Title: Image feature point matching method based on improved BRISK
Authors: Qian Shi; Yong Liu; Yichun Xu
Addresses: College of Computer and Information, China Three Gorges University, Yichang, Hubei, China ' College of Computer and Information, China Three Gorges University, Yichang, Hubei, China ' College of Computer and Information, China Three Gorges University, Yichang, Hubei, China
Abstract: In order to improve the accuracy of the BRISK algorithm of feature point matching, an improved BRISK feature point matching method is proposed by combining the SIFT algorithm idea with high matching accuracy. Firstly, the Gaussian image pyramid is established, then the FAST algorithm is used to detect the feature points of each image layer in the image pyramid, and the intensity centroid method is used to determine the direction of the feature points, so the feature points obtained have scale information and direction information. Then, using the BRISK descriptor to describe the feature points, using the Hamming distance to measure the similarity of the feature descriptors, and calculating the matching accuracy and matching time. The experimental results show that the accuracy of the method proposed in this paper is higher than that of the BRISK algorithm and the time is better than that of the SIFT algorithm and the BRISK algorithm.
Keywords: image matching; feature point matching; intensity centroid; improved BRISK; image pyramid.
DOI: 10.1504/IJWMC.2021.114129
International Journal of Wireless and Mobile Computing, 2021 Vol.20 No.2, pp.132 - 138
Received: 20 Aug 2020
Accepted: 14 Sep 2020
Published online: 09 Apr 2021 *