Infrared Target-Background Separation Based on Weighted Nuclear Norm Minimization and Robust Principal Component Analysis

The target detection ability of an infrared small (ISTD) system is advantageous in many applications. highly varied nature the background image and characteristics make process extremely difficult. To address this issue, study proposes patch model using non-convex (IPNCWNNM) weighted nuclear norm minimization (WNNM) robust principal component analysis (RPCA). As observed most advanced methods images (IPI), edges, sometimes a crowded background, can be detected as targets due to extreme shrinking singular values (SV). Therefore, WNNM RPCA have been utilized paper, where varying weights are assigned SV rather than same for all existing (NNM) IPI-based methods. alternate direction method multiplier (ADMM) also employed mathematical evaluation proposed work. evaluations demonstrated that terms suppression proficiency, suggested technique performed better cited baseline