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 "model recognition" wg kryterium: Wszystkie pola

  Lista filtrów
Wyrównaj
Wyświetlanie 1-15 z 47
Tytuł:
Quantum Foundations of Resonant Recognition Model
Autorzy:
Keković, G.
Raković, D.
Tošić, B.
Davidović, D.
Cosić, I.
Powiązania:
https://bibliotekanauki.pl/articles/1537869.pdf
Data publikacji:
2010-05
Wydawca:
Polska Akademia Nauk. Instytut Fizyki PAN
Tematy:
87.14.E-
87.15.-v
87.15.B-
87.15.Cc
87.15.hp
87.15.kp
Opis:
Biomolecular recognition is an open scientific problem, which has been investigated in many theoretical and experimental aspects. In that sense, there are encouraging results within Resonant Recognition Model (RRM), based on the finding that there is a significant correlation between spectra of the numerical presentation of amino acids in the primary structure of proteins and their biological activity. It has been found through an extensive research that proteins with the same biological function have a common frequency in their numerical spectra. This frequency was found then to be a characteristic feature for protein biological function or interaction The RRM model proposes that the selectivity of protein interactions is based on resonant energy transfer between interacting biomolecules and that this energy, electromagnetic in its nature, is in the frequency range of $10^{13}$ to $10^{15}$ Hz, which incorporates infra-red (IR), visible and a small portion of the ultra-violet (UV) radiation. In this paper, the quantum mechanical basis of the RRM model will be investigated using the solution in the simplified framework of Hückel-like theory of molecular orbits.
Źródło:
Acta Physica Polonica A; 2010, 117, 5; 756-759
0587-4246
1898-794X
Pojawia się w:
Acta Physica Polonica A
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Selected aspects of digital image processing applications in ITS
Autorzy:
Mrówka, P.
Olejniczak, M.
Powiązania:
https://bibliotekanauki.pl/articles/393804.pdf
Data publikacji:
2013
Wydawca:
Polskie Stowarzyszenie Telematyki Transportu
Tematy:
digital image processing
image parametrization
traffic light classification
colour recognition
model recognition
cyfrowe przetwarzanie obrazu
parametryzacja obrazu
klasyfikacja sygnalizacji świetlnej
rozpoznawanie koloru
rozpoznawanie modelu
Opis:
Digital image processing algorithms are commonly applied in Intelligent Transport Systems (ITS). Their effective operation is conditioned on the high robustness to real-life image distortions and the computational complexity suitable for implementation on a non expensive industrial computer. The paper presents three original image analysis methods designed for the ITS, with special attention paid on aforementioned conditions. Colour image parametrization method for the traffic light state classifier was described. The algorithm utilizes CIELAB colour space properties. The method of vehicle edges parametrization for the make and model classifier was presented. The proposed representation relies on thresholded coefficients of gradient magnitude approximation in low dimensional space. The paper presents also the method of image characteristic features detection for the licence plates localization task. The detection is performed by means of appropriately designed filters with low computational complexity.
Źródło:
Archives of Transport System Telematics; 2013, 6, 3; 23-26
1899-8208
Pojawia się w:
Archives of Transport System Telematics
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A robust ensemble model for spoken language recognition
Autorzy:
Woods, Nancy
Babatunde, Gideon
Powiązania:
https://bibliotekanauki.pl/articles/118275.pdf
Data publikacji:
2020
Wydawca:
Polskie Towarzystwo Promocji Wiedzy
Tematy:
spoken language recognition
computer vision
image recognition
CNN
rozpoznawanie języka mówionego
widzenie komputerowe
rozpoznawanie obrazu
Opis:
The identity of a language being spoken has been tackled over the years via statistical models on audio samples. A drawback of these approaches is the unavailability of phonetically transcribed data for all languages. This work proposes an approach based on image classification that utilized image representations of audio samples. Our model used Neural Networks and deep learning algorithms to analyse and classify three languages. The input to our network is a Spectrogram that was processed through the networks to extract local visual and temporal features for language prediction. From the model, we achieved 95.56 % accuracy on the test samples from the 3 languages.
Źródło:
Applied Computer Science; 2020, 16, 3; 56-68
1895-3735
Pojawia się w:
Applied Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Pipelined language model construction for Polish speech recognition
Autorzy:
Sas, J.
Żołnierek, A.
