzenne, ekonomiczne i środowiskowe nie wszystkie czujniki w systemie mogą być nadmiarowe. Redundancja powinna dotyczyć wszystkich czujników, które dostarczają istotnych informacji na temat stanu systemu, natomiast dopuszczalne są błędy mniej ważnych czujników. W niniejszej pracy pokazano jak obliczać istotność informacji o systemie dostarczanych przez poszczególne czujniki z wykorzystaniem metod przetwarzania sygnałów oraz drzew decyzyjnych. Zademonstrowano również w jaki sposób parametry przetwarzania sygnałów wpływają na poprawność klasyfikacji metodą drzewa decyzyjnego, a tym samym na poprawność dostarczanych informacji. Drzew decyzyjnych używa się do obliczania i porządkowania cech w oparciu o przyrost informacji charakteryzujący poszczególne cechy. Podczas weryfikacji zastosowanej metody, drzewa decyzyjne wykorzystano do klasyfikacji uszkodzeń celem przedstawienia wpływu różnych cech na dokładność klasyfikacji. Pracę kończy analiza wyników eksperymentów pokazujących w jaki sposób zastosowana metoda pozwala na klasyfikację różnych błędów z 75-procentowym prawdopodobieństwem oraz jak różne opcje ekstrakcji cech wpływają na przyrost informacji.
    Źródło:
    Eksploatacja i Niezawodność; 2017, 19, 1; 31-42
    1507-2711
    Pojawia się w:
    Eksploatacja i Niezawodność
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Pattern Recognition Methods for Detecting Voltage Sag Disturbances and Electromagnetic Interference in Smart Grids
    Autorzy:
    Yalcin, T.
    Ozdemir, M.
    Powiązania:
    https://bibliotekanauki.pl/articles/136160.pdf
    Data publikacji:
    2016
    Wydawca:
    EEEIC International Barbara Leonowicz Szabłowska
    Tematy:
    C4.5 decision trees
    electromagnetic interference
    feature extraction
    hilbert huang transform
    power quality disturbance
    smart grids
    support vector machines
    Opis:
    Identification of system disturbances, detection of them guarantees smart grids power quality (PQ) system reliability and provides long lasting life of the power system. The key goal of this study is to find the best accuracy of identification algorithm for non-stationary, non-linear power quality disturbances such as voltage sag, electromagnetic interference in smart grids. PQube, power quality and energy monitor, was used to acquire these distortions. Ensemble Empirical Mode Decomposition is used for electromagnetic interference reduction with first intrinsic mode function. Hilbert Huang Transform is used for generating instantaneous amplitude and instantaneous frequency feature of real time voltage sag power signal. Outputs of Hilbert Huang Transform is intrinsic mode functions (IMFs), instantaneous frequency (IF), and instantaneous amplitude (IA). Characteristic features are obtained from first IMFs, IF, and IA. The six features—, the mean, standard deviation,skewness, kurtosis of both IF and IA are then calculated. These features are normalized along with the inputs classifiers. The proposed power system monitoring system is able to detect power system voltage sag disturbances and capable of recognize electromagnetic interference component. In this study based on experimental studies, Hilbert Huang Transform based pattern recognition technique was used to investigate power signal to diagnose voltage sag and in power grid. Support Vector Machines and C4.5 Decision Tree were operated and their achievements were matched for precision and CPU timing. According to the analysis, decision tree algorithm without dimensionality reduction produces the best solution.
    Źródło:
    Transactions on Environment and Electrical Engineering; 2016, 1, 3; 86-93
    2450-5730
    Pojawia się w:
    Transactions on Environment and Electrical Engineering
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Cortical pattern detection for the developing brain: a 3D vertex labeling and skeletonization approach
    Autorzy:
    Clouchoux, C.
    Kudelski, D.
    Bouyssi-Kobar, M.
    Viseur, S.
    du Plessis, A.
    Evans, A.
    Mari, J.-L.
    Limperopoulos, C.
