Binocular Stereoscopic Vision Algorithm Based on Improved SIFT Feature

摘要

Stereo matching is the most important step in binocular vision, the traditional regional stereo matching to obtain the target three-dimensional information is slow and inaccurate. This paper presents an improved SIFT algorithm. Firstly, making the epipolar constraint on the left and right image; secondly, selecting the ROI of target from the left of binocular images, and reducing running time by reducing the dimension of feature vectors and accelerating the matching speed by using BBF algorithm based on KD tree; finally, removing the false matching by using RANSAC algorithm. The improved SIFT algorithm can get the target's feature points quickly and accurately, so the 3D coordinates can be calculated by the triangulation method speedy.