Research On Binocular Stereo Matching Algorithm And Application In 3D Positioning Of Household Ceramics

The quality of the stereo matching algorithm in the binocular stereo vision technology directly affects the positioning accuracy of the target in the three-dimensional space.In order to achieve better 3D positioning accuracy,this article studies the local stereo matching algorithm and the semi-global stereo matching algorithm in the stereo matching algorithm.The main research work is as follows:(1)Aiming at the problem of poor matching accuracy of depth discontinuous regions on the image,a local stereo matching algorithm based on SLIC superpixel segmentation is proposed.The initial plane parameter map is obtained by combining,and then the idea of iterative propagation and random search in the Patch Match algorithm is used to further improve the matching accuracy of parallax.The proposed algorithm has an error of about0.7 under the Middleburry test graph.(2)Research and improve the SGBM semi-global stereo matching algorithm,and propose an SGBM semi-global stereo matching algorithm combined with the Census algorithm.The average error of the improved SGBM algorithm tested under Middleburry is 0.9.(3)The stereo matching algorithm is applied to the three-dimensional positioning of ceramics,and a hardware platform for binocular stereo vision is built for this purpose,and the binocular camera is calibrated by Zhang's calibration method.According to the calibration results,the stereo correction of daily-use ceramics is realized.The BM stereo matching algorithm,the local stereo matching algorithm proposed in this article and the improved SGBM stereo matching algorithm are used to calculate the three-dimensional positioning error of ceramics.The results show that the stereo matching algorithm proposed in this article can effectively realize the three-dimensional localization of ceramic feature points.