Powiązania:
https://bibliotekanauki.pl/articles/329841.pdf
Data publikacji:
2013
Wydawca:
Uniwersytet Zielonogórski. Oficyna Wydawnicza
Tematy:
automatic speech recognition
hidden Markov model
adaptive language model
automatyczne rozpoznawanie mowy
model Markova ukryty
model językowy adaptacyjny
Opis:
The aim of works described in this article is to elaborate and experimentally evaluate a consistent method of Language Model (LM) construction for the sake of Polish speech recognition. In the proposed method we tried to take into account the features and specific problems experienced in practical applications of speech recognition in the Polish language, reach inflection, a loose word order and the tendency for short word deletion. The LM is created in five stages. Each successive stage takes the model prepared at the previous stage and modifies or extends it so as to improve its properties. At the first stage, typical methods of LM smoothing are used to create the initial model. Four most frequently used methods of LM construction are here. At the second stage the model is extended in order to take into account words indirectly co-occurring in the corpus. At the next stage, LM modifications are aimed at reduction of short word deletion errors, which occur frequently in Polish speech recognition. The fourth stage extends the model by insertion of words that were not observed in the corpus. Finally the model is modified so as to assure highly accurate recognition of very important utterances. The performance of the methods applied is tested in four language domains.
Źródło:
International Journal of Applied Mathematics and Computer Science; 2013, 23, 3; 649-668
1641-876X
2083-8492
Pojawia się w:
International Journal of Applied Mathematics and Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Rozpoznawanie tekstur z wykorzystaniem bazy modeli
Textures recognition using model database
Autorzy:
Szymczyk, T.
Powiązania:
https://bibliotekanauki.pl/articles/159600.pdf
Data publikacji:
2011
Wydawca:
Sieć Badawcza Łukasiewicz - Instytut Elektrotechniki
Tematy:
rozpoznawanie tekstur
przetwarzanie obrazów
dopasowanie wzorców
Opis:
Problem rozpoznawania obrazów jest zagadnieniem trudnym i złożonym. Jednym z możliwych sposobów jego realizacji, jest wykorzystanie bazy modeli i porównywanie ich z obrazem badanym pochodzącym z kamery. Możliwe jest rozpoznawanie w oparciu o: kształt, kolor czy też teksturę. W artykule zaprezentowano wykorzystanie zmodyfikowanej metody dopasowania wzorców do rozpoznawania obrazów, pochodzących z rzeczywistej kamery przemysłowej. Przedstawiono także podstawowe wady i ograniczenia klasycznej metody dopasowania wzorców.
The problem of the image recognizing is difficult and complex question. The using the base of models and comparing their to image from camera is one of possible ways of his realization. Image recognizing is possible in support about shape, colour or texture also. In article modified method template matching to recognizing images from real industrial camera was presented. Moreover basic disadvantages and limitations of simple template matching method was shown in the paper.
Źródło:
Prace Instytutu Elektrotechniki; 2011, 249; 95-115
0032-6216
Pojawia się w:
Prace Instytutu Elektrotechniki
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
CAD models clustering with machine learning
Autorzy:
Machalica, Dawid
Matyjewski, Marek
Powiązania:
https://bibliotekanauki.pl/articles/139503.pdf
Data publikacji:
2019
Wydawca:
Polska Akademia Nauk. Czytelnia Czasopism PAN
Tematy:
3D shape matching
3D shape retrieval
3D model recognition
3D shape
content-based retrieval
machine learning
dopasowanie kształtu 3D
pobieranie kształtu 3D
rozpoznawanie modeli 3D
kształt 3D
pobieranie oparte na treści
uczenie maszynowe
Opis:
Similarity assessment between 3D models is an important problem in many fields including medicine, biology and industry. As there is no direct method to compare 3D geometries, different model representations (shape signatures) are developed to enable shape description, indexing and clustering. Even though some of those descriptors proved to achieve high classification precision, their application is often limited. In this work, a different approach to similarity assessment of 3D CAD models was presented. Instead of focusing on one specific shape signature, 45 easy-to-extract shape signatures were considered simultaneously. The vector of those features constituted an input for 3 machine learning algorithms: the random forest classifier, the support vector classifier and the fully connected neural network. The usefulness of the proposed approach was evaluated with a dataset consisting of over 1600 CAD models belonging to 9 separate classes. Different values of hyperparameters, as well as neural network configurations, were considered. Retrieval accuracy exceeding 99% was achieved on the test dataset.
