Remote Sensing Image Matching Based on Improved ORB Algorithm
Aiming at the problems of high false matching rate and slow matching speed in remote sensing image matching,combined with the idea of consistent local structure of similar images,a remote sensing image matching algorithm combining ORB feature extraction operator and BRISK feature description operator was proposed.First,the feature point extraction operator of the ORB algorithm was used to detect the feature points;second,the feature point description operator of the BRISK algorithm was used to establish feature descriptors for the detected feature points;then,brute force matching based on the Hamming distance was used to complete the feature point matching;finally,the LPM algorithm was used to eliminate the wrong matching points.The experimental results showed that the number of feature points,matching accuracy and running speed extracted by the proposed algorithm all met the requirements of remote sensing image matching;compared with the ORB and BRISK algorithms,it achieved a better balance between speed and accuracy,showing a good matching performance.
万方数据
维普
