Infrared Dim and Small Target Detection Based on Strengthened Robust Local Contrast Measure
Zehao Li, Shouyi Liao, Tong Zhao
Abstract
Infrared small target detection is a popular issue in the field of computer vision. Traditional methods have been researched a lot and can basically deal with target detection in usual scenarios. However, small and dim targets do not have obvious characteristics and are easily interfered by clutter in a complex background. This letter studied common target detection methods such as low-rank and sparse representation as well as local contrast measure (LCM), a new method called strengthened robust local contrast measure (SRLCM) algorithm is proposed here. By solving non-convex problem more rigorously and analyzing small target features carefully, SRLCM can achieve a better detection and segmentation performance in an irregular background.