Źródło:
Archive of Mechanical Engineering; 2019, LXVI, 2; 133-152
0004-0738
Pojawia się w:
Archive of Mechanical Engineering
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Attention-based deep learning model for Arabic handwritten text recognition
Autorzy:
Aïcha Gader, Takwa Ben
Echi, Afef Kacem
Powiązania:
https://bibliotekanauki.pl/articles/2201264.pdf
Data publikacji:
2022
Wydawca:
Szkoła Główna Gospodarstwa Wiejskiego w Warszawie. Instytut Informatyki Technicznej
Tematy:
Arabic handwriting recognition
attention mechanism
BLSTM
CNN
CTC
RNN
Opis:
This work proposes a segmentation-free approach to Arabic Handwritten Text Recog-nition (AHTR): an attention-based Convolutional Neural Network - Recurrent Neural Network - Con-nectionist Temporal Classification (CNN-RNN-CTC) deep learning architecture. The model receives asinput an image and provides, through a CNN, a sequence of essential features, which are transferred toan Attention-based Bidirectional Long Short-Term Memory Network (BLSTM). The BLSTM gives features sequence in order, and the attention mechanism allows the selection of relevant information from the features sequences. The selected information is then fed to the CTC, enabling the loss calculation and the transcription prediction. The contribution lies in extending the CNN by dropout layers, batch normalization, and dropout regularization parameters to prevent over-fitting. The output of the RNN block is passed through an attention mechanism to utilize the most relevant parts of the input sequence in a flexible manner. This solution enhances previous methods by improving the CNN speed and performance and controlling over model over-fitting. The proposed system achieves the best accuracy of97.1% for the IFN-ENIT Arabic script database, which competes with the current state-of-the-art. It was also tested for the modern English handwriting of the IAM database, and the Character Error Rate of 2.9% is attained, which confirms the model’s script independence.
Źródło:
Machine Graphics & Vision; 2022, 31, 1/4; 49--73
1230-0535
2720-250X
Pojawia się w:
Machine Graphics & Vision
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Opracowanie koncepcji i implementacja modelu rozpoznawania obrazu z wykorzystaniem elementów sztucznej inteligencji
Development of the concept and implementation of an image recognition model using elements of artificial intelligence
Autorzy:
Sierżantowicz, Anna
Ptasznik, Andrzej
Powiązania:
https://bibliotekanauki.pl/articles/1397486.pdf
Data publikacji:
2020
Wydawca:
Warszawska Wyższa Szkoła Informatyki
Tematy:
rozpoznawanie obrazu
sztuczne sieci neuronowe
sieci splotowe
InceptionV3
transfer wiedzy
computer vision
artificial neural network
connvolutional neural network
transfer learning
Opis:
W niniejszym artykule przedstawiono koncepcję i implementację modelu do rozpoznawania ras psów na podstawie zdjęcia. Do realizacji zadania wykorzystano model głębokiej sieci neuronowej bazujący na strukturze InceptionV3. Sieć została wytrenowana i przetestowana na zbiorze przypadków uczących liczącym ponad 20 tys. zdjęć 120 ras psów z zastosowaniem transferu wiedzy. Zbadano również wpływ jakości zdjęć na wyniki klasyfikacji. Sieć uzyskała bardzo dobre rezultaty zarówno w przypadku analizy typowych, jak i nietypowych zdjęć.
This article presents the concept and implementation of a model for recognizing dog breeds based on an input image. The task was performed with the use of a deep neural network model based on the InceptionV3 structure. The neural network has been trained and tested on a dataset counting more than 20,000 images of 120 dog breeds using transfer learning technique. The impact of image quality on classification results was also examined. The model obtained very good results in the analysis of both typical and unusual input images.
Źródło:
Zeszyty Naukowe Warszawskiej Wyższej Szkoły Informatyki; 2020, 14, 23; 7-26
1896-396X
2082-8349
Pojawia się w:
Zeszyty Naukowe Warszawskiej Wyższej Szkoły Informatyki
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Determination of Input Parameters of the Neural Network Model, Intended for Phoneme Recognition of a Voice Signal in the Systems of Distance Learning
Autorzy:
Akhmetov, B.
Tereykovsky, I.
Doszhanova, A.
Tereykovskaya, L.
