Artificial Intelligence - Special volume on computer vision
Feature Detection with Automatic Scale Selection
International Journal of Computer Vision
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Speeded-Up Robust Features (SURF)
Computer Vision and Image Understanding
SURF: speeded up robust features
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part I
Hi-index | 0.00 |
Stereo matching is currently one of the most important research topics in domain of computer vision. The improved SURF based on stereo matching algorithm is proposed in this paper, in order to match feature points more efficiently and accurately. The procedure of this method is following: Firstly, we used the algorithm based on Speeded-Up Robust Features (SURF) to detect and descript the feature points of image sequence, used normalized correlation (NCC) for the initial match. Secondly, we eliminated mismatching points by using random sample consensus algorithm (RANSAC). Lastly, we used the least square method for precision matching. Three Experiments and table analysis show that the matching accuracy of this algorithm is better than the traditional SIFT, SURF based on stereo matching algorithm and the running time is quite fast. So, it can be used in the pure software feature-point-based stereo vision system.