Infrared Dim Target Detection Based on Multi-Feature Fusion

Abstract

A novel infrared(IR) dim target detection algorithm based on multi-feature fusion is proposed in this paper. Firstly, the gray residuals map is obtained by calculating the 8-directional local gray residual. Secondly, the image is divided into a series of local image patches by using a sliding window, then the intensity mean of local image patches is constrained to achieve the local intensity mean constrained map. Lastly, the local image patch is further divided into 12 directional blocks, and the gradient direction constrained saliency map can be obtained by constraining the gradient direction of the pixels in each directional block. Then, the final saliency map is obtained from the above three feature maps by dot product operation, and dim targets are separated by threshold segmentation. The experimental results show that the proposed algorithm not only can effectively suppress background clutter and eliminate false targets, but also can accurately detect dim targets with high detection rate.