Infrared and visible sequential images registration based on the motion characteristics
Zhong-zhong Xu, Zhi-hua Huang, Zhi-qiang Chi · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2014
The research has been done on how to deal with the correspondences of sequential images, and the difficulties caused by multi-sensor. The paper proposes and realizes the registration methods addressing visible and infrared sequences, and the experiments show the validity and the accuracy of the proposed methods. At the first, the research on the sequence-to-sequence rough registration by their mass centers is studied. In the method, we represent the trajectory in the form of curves, and then use the curve-matching method to match the trajectories, and finally acquire the correspondence of the two sequences both in time and space. Then, compared to the rough matching methods above, an automatic visual–infrared image sequence accurate registration method based on Co-motion and the RANSAC algorithm with the RSCJ algorithm for higher efficiency are proposed.