摘要
SURF是在SIFT基础上提出的一种图像特征点提取算法。针对传统算法误匹配点多和计算量大等问题,提出一种基于改进SURF的快速图像匹配算法。该算法通过引入对角降维与角度删减方法,分别对SURF算法中特征点描述子进行降维和误匹配点剔除,以提升匹配速度和精确度。实验结果表明,与传统算法相比,该算法提高了1%~10%的匹配正确率,以及8%~30%的效率。
Speeded up robust features(SURF)is an image feature point extraction algorithm based on scale-invariant feature transform(SIFT).In order to solve the problems of multiple error matching points and heavy computation in traditional algorithms,this paper proposes a fast image matching algorithm based on the improved SURF.By introducing diagonal dimension reduction and angle reduction methods,the algorithm reduces the dimension of feature point descriptors in SURF and eliminates mismatched points respectively.The algorithm improves matching speed and accuracy.The experiment shows that compared with the traditional algorithm,the proposed algorithm improves the matching accuracy from 1%to 10%and the efficiency from 8%to 30%.
作者
黄云彬
焦良葆
曹雪虹
HUANG Yun-bin;JIAO Liang-bao;CAO Xue-hong(School of Communications and Information Technology,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;Institute of Artificial Intelligence Industrial Technology,Nanjing Institute of Technology,Nanjing 211167,China)
出处
《软件导刊》
2020年第9期220-223,共4页
Software Guide
关键词
SURF算法
特征点匹配
对角降维
角度删减
SURF algorithm
feature point matching
diagonal dimension reduction
angle reduction
