8 March 2017 Infrared and visible image fusion with the target marked based on multi-resolution visual attention mechanisms
Author Affiliations +
Proceedings Volume 10255, Selected Papers of the Chinese Society for Optical Engineering Conferences held October and November 2016; 102552V (2017) https://doi.org/10.1117/12.2264771
Event: Selected Papers of the Chinese Society for Optical Engineering Conferences held October and November 2016, 2016, Jinhua, Suzhou, Chengdu, Xi'an, Wuxi, China
Abstract
During traditional multi-resolution infrared and visible image fusion processing, the low contrast ratio target may be weakened and become inconspicuous because of the opposite DN values in the source images. So a novel target pseudo-color enhanced image fusion algorithm based on the modified attention model and fast discrete curvelet transformation is proposed. The interesting target regions are extracted from source images by introducing the motion features gained from the modified attention model, and source images are performed the gray fusion via the rules based on physical characteristics of sensors in curvelet domain. The final fusion image is obtained by mapping extracted targets into the gray result with the proper pseudo-color instead. The experiments show that the algorithm can highlight dim targets effectively and improve SNR of fusion image.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yadong Huang, Yadong Huang, Kun Gao, Kun Gao, Chen Gong, Chen Gong, Lu Han, Lu Han, Yue Guo, Yue Guo, } "Infrared and visible image fusion with the target marked based on multi-resolution visual attention mechanisms", Proc. SPIE 10255, Selected Papers of the Chinese Society for Optical Engineering Conferences held October and November 2016, 102552V (8 March 2017); doi: 10.1117/12.2264771; https://doi.org/10.1117/12.2264771
PROCEEDINGS
8 PAGES


SHARE
Back to Top