    Powiązania:
    https://bibliotekanauki.pl/articles/333059.pdf
    Data publikacji:
    2010
    Wydawca:
    Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
    Tematy:
    rozwój mózgu
    ocena krzywizny
    wychwytywanie cech
    sulcal pattern
    brain development
    in-vivo MRI
    cortical surface
    curvature estimation
    morphological operators
    feature extraction
    Opis:
    Normal brain development is associated with expansion and folding of the cerebral cortex in a normal sequence of gyral–sulcal formation. We propose a global approach for measuring the cortical folding pattern of the developing brain. Our method measures geometric features directly on the cortical surface mesh, based on vertex labeling and skeletonization. The resulting extraction provides an accurate representation of global cortical organization. We applied this method to 17 young infants in order to characterize the evolution of cortical organization in the developing brain.
    Źródło:
    Journal of Medical Informatics & Technologies; 2010, 16; 161-166
    1642-6037
    Pojawia się w:
    Journal of Medical Informatics & Technologies
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Rgb-D face recognition using LBP-DCT algorithm
    Autorzy:
    Kumar, Sunil B L
    Kumari, Sharmila M
    Powiązania:
    https://bibliotekanauki.pl/articles/1956066.pdf
    Data publikacji:
    2021
    Wydawca:
    Polskie Towarzystwo Promocji Wiedzy
    Tematy:
    RGB-D
    kinect
    local binary pattern
    pattern recognition
    feature extraction
    histogram
    face recognition
    lokalny wzorzec binarny
    rozpoznawanie wzorców
    wyodrębnianie cech
    rozpoznawanie twarzy
    Opis:
    Face recognition is one of the applications in image processing that recognizes or checks an individual's identity. 2D images are used to identify the face, but the problem is that this kind of image is very sensitive to changes in lighting and various angles of view. The images captured by 3D camera and stereo camera can also be used for recognition, but fairly long processing times is needed. RGB-D images that Kinect produces are used as a new alternative approach to 3D images. Such cameras cost less and can be used in any situation and any environment. This paper shows the face recognition algorithms’ performance using RGB-D images. These algorithms calculate the descriptor which uses RGB and Depth map faces based on local binary pattern. Those images are also tested for the fusion of LBP and DCT methods. The fusion of LBP and DCT approach produces a recognition rate of 97.5% during the experiment
    Źródło:
    Applied Computer Science; 2021, 17, 3; 73-81
    1895-3735
    Pojawia się w:
    Applied Computer Science
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Additional data preprocessing and feature extraction in automatic classification of heartbeats
    Dodatkowe przetwarzanie wstępne i ekstrakcja cech w procesie automatycznej klasyfikacji rytmu serca
    Autorzy:
    Tadejko, P.
    Powiązania:
    https://bibliotekanauki.pl/articles/341075.pdf
    Data publikacji:
    2007
    Wydawca:
    Politechnika Białostocka. Oficyna Wydawnicza Politechniki Białostockiej
    Tematy:
    ECG
    przetwarzanie wstępne
    morfologia matematyczna
    filtrowanie ECG
    ekstrakcja cech
    klasyfikacja rytmu serca
    preprocessing
    mathematical morphology
    ECG filtering
    wavelet approximation
    feature extraction
    heartbeat classification
    Opis:
    The paper presents the classification performance of an automatic classifier of the electrocardiogram (ECG) for the detection abnormal beats with new concept of feature extraction stage. Feature sets were based on ECG morphology and RR-intervals. This paper compares two strategies for classification of annotated QRS complexes: based on original ECG morphology features and proposed new approach - based on preprocessed ECG morphology features. The mathematical morphology filtering and wavelet trans-form is used for the preprocessing of ECG signal. Within this framework, the problem of choosing an appropriate structuring element in mathematical morphology filtering in signal processing was studied. Configuration adopted a Kohonen self-organizing maps (SOM) and Support Vector Machine (SVM) for analysis of signal features and clustering. In this study, a classifiers was developed with LVQ and SVM algorithms using the data from the records recommended by ANSI/AAMI EC57 standard. The performance of the algorithm is evaluated on the MIT-BIH Arrhythmia Database following the AAMI recommendations. Using this method the results of identify beats either as normal or arrhythmias was improved.
