An Improved Adaptive Threshold BRISK Feature Matching Algorithm Based on SURF

This paper proposes an adaptive threshold image matching strategy based on the SURF algorithm and the improved BRISK algorithm. The process of algorithm can be divided into four parts. First, histogram equalization is applied to enhanced the image. Then the SURF threshold is determined by calculating the complexity of the image. After that, the SURF detection is used to obtain the image feature points. Then the feature points are processed by BRISK descriptor. Finally, the Lowe's algorithm is adopted to obtain necessary matching points. The results of experiment verified that, compared with SURF algorithm in speed, the improved algorithm has computational advantages, and the matching accuracy is obviously enhanced compared to the BRISK algorithm.