Ring Local Contrast Measure Weighted by Spatial–Temporal Prior for Infrared Small Target Detection

Infrared (IR) small target detection faces two major challenges: single-frame-based methods are highly sensitive to random noise, while multi-frame-based approaches often neglect the dynamic nature of background scenes, resulting in high false alarm rates. To address these issues, this paper proposes a novel ring local contrast measure weighted by spatial–temporal prior (WSTRLCM). WSTRLCM comprises two key modules: a spatial-temporally enhanced ring local contrast and a spatial-temporal weighting factor. Specifically, a spatial–temporal enhancement scheme is first applied to the original image sequence to boost target saliency, and the ring contrast computation is optimized for improved efficiency. For the weighting factor, a small target enhancement strategy based on image block gradients is introduced. By incorporating structure tensor analysis, effective back-ground suppression is achieved, resulting in a refined spatial-temporal enhancement sequence. Furthermore, a tailored multi-frame difference approach, designed according to target motion characteristics, is employed to generate the final weighting factor. The Hadamard product of the local contrast and the weighting factor is then computed, and adaptive thresholding is applied to obtain the final detection map. Experimental results on public datasets demonstrate that WSTRLCM achieves superior performance compared to several state-of-the-art methods, exhibiting enhanced target detection capability, strong background suppression. The code is available at https://github.com/biangbiangliang/WSTRLCM.