    Artykuł prezentuje nowe podejście do problemu klasyfikacji zapisów ECG w celu detekcji zachowań chorobowych. Podstawą koncepcji fazy ekstrakcji cech jest proces przetwarzania wstępnego sygnału ECG z wykorzystaniem morfologii matematycznej oraz innych transformacji. Morfologia matematyczna bazując na teorii zbiorów, pozwala zmienić charakterystyczne elementy sygnału. Dwie podstawowe operacje: dylatacja i erozja pozwalają na uwydatnienie lub redukcję wielkości i kształtu określonych elementów w danych. Parametry charakterystyki zapisów ECG stanowią bazę dla wektora cech. Do klasyfikacji przebiegów ECG w pracy wykorzystano samoorganizujące się mapy (SOM) Kohonena z klasyfikatorem LVQ oraz algorytm Support Vector Machines (SVM). Eksperymenty przeprowadzono klasyfikując sygnały pomiędzy trzynaście kategorii rekomendowanych przez standard ANSI/AAMI EC57, to jest: prawidłowy rytm serca i 12 arytmii. Zaproponowany w artykule algorytm opiera się na wykorzystaniu elementarnych operacji morfologii matematycznej i ich kombinacji. Ocenę wyników eksperymentów przeprowadzono na sygnałach z bazy MIT/BIH. Na tej podstawie zaproponowano wyjściową architekturę bloku filtrów morfologicznych dla celów ekstrakcji cech oraz unifikacji wejściowego sygnału ECG jako danych wejściowych do budowy wektora cech.
    Źródło:
    Zeszyty Naukowe Politechniki Białostockiej. Informatyka; 2007, 2; 155-173
    1644-0331
    Pojawia się w:
    Zeszyty Naukowe Politechniki Białostockiej. Informatyka
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Comparative Investigation on Objective Evaluation Methods for Fabric Smoothness
    Badanie porównawcze obiektywnych metod oceny gładkości tkaniny
    Autorzy:
    Liu, Chengxia
    Zheng, Xiaoping
    Powiązania:
    https://bibliotekanauki.pl/articles/231925.pdf
    Data publikacji:
    2020
    Wydawca:
    Sieć Badawcza Łukasiewicz - Instytut Biopolimerów i Włókien Chemicznych
    Tematy:
    AATCC replicas
    3D laser scanning
    image processing
    feature extraction
    fabric smoothness
    repliki AATCC
    skanowanie laserowe 3D
    przetwarzanie obrazu
    ekstrakcja cech
    gładkość tkaniny
    Opis:
    An objective method for fabric smoothness usually comprises two widely used approaches: 3D laser scanning and 2D image processing, which are represented by GLCM in this work. To make a comparison of them and find out which one is more effective, four 3D parameters (variance, roughness, torsion and interquartile deviation) and eight 2D parameters (mean value and standard deviation of energy, entropy, contrast and correlation) were extracted for AATCC SA replicas and fabrics. Results show that both 3D laser scanning and 2D image processing technology can be used to study smoothness. With regard to accuracy, the 3D laser scanning method is better than the 2D image processing method. Roughness in 3D parameters and the standard deviation of Entropy in 2D parameters have the highest correlation coefficient with the wrinkling grade of replicas, -0.965 and -0.917 respectively. The verification experiment of fabrics proves that roughness can characterise the wrinkling degree better as well. Furthermore, through the work of this paper, we find that the wrinkling degree differences between two adjacent AATCC SA replicas are not the same; the difference between SA-1 and SA-2 is significant, while that between SA-3 and SA-3.5 as well as SA-4 and SA-5 is not so obvious. It is advisable that the AATCC SA replicas for grades 3, 3.5, 4 and 5 be adjusted or improved.
