30 October 2009 Contour-based multisensor image registration using gradient vector features
Author Affiliations +
Proceedings Volume 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis; 74954A (2009) https://doi.org/10.1117/12.832594
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China
Abstract
Multi-sensor image registration plays an important role in the automatic targets acquirement (ATA). In this paper, two kinds of new multi-sensor image registration methods based on contour feature are put forward. One is gradient vector correlation coefficient method which is based on contour characteristics of the image, the other is called relative gradient difference method which is in fact a measurement of differences between out-contour area and inner area of the target. Lots of experimental data composed of satellite images and real-time infrared images are used to verify the above two algorithms, and the results show that the two proposed methods can have good performances for both day and night infrared images.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Delie Ming, Delie Ming, Mi Wang, Mi Wang, Pengfei Gao, Pengfei Gao, "Contour-based multisensor image registration using gradient vector features", Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 74954A (30 October 2009); doi: 10.1117/12.832594; https://doi.org/10.1117/12.832594
PROCEEDINGS
8 PAGES


SHARE
Back to Top