Powiązania:
https://bibliotekanauki.pl/articles/226378.pdf
Data publikacji:
2018
Wydawca:
Polska Akademia Nauk. Czytelnia Czasopism PAN
Tematy:
neural networks
phonemes
recognition of a voice signal
system of distance learning
mel-cepstral coefficients
spectral analysis
Opis:
The article is devoted to the problem of voice signals recognition means introduction in the system of distance learning. The results of the conducted research determine the prospects of neural network means of phoneme recognition. It is also shown that the main difficulties of creation of the neural network model, intended for recognition of phonemes in the system of distance learning, are connected with the uncertain duration of a phoneme-like element. Due to this reason for recognition of phonemes, it is impossible to use the most effective type of neural network model on the basis of a multilayered perceptron, at which the number of input parameters is a fixed value. To mitigate this shortcoming, the procedure, allowing to transform the non-stationary digitized voice signal to the fixed quantity of mel-cepstral coefficients, which are the basis for calculation of input parameters of the neural network model, is developed. In contrast to the known ones, the possibility of linear scaling of phoneme-like elements is available in the procedure. The number of computer experiments confirmed expediency of the fact that the use of the offered coding procedure of input parameters provides the acceptable accuracy of neural network recognition of phonemes under near-natural conditions of the distance learning system. Moreover, the prospects of further research in the field of development of neural network means of phoneme recognition of a voice signal in the system of distance learning is connected with an increase in admissible noise level. Besides, the adaptation of the offered procedure to various natural languages, as well as to other applied tasks, for instance, a problem of biometric authentication in the banking sector, is also of great interest.
Źródło:
International Journal of Electronics and Telecommunications; 2018, 64, 4; 425-432
2300-1933
Pojawia się w:
International Journal of Electronics and Telecommunications
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Wpływ transmisji głosu z wykorzystaniem telefonii internetowej VoIP na skuteczność automatycznego systemu kryminalistycznej identyfikacji mówców opartego na metodzie EM-UBM-MAP
The impact of voice transmission using VoIP Internet Telephony on the efficiency of forensic of an automatic speaker recognition system based on EM-UBM-MAP*
Autorzy:
Maciejko, Waldemar
Powiązania:
https://bibliotekanauki.pl/articles/499794.pdf
Data publikacji:
2014
Wydawca:
Centralne Laboratorium Kryminalistyczne Policji
Tematy:
Voice over Internet Protocol
kryminalistyczna identyfikacja mówców
model Gilberta
utrata pakietów
kompresja dźwięku
standard H.323
the forensic speaker recognition Gilbert model, packet loss
audio compression
the H.323 standard
Opis:
W niniejszej pracy zbadano wpływ transmisji pakietowej na skuteczność systemu automatycznej identyfikacji mówców pracującego w oparciu o bayesowski klasyfikator LR. Modelowanie statystyczne w systemie prowadzone jest z wykorzystaniem algorytmów EM-GMM oraz MAP. W badaniach założono, że z transmisją pakietową wiążą się dwa zjawiska: utrata pakietów oraz kodowanie sygnału. W badaniach wykorzystano niektóre kodeki audio standardu H.323 ITU-T. Zjawisko utraty pakietów przybliżono za pomocą dwustanowego modelu Gilberta. Wyniki badań przedstawiono w postaci charakterystyk Tippetta.
In this study, the effect of packet transmission on the effectiveness of the automatic speaker recognition system was examined, working on the basis Bayesian classifier LR. Statistical modelling of the system is carried out using algorithms EM-GMM and MAP. The study assumed that the packet transmission is associated with the occurrence of loss and encoding of the signal. In the research, some codecs of the audio standard H.323 ITU-T were used. The occurrence of the packet loss was described by means of a Gilbert digital model. Test results of performed tests are presented in Tippett plots.