    Obiektywna metoda oceny gładkości tkaniny zwykle obejmuje dwa szeroko stosowane podejścia: skanowanie laserowe 3D i przetwarzanie obrazu 2D, które w przedstawionej pracy są reprezentowane przez GLCM. Aby dokonać ich porównania i dowiedzieć się, który sposób jest bardziej skuteczny, wyodrębniono cztery parametry 3D (wariancja, chropowatość, skręcanie i odchylenie międzykwartylowe) i osiem parametrów 2D (wartość średnia i odchylenie standardowe energii, entropia, kontrast i korelacja). Wyniki pokazały, że do badania gładkości można wykorzystać zarówno skanowanie laserowe 3D, jak i technologię przetwarzania obrazu 2D. Pod względem dokładności metoda skanowania laserowego 3D jest lepsza, niż metoda przetwarzania obrazu 2D. Chropowatość parametrów 3D i odchylenie standardowe entropii w parametrach 2D mają najwyższy współczynnik korelacji z klasą marszczenia, odpowiednio -0,965 i -0,917. Eksperyment weryfikacyjny tkanin dowodzi, że szorstkość może lepiej scharakteryzować stopień marszczenia. Ponadto dzięki zaprezentowanym w pracy wynikom stwierdzono, że różnice stopnia marszczenia między dwiema sąsiadującymi replikami AATCC SA nie byłytakie same; różnica między SA-1 i SA-2 była znacząca, podczas gdy różnica między SA-3 i SA-3.5, a także SA-4 i SA-5 nie byłatak oczywista. Wskazane jest, aby repliki AATCC SA dla klas 3, 3.5, 4 i 5 były dostosowane lub ulepszone.
    Źródło:
    Fibres & Textiles in Eastern Europe; 2020, 2 (140); 43-49
    1230-3666
    2300-7354
    Pojawia się w:
    Fibres & Textiles in Eastern Europe
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Application of Support Vector Machines in automatic human face recognition
    Autorzy:
    Kawulok, M.
    Powiązania:
    https://bibliotekanauki.pl/articles/333790.pdf
    Data publikacji:
    2005
    Wydawca:
    Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
    Tematy:
    automatyczne rozpoznanie twarzy
    metoda wektorów nośnych
    wykrywanie twarzy
    wybór cech
    fuzja wielometodowa
    automatic face recognition
    support vector machines
    face detection
    feature extraction
    multi-method fusion
    Opis:
    This paper presents the possibilities of applying the Support Vector Machines (SVM) in the process of automatic human face recognition. It is described how the existing methods of face recognition can be improved by the SVM. Moreover, a new approach to the multi-method fusion utilising the SVM is proposed. Usefulness of all the methods described in the paper improving the face recognition effectiveness by the SVM is confirmed by the experimental results.
    Źródło:
    Journal of Medical Informatics & Technologies; 2005, 9; 143-150
    1642-6037
    Pojawia się w:
    Journal of Medical Informatics & Technologies
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Customer product review summarization over time for competitive intelligence
    Autorzy:
    Amarouche, Kamal
    Benbrahim, Houda
    Kassou, Ismail
    Powiązania:
    https://bibliotekanauki.pl/articles/950925.pdf
    Data publikacji:
    2018
    Wydawca:
    Sieć Badawcza Łukasiewicz - Przemysłowy Instytut Automatyki i Pomiarów
    Tematy:
    feature extraction
    fuzzy logic
    competitive intelligence
    opinion mining
    opinion summarization
    sentiment analysis
    SentiWordNet
    ekstrakcja cech
    logika rozmyta
    wywiad konkurencyjny
    eksploracja opinii
    podsumowanie opinii
    analiza nastrojów
    Opis:
    Nowadays, Customer’s product reviews can be widely found on the Web, be it in personal blogs, forums, or ecommerce websites. They contain important products’ information and therefore became a new data source for competitive intelligence. On that account, these reviews need to be analyzed and summarized in order to help the leader of an entity (company, brand, etc.) to make appropriate decisions in an efective way. However, most previous review summarization studies focus on summarizing sentiment distribution toward different product features without taking into account that the real advantages and disadvantages of a product clarify over time. For this reason, in this work we aim to propose a new system for product opinion summarization which depends on the time when reviews are expressed and that covers the sentiments change about product features. The proposed system firstly, generates a summary based on product features in order to give more accurate and efficient information about different features. secondly, classify the product based on its features in its appropriate class (good, medium or bad product) using a fuzzy logic system. The experimental results demonstrate the effectiveness of the proposed system to generate the real image of a product and its features in reviews.
