Title :
Cross-Spectral Stereo Matching Based on Local Self-Similarities and Image Moments
Author :
Mouats, Tarek ; Aouf, Nabil
Author_Institution :
Dept. of Inf. & Syst. Eng. (DISE), Cranfield Univ., Shrivenham, UK
Abstract :
Recent developments in the infrared industry and the availability of relatively cheaper infrared cameras attracted more attention towards the simultaneous utilization of visible and infrared images and solving the correspondence problem between them. The Microsoft Kinect sensor represents one of the most widespread examples of low cost multimodal setups. In this context, we investigate the feasibility of matching features extracted from cross-spectral stereo cameras using a sparse approach. First, a set of stable features are extracted from both images using the Scale Invariant Feature Transform (SIFT). Then, descriptors are computed from local self-similarities around the selected key points. The combination of image moments is also investigated and expected to improve the matching process. Experimental results show the performance of this approach to the multimodal correspondence challenge.
Keywords :
cameras; feature extraction; image matching; image sensors; infrared imaging; spectral analysis; stereo image processing; Microsoft Kinect sensor; SIFT; cross-spectral stereo cameras; cross-spectral stereo matching; features extraction matching; image moments; infrared cameras; infrared images; infrared industry; local self-similarities; matching process; multimodal correspondence; scale invariant feature transform; sparse approach; visible images; Cameras; Correlation; Detectors; Feature extraction; Image representation; Stereo vision; Three-dimensional displays; SIFT; multimodal feature matching; self-similarity; stereo correspondence;
Conference_Titel :
Systems, Man, and Cybernetics (SMC), 2013 IEEE International Conference on
Conference_Location :
Manchester
DOI :
10.1109/SMC.2013.691