1 November 2016 Improved temporal high-pass filter non-uniformity correction algorithm based on shearlet transform
Author Affiliations +
Proceedings Volume 10157, Infrared Technology and Applications, and Robot Sensing and Advanced Control; 1015738 (2016) https://doi.org/10.1117/12.2247347
Event: International Symposium on Optoelectronic Technology and Application 2016, 2016, Beijing, China
Abstract
The drawback of temporal high-pass non-uniformity correction algorithm, ghosting and the image blurring, severely degrades the correction quality. In this paper, an improved non-uniformity correction algorithm based on shearlet transform is proposed. First, the proposed algorithm decomposes the original infrared image into one low frequency sub-band and a group of high frequency sub-bands by the shearlet transform. As a powerful mathematical tool, the decomposition of image by shearlet can reveal the detail of the image accurately. As the high frequency sub-bands contain the most of FPN, the FPN is estimated from the high frequency sub-bands by temporal high-pass. Then, the goal of non-uniformity correction can be achieved by subtracting the estimated FPN from the original high frequency sub-bands. At last, the corrected infrared image can be obtained by the inverse shearlet transform. The performance of the proposed algorithm is thoroughly studied with real infrared image sequences. Experimental results indicate that the proposed algorithm can reduce the non-uniformity with less ghosting artifacts but also overcome the problems of image blurring in static areas.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Sheng-hui Rong, Sheng-hui Rong, Hui-xin Zhou, Hui-xin Zhou, Han-lin Qin, Han-lin Qin, Kun Qian, Kun Qian, Dong Zhao, Dong Zhao, Bing-jian Wang, Bing-jian Wang, } "Improved temporal high-pass filter non-uniformity correction algorithm based on shearlet transform", Proc. SPIE 10157, Infrared Technology and Applications, and Robot Sensing and Advanced Control, 1015738 (1 November 2016); doi: 10.1117/12.2247347; https://doi.org/10.1117/12.2247347
PROCEEDINGS
8 PAGES


SHARE
Back to Top