A Self-calibration Method Of Camera Based On Binocular Stereo Vision

Computer binocular stereo vision is an important branch. It mostly according as simulates the mode of processing the scape in human's vision. Make uses of two vidicon that their positions are relatively fixed to record the images of a scenery from slightly different visual angle, and calculating corresponding points in both images can obtain three-dimensioned information of the scenery.Three dimensional measurement based on binocular stereo vision is a sort of passive non-contact measurement way. This measure can chalk up the depth information of the object besides obtainning the width and height information,ect.At present, It has been abroadly used in three-dimensional reconstruction, brainpower mobile robots navigation, industrial supervis, etc. With the Three-dimensional measurement develop and based on binocular stereo vision has been researched widely as well as popular domestic.On the basis of computer binocular stereo vision theory,this paper deeply investigated three dimensional measurement based on binocular stereo vision, and the point of this paper is a self-calibration method of camera based on binocular stereo vision. The application and measure of binocular stereo vision system in the short baseline computer was also studied in this paper.The chiefly work of this paper was to realize a new self-calibration method of camera intrinsic parameters. The self-calibration of camera intrinsic parameters was completed by two cameras detecting at two different positions and extracting the description operators of these feature points. This method has the advantage of simply operation, low requirement for hardware equipments, easy to realize, and so on. It is suitable for calibrating the camera in the measurement environment.