    Źródło:
    Journal of Automation Mobile Robotics and Intelligent Systems; 2018, 12, 4; 70-82
    1897-8649
    2080-2145
    Pojawia się w:
    Journal of Automation Mobile Robotics and Intelligent Systems
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Ensemble of feature extraction methods to improve the structural damage classification in a wind turbine foundation
    Autorzy:
    Leon-Medina, Jersson X.
    Parés, Núria
    Anaya, Maribel
    Tibaduiza, Diego A.
    Pozo, Francesc
    Powiązania:
    https://bibliotekanauki.pl/articles/27311417.pdf
    Data publikacji:
    2023
    Wydawca:
    Polska Akademia Nauk. Czasopisma i Monografie PAN
    Tematy:
    structural health monitoring
    wind turbine foundation
    damage classification
    machine learning
    feature extraction
    XGBoost
    monitorowanie stanu konstrukcji
    fundament turbiny wiatrowej
    klasyfikacja uszkodzeń
    uczenie maszynowe
    ekstrakcja cech
    Opis:
    The condition monitoring of offshore wind power plants is an important topic that remains open. This monitoring aims to lower the maintenance cost of these plants. One of the main components of the wind power plant is the wind turbine foundation. This study describes a data-driven structural damage classification methodology applied in a wind turbine foundation. A vibration response was captured in the structure using an accelerometer network. After arranging the obtained data, a feature vector of 58 008 features was obtained. An ensemble approach of feature extraction methods was applied to obtain a new set of features. Principal Component Analysis (PCA) and Laplacian eigenmaps were used as dimensionality reduction methods, each one separately. The union of these new features is used to create a reduced feature matrix. The reduced feature matrix is used as input to train an Extreme Gradient Boosting (XGBoost) machine learning-based classification model. Four different damage scenarios were applied in the structure. Therefore, considering the healthy structure, there were 5 classes in total that were correctly classified. Five-fold cross validation is used to obtain a final classification accuracy. As a result, 100% of classification accuracy was obtained after applying the developed damage classification methodology in a wind-turbine offshore jacket-type foundation benchmark structure.
    Źródło:
    Bulletin of the Polish Academy of Sciences. Technical Sciences; 2023, 71, 3; art. no. e144606
    0239-7528
    Pojawia się w:
    Bulletin of the Polish Academy of Sciences. Technical Sciences
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    A note on Töeplitz matrix-based model in biometrics
    Autorzy:
    Saeed, K.
    Powiązania:
    https://bibliotekanauki.pl/articles/332882.pdf
    Data publikacji:
    2008
    Wydawca:
    Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
    Tematy:
    macierze Teoplitza
    multibiometria
    wybór cech obrazu
    opis obrazu
    bezpieczeństwo systemów biometrycznych
    fałszowania w biometrii
    Toeplitz matrices
    multibiometrics
    feature extraction
    image description
    security in biometric systems
    spoofing in biometrics
    Opis:
    This paper presents a summary of the work presented as an invited paper at MIT 2008 International Conference. The work comprises a general note on the problems we meet in our everyday contact with biometrics and their different systems. A particular attention is paid to the anti-spoofing approaches in having a safe and convenient system of human verification for personal identification. A conclusion is drawn that neither stand-alone nor multi-system Biometrics are ideal and convenient to people for their daily necessity of being identified. The author suggests a system that may seem practical in banks and cash machines, for example, in which a biometric system is used (fingerprint or face identification for example) in conjunction with the popular means of account securing, the PIN code.
