Abstract
Seeking reliable correspondence between multispectral images is a fundamental and important task in computer vision. To overcome the nonlinearity problem occurring in multispectral image matching, a novel, edge-feature-based maximum clique-matching frame (EMCM) is proposed, which contains three main parts: (1) a novel strong edge binary feature descriptor, (2) a new correspondence-ranking algorithm based on keypoint distinctiveness analysis algorithms in the feature space of the graph, and (3) a false match removal algorithm based on maximum clique searching in the correspondence space of the graph considering both position and angle consistency. Extensive experiments are conducted on two standard multispectral image datasets with respect to the three parts. The feature-matching experiments suggest that the proposed feature descriptor is of high descriptiveness, robustness, and efficiency. The correspondence-ranking experiments validate the superiority of our correspondences-ranking algorithm over the nearest neighbor algorithm, and the coarse registration experiments show the robustness of EMCM with varied interferences.
Highlights
Because of the differences in imaging mechanisms, multispectral imaging devices can acquire information of scenes under different band conditions, and they make up for the deficiency of single-band imaging sensors
We propose a novel, edge-feature-based maximum clique-matching frame (EMCM), which contains three main steps: (1) structural feature extraction, (2) feature correspondence ranking, and (3) improved maximum clique matching
We constructed the putative correspondences through a nearest neighbor search, as shown in Figure 2i, and we reweighted each correspondence by considering the distinctiveness of each keypoint
Summary
Because of the differences in imaging mechanisms, multispectral imaging devices can acquire information of scenes under different band conditions, and they make up for the deficiency of single-band imaging sensors. Different from the EOH descriptor, the Log-Gabor histogram descriptor (LGHD) [34] and multispectral feature descriptor (MFD) [35] use multiscale and multi-oriented Log-Gabor filters to replace the multi-oriented spatial filters In spite of these three algorithms having a certain matching effect on multispectral images, there are still some shortcomings: (1) A large number of common edge features of multispectral images are not fully considered; (2) when using edge information, numerous low-value, repetitive feature structures will be extracted; (3) when constructing feature descriptors for keypoints, the encoding method is not concise, and the data storage efficiency is low; and (4) in the process of feature description, too many mismatched feature points will remain and participate in subsequent matching.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
