Infrared bi-polar small target detection via novel ring morphological transformation

Abstract

Despite the progress of the infrared small target detection algorithms in recent years, there are two unresolved challenges in small target detection remain. (1) fast and robust infrared small target detection is still a challenge, and (2) existing algorithms are less effective in detecting bi-polar targets. To solve these problems, we propose a small target detection algorithm based on a novel ring morphological transformation. In the proposed algorithm, we reconstruct the structural elements in four directions based on the ring top-hat transformation and the relationship between the target and circularity. Then, we organically combine the open and closed operations in the four directions to make the algorithm capable of detecting bi-polar targets. The final results are obtained by fusing the results of the open and closed operations in the four directions. Extensive experiments on six datasets demonstrate the effectiveness and robustness of the proposed algorithm. Moreover, the low complexity makes the proposed algorithm satisfy the purpose of real-time detection.