    Źródło:
    Journal of Medical Informatics & Technologies; 2008, 12; 29-31
    1642-6037
    Pojawia się w:
    Journal of Medical Informatics & Technologies
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Speech emotion recognition using wavelet packet reconstruction with attention-based deep recurrent neutral networks
    Autorzy:
    Meng, Hao
    Yan, Tianhao
    Wei, Hongwei
    Ji, Xun
    Powiązania:
    https://bibliotekanauki.pl/articles/2173587.pdf
    Data publikacji:
    2021
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    speech emotion recognition
    voice activity detection
    wavelet packet reconstruction
    feature extraction
    LSTM networks
    attention mechanism
    rozpoznawanie emocji mowy
    wykrywanie aktywności głosowej
    rekonstrukcja pakietu falkowego
    wyodrębnianie cech
    mechanizm uwagi
    sieć LSTM
    Opis:
    Speech emotion recognition (SER) is a complicated and challenging task in the human-computer interaction because it is difficult to find the best feature set to discriminate the emotional state entirely. We always used the FFT to handle the raw signal in the process of extracting the low-level description features, such as short-time energy, fundamental frequency, formant, MFCC (mel frequency cepstral coefficient) and so on. However, these features are built on the domain of frequency and ignore the information from temporal domain. In this paper, we propose a novel framework that utilizes multi-layers wavelet sequence set from wavelet packet reconstruction (WPR) and conventional feature set to constitute mixed feature set for achieving the emotional recognition with recurrent neural networks (RNN) based on the attention mechanism. In addition, the silent frames have a disadvantageous effect on SER, so we adopt voice activity detection of autocorrelation function to eliminate the emotional irrelevant frames. We show that the application of proposed algorithm significantly outperforms traditional features set in the prediction of spontaneous emotional states on the IEMOCAP corpus and EMODB database respectively, and we achieve better classification for both speaker-independent and speaker-dependent experiment. It is noteworthy that we acquire 62.52% and 77.57% accuracy results with speaker-independent (SI) performance, 66.90% and 82.26% accuracy results with speaker-dependent (SD) experiment in final.
    Źródło:
    Bulletin of the Polish Academy of Sciences. Technical Sciences; 2021, 69, 1; art. no. e136300
    0239-7528
    Pojawia się w:
    Bulletin of the Polish Academy of Sciences. Technical Sciences
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Speech emotion recognition using wavelet packet reconstruction with attention-based deep recurrent neutral networks
    Autorzy:
    Meng, Hao
    Yan, Tianhao
    Wei, Hongwei
    Ji, Xun
    Powiązania:
    https://bibliotekanauki.pl/articles/2090711.pdf
    Data publikacji:
    2021
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    speech emotion recognition
    voice activity detection
    wavelet packet reconstruction
    feature extraction
    LSTM networks
    attention mechanism
    rozpoznawanie emocji mowy
    wykrywanie aktywności głosowej
    rekonstrukcja pakietu falkowego
    wyodrębnianie cech
    mechanizm uwagi
    sieć LSTM
    Opis:
    Speech emotion recognition (SER) is a complicated and challenging task in the human-computer interaction because it is difficult to find the best feature set to discriminate the emotional state entirely. We always used the FFT to handle the raw signal in the process of extracting the low-level description features, such as short-time energy, fundamental frequency, formant, MFCC (mel frequency cepstral coefficient) and so on. However, these features are built on the domain of frequency and ignore the information from temporal domain. In this paper, we propose a novel framework that utilizes multi-layers wavelet sequence set from wavelet packet reconstruction (WPR) and conventional feature set to constitute mixed feature set for achieving the emotional recognition with recurrent neural networks (RNN) based on the attention mechanism. In addition, the silent frames have a disadvantageous effect on SER, so we adopt voice activity detection of autocorrelation function to eliminate the emotional irrelevant frames. We show that the application of proposed algorithm significantly outperforms traditional features set in the prediction of spontaneous emotional states on the IEMOCAP corpus and EMODB database respectively, and we achieve better classification for both speaker-independent and speaker-dependent experiment. It is noteworthy that we acquire 62.52% and 77.57% accuracy results with speaker-independent (SI) performance, 66.90% and 82.26% accuracy results with speaker-dependent (SD) experiment in final.
