摘要
针对传统图像匹配算法面临的特征提取的信息较少,匹配成功率不高以及匹配速率较低等问题,提出基于SURF特征提取和Brute-Force搜索的图像匹配算法。利用SURF算法中的Hessian矩阵来获取图像中鲁棒性较好的突变点,并使用不同尺寸的滤波器同时处理尺寸空间多层图像的突变点,以此来提高匹配速率,最后采用Brute-Force搜索算法对图像特征点进行最佳匹配,以此来提高匹配成功率。实验表明,该算法在图像匹配效果和匹配效率方面都表现良好。
The image matching algorithm based on SURF feature extraction and Bruce-Force search is proposed based on the fact that the features of traditional image matching algorithm are less information,the matching success rate is low and the matching rate is low.The hessian matrix in the SURF algorithm is used to obtain the robust points with better robustness in the image,and different size filters are used to simultaneously process the mutation points of the multi-layer image in the size space to improve the matching rate.Finally,the Bruce-Force is used.The search algorithm best matches the image feature points to improve the matching success rate.Experiments show that the algorithm performs well in image matching and matching efficiency.
作者
厉彦福
LI Yanfu(College of Geomatics,Shandong University of Science and Technology,Qingdao Shandong,266590,China)
出处
《北京测绘》
2019年第11期1352-1355,共4页
Beijing Surveying and Mapping
