At present, satellite remote sensing technology has become a research hotspot for establishing space sensing systems at home and abroad. In remote sensing image processing, detecting and matching feature points in remote sensing images play a key role. In this article, three mainstream image feature extraction and matching methods are introduced, namely the SIFT method, SURF method, and ORB method. Experiments are designed to test and compare the operating efficiency and matching accuracy of these three feature point extraction and matching algorithms in remote sensing image scenes. The conclusion is as follows: ORB method is the fastest, SURF is second, and SIFT is the slowest. In terms of correct matching rate, the SURF algorithm and ORB algorithm have the advantage and show a high correct matching rate in remote sensing image scenes.
Research on feature matching algorithm of remote sensing scenes
12.10.2022
1415348 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Image Feature Matching Algorithm Research on Topography Measurement
British Library Online Contents | 2008
|