Design And Implementation Binocular Vision Technology In Blind Assisting System

The thesis considers the fact that the blinds that lost their vision could hardly get enough information to understand the situation of surrounding,using binocular stereo vision for two tasks of blind assisting system’s,which are objective orientation and real-time object tracking. The paper,using modem digital image processing methods,stereo vision methods,moving target detection and tracking algorithms,launch in-depth research.The study of two-camera visual platforms includes image grabbing,cameras calibration,image processing, image rectification, stereo matching,three-dimensional reconstruction and movement analysis.Firstly,at the conclusion of sever camera calibration methods,we make some amelioration to realize the binocular stereo camera system calibration.The image coordinate of the Calibration board’s corners are obtained by ways of hand working method and the Harris comer detecting,and then the camera parameters can be obtained by the calculation and the experiment.According to the data,image preprocessing and image rectification are taken, in order to improve stereo matching and3D rebuilding efficiency.Secondly,stereo matching algorithm is deeply studied.According to the advantages and disadvantages of various algorithms,and considering the real time of the system,1use SAD regional intensity-based correlation method to get disparity.The3D coordinate can be obtained according to the basic principle of the stereo vision and the coordinate calculation methods, and then compare the feature spatial information result with the result from the actual measured distance,the error will be obtained, then the error analysis is taken.Finally,the presently common methods on the moving target detection and tracking of moving objects are importantly introduced.In the processing of moving object tracking,the paper based on Camshift algorithm to realize the self-motion tracking of colorful sequential imaging.What’s more, to remedy these drawbacks of the conventional Camshift algorithm,a modified Camshift which Combine the SURF and Camshift algorithm is proposed in this paper.Experiments show that the algorithm has avoided the mistake in the process of tracking and reduced the operation consumedly,it can track the object accurately in a real-time way.The results of experiment show that this system complete two tasks,and it is the foundation of further function of the blind assisting system.