BIO-INSPIRED IMPLEMENTATION OF ARTIFICIAL VISION MOTION DETECTION SYSTEMS

Motion estimation is one of the most studied areas in artificial vision. Several algorithms with biologic inspiration have already been created for this goal. The main reason is that motion can be used in many applications, from both video compression and codification to perceptual tasks like surveillance and activity identification. Human brain can solve motion detection reliably and in a simple way, making its behaviour attractive this to be applied in robotics. In this paper the main functions of three areas involved in the motion perception process (through visual path) are described. Those functions have been used as bioinspiration to create computational models. This study was motivated by the motion information relevance in gesture recognition and imitation for robot programming.