Real-time Video Vectorization Method for the Purpose of Surroundings Recognition and Mobile Robot Navigation

In thise paper a very simple method of the visual information interpretation for recognition and navigation purposes is presented and discussed. The proposed method consists of six steps: image acquisition, edge detection, fast edge vectorization using a high number of short preliminary vectors, aggregation of the preliminary vectors into the form of final vectors. The next stages of the visualthe visual information interpretation for recognition and navigation purposes will be description of the objects’ shapes by means of the final vectors, object recognition and/or robot navigation on the base of comparison between actual shape description and templates memorized during the programming/training process, but they are not discussed in this paper. The main advantage of the proposed method is a simple and time-effective algorithm, which can be performed in real time also by a simple and cheap processor, working as a “brain” of the considered robot.