Image dynamic matching for stereo vision based on structural information between lines
Qiang Wen · Kongzhi yu juece · 2003
The stereo correspondence problem is the problem of matching two images of the same scene from different viewing positions. A dynamic programming approach is used to solve the correspondence problem in stereo vision. The similarity measure between the straight lines in left and right image is divided into local similarity measure (LSM) and global similarity measure (GSM). The structural information is added to GSM. Some rules based on the structural network of each image are used to eliminate the incorrect matching between edge lines of left image and right image. The simulation result shows that the image matching achieves a high success rate and needs less time when improved dynamic programming is used for stereo matching.