Źródło:
Problemy Kryminalistyki; 2014, 283
0552-2153
Pojawia się w:
Problemy Kryminalistyki
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Refleksje nad modelem rozpoznawania przez sądy spraw cywilnych z zakresu prawa rolnego
Some reflections on the model for the recognition by courts of civil law cases in agricultural law
Riflessioni sul modelli di riconoscimento delle cause civili in materia di diritto agrario da parte dei tribunali
Autorzy:
Mucha, Joanna
Powiązania:
https://bibliotekanauki.pl/articles/2137429.pdf
Data publikacji:
2022-06-08
Wydawca:
Uniwersytet im. Adama Mickiewicza w Poznaniu
Tematy:
cause in materia di diritto agrario
procedimenti giudiziari in matria di diritto agrario
procedimenti separati
specializzazione dei giudici
agricultural law cases
judicial proceedings in agricultural law cases
separate proceedings
specialisation of judges
sprawy z zakresu prawa rolnego
postępowanie sądowe w sprawach z zakresu prawa rolnego
postępowania odrębne
specjalizacja sędziów
Opis:
Celem rozważań – wobec braku w kodeksie postępowania cywilnego zunifikowanej regulacji poświęconej rozpoznawaniu przez sądy spraw z zakresu prawa rolnego – jest ustalenie, czy w świetle obowiązujących przepisów, a także utrwalonych poglądów doktryny istnieją przesłanki do wyodrębnienia procesowego modelu załatwiania tego rodzaju spraw. W artykule przeanalizowano dwie niewykluczające się wzajemnie opcje: dopuszczalność utworzenia postępowania odrębnego w sprawach z zakresu prawa rolnego oraz poddanie ich rozpoznawania sądownictwu wyspecjalizowanemu. W konkluzji rozważań przyjęto, że ze względu na trudność w realizacji kluczowego dla kwalifikacji spraw do postępowań odrębnych kryterium przedmiotowego, odnoszącego się do specyfiki spraw, a nie właściwości podmiotów, pierwszą koncepcję należy odrzucić. Takie rozwiązanie nie ma oparcia także w argumentach historycznych. Bardziej realne wydaje się rozwiązanie oparte na utworzeniu wyspecjalizowanych wydziałów lub składów orzekających w ramach funkcjonującej struktury sądownictwa.
L’articolo si propone – visto che nel codice di procedura civile manca una disciplina unificata dedicata al riconoscimento delle cause in materia di diritto agrario da parte dei tribunali – di stabilire se, alla luce delle disposizioni in vigore e delle consolidate opinioni dottrinali, esistano premesse che permettano di individuare un modello procedurale in grado di far affrontare questo tipo di cause. L’articolo analizza due opzioni non mutuamente esclusive: da una parte l’ammettere di avviare procedimenti separati in materia di diritto agrario, dall’altra il sottoporre il loro riconoscimento a tribunali specializzati. In conclusione, si è ritenuto che, data la difficoltà di attuare un criterio oggettivo, relativo alla specificità delle cause, e non alle proprietà dei soggetti, ma cruciale nel qualificare le cause per i procedimenti separati. il primo concetto vada scartato. Tale soluzione non è neanche supportata da argomentazioni storiche. Più realistica sembra la soluzione che presuppone di creare sezioni o collegi giudicanti specializzati da inserire all’interno di una struttura già funzionante della magistratura.
The aim of the considerations – in the absence in the Code of Civil Procedure of a unified regulation devoted to the recognition by courts of cases in agricultural law – is to determine whether, in the light of the current legislation as well as the established views of the doctrine, there are premises for the separation of the procedural model of handling this type of cases. Two not mutually exclusive options are analysed: the admissibility of instituting separate proceedings in agricultural law cases, and recognition of such cases by specialised jurisdiction. In conclusion it is assumed that due to the difficulty of implementing the essential criterion for the qualification of cases to separate proceedings pertaining to their specificity rather than the proper jurisdictions of the entities involved, the first concept should be rejected. Moreover, this concept cannot be supported by any historical reasons. A more realistic solution seems to be based on the creation of specialised divisions or panels of judges adjudicating within the functioning structure of the judiciary.
Źródło:
Przegląd Prawa Rolnego; 2022, 1(30); 157-186
1897-7626
2719-7026
Pojawia się w:
Przegląd Prawa Rolnego
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Statistical proper name recognition in Polish economic texts
Autorzy:
Marcińczuk, M.
Piasecki, M.
Powiązania:
https://bibliotekanauki.pl/articles/206385.pdf
Data publikacji:
2011
Wydawca:
Polska Akademia Nauk. Instytut Badań Systemowych PAN
Tematy:
proper name recognition
named entity recognition
machine learning
hidden Markov model
rule-base approach
dictionary-base approach
Opis:
In the paper we present a Proper Name Recognition algorithm based on the Hidden Markov Model (HMM). Recognition of the Proper Names (PN) is treated as the basis for Named Entity Recognition problem in general. The proposed method is based on combining domain-dependent method based on HMM with domain independent methods based on gazetteers and hand-written rules for recognition and post-processing that capture the general properties of Polish PN structure. A large gazetteer with entries described morphologically was acquired from the web. The HMM re-scoring mechanism was applied as a basis for integration of different knowledge sources in PN recognition. Results of experiments on a domain corpus of Polish stock exchange reports, used for training and testing, are presented. A cross-domain evaluation on two other corpora is also presented. Adaptability of the method was analysed by applying the trained model to two other domain corpora.
