Color and Texture Based Image Matching Algorithm

In order to solve the problem of traditional SURF (Speed Up Robust Feature) based only on image gray information features while ignoring image color and texture information, a SURF feature matching algorithm based on color and texture is proposed. First, extract the image feature points through the Hesssian matrix, describe the feature points through the SURF algorithm, and add the normalized RG color space feature point neighborhood difference information to the feature descriptor to form an improved SURF feature vector; then The two-way matching strategy is used to match the feature points, and then the random sampling consensus algorithm is used to eliminate the mismatched points of the rough matching point set, and reduce the mismatch rate. The experimental results show that the algorithm has a higher performance than the traditional algorithm under the perspective change. Matching accuracy and higher robustness.