摘要
AKAZE特征检测算法具有鲁棒性好,匹配率高等特点,为解决其实时性差的问题,提出将ORB与AKAZE相结合的改进算法。利用oFAST算法检测特征点然后采用M-LDB算法计算其描述符,使用汉明距离进行图像粗匹配,最后用RANSAC算法剔除误匹配点,得出匹配结果。经反复的实验对比证明,改进后的算法与ORB算法相比匹配正确率更高。与AKAZE算法相比匹配速度更快。且改进后的算法在不同模糊程度、不同JPEG图像压缩、不同光照程度以及不同旋转角度变化下的图像匹配性能良好。
The AKAZE feature detection algorithm has the characteristics of good robustness and high matching rate.To solve the problem of poor real-time performance,an improved algorithm combining ORB and AKAZE is proposed.The feature points are detected by the oFAST algorithm and then the descriptors are calculated by the M-LDB algorithm.The Hanming distance is used to perform image rough matching.Finally,the RANSAC algorithm is used to eliminate the mismatched points,and the matching result is obtained.The repeated experimental comparison proves that the improved algorithm has higher matching accuracy than the ORB algorithm.Matching speed is faster than the AKAZE algorithm.And the improved algorithm has good image matching performance under different blurring degrees,different JPEG image compression,different illumination levels and different rotation angle changes.
作者
赵柏山
张楠
ZHAO Baishan;ZHANG Nan(School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870)
出处
《计算机与数字工程》
2021年第8期1651-1655,共5页
Computer & Digital Engineering
基金
辽宁省自然科学基金项目(编号:2019-ZD-0213)资助。