Research on an improved SURF matching algorithm for door handle images

Aiming at the requirement of fastness and accuracy of the feature extraction and matching of the door handle image, an improved SURF algorithm is proposed. The algorithm mainly improves the problem that the image is more difficult to extract a large number of information points at smooth area.increases the edge detection algorithm, obtains the edge information of the image and then performs the morphological processing, And obtains the key edge information of the door handle image by the dilation operation and the Opening Operation, and then extracts the key points, and obtains more obvious information. Compared with the original SURF algorithm, the accuracy of the improved SURF algorithm is better than that of the original SURF algorithm at smooth area and the matching accuracy is improved obviously. Because the total number of extracted feature points decreases, so the matching time is greatly saved, thus enhancing the real-time performance of the algorithm.