Infrared Small-Target Detection Based on Background-Suppression Proximal Gradient and GPU Acceleration

Patch-based methods improve the performance of infrared small target detection, transforming the detection problem into a Low-Rank Sparse Decomposition (LRSD) problem. However, two challenges hinder the success of these methods: (1) The interference from strong edges of thebackground, and (2) the time-consuming nature of solving the model. To tackle these two challenges,we propose a novel infrared small-target detection method using a Background-SuppressionProximal Gradient (BSPG) and GPU parallelism. We first propose a new continuation strategy tosuppress the strong edges. This strategy enables the model to simultaneously consider heterogeneouscomponents while dealing with low-rank backgrounds. Then, the Approximate Partial SingularValue Decomposition (APSVD) is presented to accelerate solution of the LRSD problem and furtherimprove the solution accuracy. Finally, we implement our method on GPU using multi-threadedparallelism, in order to further enhance the computational efficiency of the model. The experimentalresults demonstrate that our method out-performs existing advanced methods, in terms of detectionaccuracy and execution time.