Infrared Small Maritime Target Detection Based on Integrated Target Saliency Measure

Robust and effective detection of a small target in an infrared maritime image is key technology search tracking applications. Infrared challenging task due to the factors such as dim targets various complex backgrounds caused by sun glitters strong waves. In this article, integrated saliency measure (ITSM) based on local nonlocal spatial information proposed improve performance. We combine heterogeneity property self-correlation background with targets' sparsity separate real from clutters. First, calculation cross-window standard deviation (CSD) extract candidate preliminarily, which enhances intensity difference between neighboring background. Meanwhile, low-rank representation (LRR) applied prediction removal, followed adaptive threshold segmentation enhance saliency. Finally, we integrate results obtained two steps mentioned above further suppress Then, are extracted iterative map so generate seed map, addition, expansion strategy exploited keep full areas. Experimental six datasets show that method outperforms state-of-the-art methods terms robustness accuracy.