Shearlet-based infrared target detection algorithm on complex sea

Aiming at the infrared target detection issue on complex sea, using shearlet anisotropy and geometry characteristic of its coefficients, a shearlet-based infrared target detection algorithm on complex sea is proposed. The algorithm adopts shearlet to acquire edge image and obtain the direction information of the edge, which greatly improves the probability of detection and identification of the sea-sky line. Then, according to the sea-sky line position, the edge image is weighted, which restrains sea clutter and keeps the target information. After the cut-line-mean filtering, the weighted edge image is segmented. Finally, the segmented image is processed with mathematical morphology and the target is detected. Experimental results show that using the proposed algorithm the image target can be detected from a single frame image, and the algorithm can achieve good detection results under the complex sea background with sun bright band, sea clutter and other interference.