Źródło:
Control and Cybernetics; 2011, 40, 2; 393-418
0324-8569
Pojawia się w:
Control and Cybernetics
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Optimal acoustic model complexity selection in polish medical speech recognition
Autorzy:
Sas, J.
Poreba, T.
Powiązania:
https://bibliotekanauki.pl/articles/333361.pdf
Data publikacji:
2011
Wydawca:
Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
Tematy:
rozpoznawanie mowy
modele języka
medyczne systemy informacji
speech recognition
language models
medical information systems
Opis:
In the paper, the method of acoustic model complexity level selection for automatic speech recognition is proposed. Selection of the appropriate model complexity affects significantly the accuracy of speech recognition. For this reason the selection of the appropriate complexity level is crucial for practical speech recognition applications, where end user effort related to the implementation of speech recognition system is important. We investigated the correlation between speech recognition accuracy and two popular information criteria used in statistical model evaluation: Bayesian Information Criterion and Akaike Information Criterion computed for applied acoustic models. Experiments carried out for language models related to general medicine texts and radiology diagnostic reporting in CT and MR showed strong correlation of speech recognition accuracy and BIC criterion. Using this dependency, the procedure of Gaussian mixture count selection for acoustic model was proposed. Application of this procedure makes it possible to create the acoustic model maximizing the speech recognition accuracy without additional computational costs related to alternative cross-validation approach and without reduction of training set size, which is unavoidable in the case of cross-validation approach.
Źródło:
Journal of Medical Informatics & Technologies; 2011, 17; 115-122
1642-6037
Pojawia się w:
Journal of Medical Informatics & Technologies
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Application of local bidirectional language model to error correction in polish medical speech recognition
Autorzy:
Sas, J.
Powiązania:
https://bibliotekanauki.pl/articles/333597.pdf
Data publikacji:
2010
Wydawca:
Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
Tematy:
rozpoznawanie mowy
modele języka
medyczne systemy informacji
speech recognition
language models
medical information systems
Opis:
In the paper, the method of short word deletion errors correction in automatic speech recognition is described. Short word deletion errors appear to be a frequent error type in Polish speech recognition. The proposed speech recognition process consists of two stages. At the first stage the utterance is recognized by a typical speech recognizer based on forward bigram language model. At the second stage the word sequence recognized by the first stage recognizer is analyzed and such pairs of adjacent words in the recognized sequence are localized, which are likely to be separated by a short word like conjunction or preposition. The probability of short word appearance in context of found words is evaluated using centered trigrams and backward bigram language model for short words prone to deletion. The set of probabilistic language properties used to correct deletions is called here Local Bidirectional Language Model (in contrast to purely forward or backward model used typically in speech recognition). The decision of short word insertion is based on comparison of deletion error probability of the first stage recognizer and the error probability of the decision based only on centered trigrams and backward model. Despite its simplicity, the method proved to be effective in correcting deletion errors of most frequently appearing Polish prepositions. The method was tested in application to medical spoken reports recognition, where the overall short word deletion error rate was reduced by almost 45%.
Źródło:
Journal of Medical Informatics & Technologies; 2010, 15; 127-134
1642-6037
Pojawia się w:
Journal of Medical Informatics & Technologies
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Multimodal face recognition method with two-dimensional hidden Markov model
Autorzy:
Bobulski, J.
Powiązania:
https://bibliotekanauki.pl/articles/201711.pdf
Data publikacji:
2017
Wydawca:
Polska Akademia Nauk. Czytelnia Czasopism PAN
Tematy:
pattern recognition
biometrics
3D face recognition
hidden Markov model
rozpoznawanie wzorców
biometria
rozpoznawanie twarzy 3D
ukryty model Markowa
Opis:
The paper presents a new solution for the face recognition based on two-dimensional hidden Markov models. The traditional HMM uses one-dimensional data vectors, which is a drawback in the case of 2D and 3D image processing, because part of the information is lost during the conversion to one-dimensional features vector. The paper presents a concept of the full ergodic 2DHMM, which can be used in 2D and 3D face recognition. The experimental results demonstrate that the system based on two dimensional hidden Markov models is able to achieve a good recognition rate for 2D, 3D and multimodal (2D+3D) face images recognition, and is faster than ICP method.
Źródło:
Bulletin of the Polish Academy of Sciences. Technical Sciences; 2017, 65, 1; 121-128
0239-7528
Pojawia się w:
Bulletin of the Polish Academy of Sciences. Technical Sciences
Dostawca treści:
Biblioteka Nauki
Artykuł
Wyświetlanie 1-15 z 47

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