Journal
APPLIED OPTICS
Volume 55, Issue 23, Pages 6480-6490Publisher
OPTICAL SOC AMER
DOI: 10.1364/AO.55.006480
Keywords
-
Categories
Funding
- National Natural Science Foundation of China (NSFC) [61403033]
Ask authors/readers for more resources
Because of the poor lighting conditions at night time, visible images are often fused with corresponding infrared (IR) images for context enhancement of the scenes in night vision. In this paper, we present a novel night-vision context enhancement algorithm through IR and visible image fusion with the guided filter. First, to enhance the visibility of poorly illuminated details in the visible image before the fusion, an adaptive enhancement method is developed by incorporating the processes of dynamic range compression and contrast restoration based on the guided filter. Then, a hybrid multi-scale decomposition based on the guided filter is introduced to inject the IR image information into the visible image through a multi-scale fusion approach. Moreover, a perceptual-based regularization parameter selection method is used to determine the relative amount of the injected IR spectral features by comparing the perceptual saliency of the IR and visible image information. This fusion method can successfully transfer the important IR image information into the fused image, and simultaneously preserve the details and background scenery in the input visible image. Experimental results show that the proposed algorithm is able to achieve better context enhancement results in night vision. (C) 2016 Optical Society of America.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available