Small target detection based on weighted self-information map

The target growing effect and large calculation are two major drawbacks in small target detection based on the self-information map. Consequently, a weighted self-information map is proposed to overcome those disadvantages in this paper. The removal of the target growing effect from the weighted self-information map is proved theoretically, which is capable of improving the capability of small target detection. Based on the weighted self-information map and the improved region growing technology, a small-target detection approach is constructed. The signal-to-noise ratio, peak signal-to-noise ratio, region nonuniformity, the probability of detection and the probability of false alarm are adopted to demonstrate the performance of the proposed approach. Both quantitative analysis and qualitative comparison confirm the validity and efficiency of the proposed approach.