    Źródło:
    Bulletin of the Polish Academy of Sciences. Technical Sciences; 2021, 69, 1; e136300, 1--12
    0239-7528
    Pojawia się w:
    Bulletin of the Polish Academy of Sciences. Technical Sciences
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Research on communication emitter identification based on semi-supervised dimensionality reduction in complex electromagnetic environment
    Autorzy:
    Ge, Wei
    Qi, Lin
    Tong, Lin
    Zhu, Jun
    Zhang, Jing
    Zhao, Dongyang
    Li, Ke
    Powiązania:
    https://bibliotekanauki.pl/articles/27311449.pdf
    Data publikacji:
    2023
    Wydawca:
    Polska Akademia Nauk. Czasopisma i Monografie PAN
    Tematy:
    communication emitter identification
    feature extraction
    dimensionality reduction
    VMD
    ESDA
    variational mode decomposition
    exponential semi-supervised discriminant analysis
    identyfikacja emitera komunikacyjnego
    ekstrakcja cech
    redukcja wymiarowości
    rozkład w trybie wariacyjnym
    analiza dyskryminacyjna wykładnicza półnadzorowana
    Opis:
    The individual identification of communication emitters is a process of identifying different emitters based on the radio frequency fingerprint features extracted from the received signals. Due to the inherent non-linearity of the emitter power amplifier, the fingerprints provide distinguishing features for emitter identification. In this study, approximate entropy is introduced into variational mode decomposition, whose features performed in each mode which is decomposed from the reconstructed signal are extracted while the local minimum removal method is used to filter out the noise mode to improve SNR. We proposed a semi-supervised dimensionality reduction method named exponential semi-supervised discriminant analysis in order to reduce the high-dimensional feature vectors of the signals, and LightGBM is applied to build a classifier for communication emitter identification. The experimental results show that the method performs better than the state-of-the-art individual communication emitter identification technology for the steady signal data set of radio stations with the same plant, batch and model.
    Źródło:
    Bulletin of the Polish Academy of Sciences. Technical Sciences; 2023, 71, 4; art. no. e145766
    0239-7528
    Pojawia się w:
    Bulletin of the Polish Academy of Sciences. Technical Sciences
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Characterization of symbolic rules embedded in deep DIMLP networks : a challenge to transparency of deep learning
    Autorzy:
    Bologna, G.
    Hayashi, Y.
    Powiązania:
    https://bibliotekanauki.pl/articles/91545.pdf
    Data publikacji:
    2017
    Wydawca:
    Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych
    Tematy:
    ensemble
    Deep Learning
    rule extraction
    feature detectors
    Opis:
    Rule extraction from neural networks is a fervent research topic. In the last 20 years many authors presented a number of techniques showing how to extract symbolic rules from Multi Layer Perceptrons (MLPs). Nevertheless, very few were related to ensembles of neural networks and even less for networks trained by deep learning. On several datasets we performed rule extraction from ensembles of Discretized Interpretable Multi Layer Perceptrons (DIMLP), and DIMLPs trained by deep learning. The results obtained on the Thyroid dataset and the Wisconsin Breast Cancer dataset show that the predictive accuracy of the extracted rules compare very favorably with respect to state of the art results. Finally, in the last classification problem on digit recognition, generated rules from the MNIST dataset can be viewed as discriminatory features in particular digit areas. Qualitatively, with respect to rule complexity in terms of number of generated rules and number of antecedents per rule, deep DIMLPs and DIMLPs trained by arcing give similar results on a binary classification problem involving digits 5 and 8. On the whole MNIST problem we showed that it is possible to determine the feature detectors created by neural networks and also that the complexity of the extracted rulesets can be well balanced between accuracy and interpretability.
    Źródło:
    Journal of Artificial Intelligence and Soft Computing Research; 2017, 7, 4; 265-286
    2083-2567
    2449-6499
    Pojawia się w:
    Journal of Artificial Intelligence and Soft Computing Research
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
      Wyświetlanie 1-69 z 69

      Ta witryna wykorzystuje pliki cookies do przechowywania informacji na Twoim komputerze. Pliki cookies stosujemy w celu świadczenia usług na najwyższym poziomie, w tym w sposób dostosowany do indywidualnych potrzeb. Korzystanie z witryny bez zmiany ustawień dotyczących cookies oznacza, że będą one zamieszczane w Twoim komputerze. W każdym momencie możesz dokonać zmiany ustawień dotyczących cookies