Aiming at the problem that the matching accuracy for the low-texture areas in the stereo matching is so low that the matching failure is likely to occur. This paper proposes a stereo matching algorithm for low-texture areas. Firstly, using the horizontal and vertical gradient of the pixel and the specificity of the pixel itself to perform the calculation of the matching cost. At the same time, introducing the HSV color space to obtain the adaptive support arm length of each pixel in the image, thereby obtaining the aggregate area of each pixel. Then, in the acquired aggregation area, by using the bilateral filtering method to calculate the aggregation cost. Finally, in the disparity computation and refinement stage, Left-Right Consistency (LRC) check is combined with the iterative guided filtering method to reduce the error matching. At the end of the article, by using the standard images on the Middlebury platform to experiment. The experimental results are compared with the traditional experimental results, which proves the effectiveness of the proposed algorithm in the low-texture areas.
In order to obtain the full view image, proposing a fast method of generating full view images based on camera. the method based on multi-camera equipment and the spherical transform of the camera image acquisition using sphere transformation model. the improved SIFT algorithm is used to extract keypoints and match keypoints, and then calculating the transformation matrix based on the keypoints that filtered. finally, according to the projection matrix and transform matrix of each image, the images of subsequent cameras are spliced together. Using this method to generate full view images. the speed is fast and real-time.