Research on Image Matching Based on Edge Point Feature

The feature point extraction method is proposed combining with LoG algorithm, and Scale Invariant Feature Transform algorithm is applied to image feature extraction. First of all, LoG algorithm is used to compute the edge points, the edge points are sorted by the gradient values, and the points of lager gradient are selected as feature points; Then compute the feature vector of feature points with SIFT algorithm, and the minimum distance algorithm is used to find the matching points of two images; Finally, the wrong matching points are excluded by the ratio algorithm of the most relevant point and the second relevant point. The experimental results prove that the method proposed in this paper has excellent adaptive and accurate characteristics for the illumination, transfer and the rotation transform and the match accuracy was over 90%.