Weighted Adaptive Schatten p-Norm and Transpose Variability Model for Infrared Maritime Small Target Detection

With the increasing frequency of marine accidents, the infrared search and tracking system plays a leading role in the detection of maritime distress, both for people and ships. However, the existing traditional infrared weak small target methods usually cause high missing alarm or false alarm in nonuniform sea wave background. Although the tensor model based on low-rank and sparse decomposition (LRSD) has better detection performance for infrared weak small target detection in heterogeneous background, some outstanding issues still exist: 1) incomplete construction of a tensor model; 2) inaccuracy of background rank approximation; and 3) poor robustness to sparse clutter and noise maritime scene. To solve three problems, a weighted adaptive Schatten p-norm and spatial-temporal tensor transpose variability (WASpN-STTTV) model is proposed in this article. First, the spatial-temporal tensor transpose variability (STTTV) model is constructed based on spatial-temporal tensor (STT) and different transpose operators, to reduce the loss of information in the observation tensor. Second, inspired by the Schatten p-norm and weighting strategy, the weighted adaptive Schatten p-norm is introduced into the infrared maritime target recovery model to better approximate the tensor rank. Third, an improved structure tensor is applied to extract the prior target information and suppress sparse clutter. The tensor anisotropic total variation (TATV) regular term is introduced optimization model to remove noise. Finally, an optimization algorithm derived from the alternating direction method of multipliers (ADMM) algorithm framework is designed to solve the WASpN-STTTV model. Extensive experimental results reveal that the proposed method is superior to state-of-the-art baseline methods in terms of performance.