Based Computer Binocular Stereo Vision Technology, Human Rehabilitation System

With researches on binocular stereovision and development of domestic rehabilitation systems and overseas ones,. this paper present application of computerized binocular stereovision to rehabilitation system and research on auto recognition of physical rehabilitation procession based on computerized binocular stereovision.The goal of rehabilitative treatments is to enable a person who has experienced a stroke to regain the highest possible level of independence and be as productive as possible. Mostly, rehabilitative therapy is a dynamic procession in which motions out of control are corrected with a series of sensors.Computerized binocular stereovision is one of the most important parts of the robot research. With more advanced researches, computerized binocular stereovision reveals its broad prospect in biomedicine, manufacture and information et al. This paper deals with the computerized binocular visional technology that gain the three-dimensional scenes from two two-dimensional pictures. The main purpose is to reconstruct the three-dimensional pixels with two photos of the same limb under rehabilitative trains.As one of base vision technology theories, stereo matching can bring up with depth information which monocular vision does not can get and construct basic models for multi vision as well. So, stereo matching is an important subject of computerized vision. This paper concludes the stereo matching methods from some aspects such as primitive factor selection, matching rules and algorithm organization. This paper presents a matching measure algorithm that integrate feature points manually, feature points matching based on gray relationship and lateral matching. The main point of this algorithm is to count out three-dimensional pixels of dimensional points according to parameters gained by matching.Experience suggests that the theory of this algorithm is very correct and satisfy the field commands of scale detection and location detection. This algorithm bring up with good base on which the computerized binocular system is combined withrehabilitative treatment system.