Infrared Small Target Detection Based on the Difference Variance Weighted Enhanced Local Contrast Measure

Abstract

Infrared Search and Tracking System (IRST) has been widely applied in many fields, but it is still challenging to detect small infrared targets in complex backgrounds. To address this problem, this paper proposes a detection framework known as Difference Variance Weighted Enhanced Local Contrast Measure (DVWELCM). First, an enhanced local contrast measure (ELCM) is used to enhance small targets and suppress complex background while improving signal clutter ratio (SCR). Second, a weighting function of the difference variance is adopted to further reduce the influence of the background and improve the robustness. Finally, by integrating enhanced local contrast measure (ELCM) and difference variance weighting (DVW), an adaptive threshold segmentation method is used to extract the real target. Extensive experiments have been performed on data sets in different scenarios. The results show that compared with the existing methods, the proposed method has better detection performance in complex backgrounds.