8 October 2015 Retina-inspired redundant dictionary for infrared small target detection based on sparse representation
Author Affiliations +
Proceedings Volume 9678, AOPC 2015: Telescope and Space Optical Instrumentation; 96780O (2015) https://doi.org/10.1117/12.2199395
Event: Applied Optics and Photonics China (AOPC2015), 2015, Beijing, China
Abstract
A novel redundant dictionary of sparse representation is proposed to detect infrared small targets. Redundant dictionary plays an important role in sparse representation, and is the key point to guarantee the sparseness of solution. Different from traditional intensity dictionary, in this letter, the gradient information and retinal sampling pattern are used to construct the novel redundant dictionary, in which pixels near the centre have higher sampling density. It can effectively reduce the influence of background clutter around target. Consequently, it is easy to discriminate target from background according to the sparsity of the representation coefficient vector. Experimental results demonstrate that the presented method is robust and efficient.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Miao Li, Miao Li, Yunli Long, Yunli Long, Wei An, Wei An, Yiyu Zhou, Yiyu Zhou, } "Retina-inspired redundant dictionary for infrared small target detection based on sparse representation", Proc. SPIE 9678, AOPC 2015: Telescope and Space Optical Instrumentation, 96780O (8 October 2015); doi: 10.1117/12.2199395; https://doi.org/10.1117/12.2199395
PROCEEDINGS
5 PAGES


SHARE
Back to Top