Derivative Entropy-Based Contrast Measure for Infrared Small-Target Detection

Robust and effective detection of small target and false alarm (FA) suppression are the key techniques in infrared search and track systems. In this paper, the derivative entropy-based contrast measure (DECM) is proposed for small-target detection under various complex background clutters. Initially, different directional derivatives of an infrared image are calculated based on the facet model. Then, by analyzing the derivative properties of small target, the primitive entropy formula is improved by incorporating derivative information. With the improved entropy, the contrast measure is constructed to enhance small target and suppress background clutters in each derivative subband. Finally, the contrast measure maps derived from derivative subbands are fused together. The small target could be segmented easily from the fusion result. Experimental results demonstrate that DECM could effectively enhance dim small targets and suppress complex background clutters. Besides, DECM is also robust to infrared small-target images with noises of different levels. The detection results achieve higher detection ratio and lower FA compared with those of other methods under various infrared scenes.

I. Introduction

Infrared search and track (IRST) systems have been widely applied to infrared warning and defense alertness projects [1], [2]. Compared with radar systems and visible light systems, IRST systems have good resolution, strong smoke penetration, and good concealment [3], [4]. As a key technique in IRST systems, the early detection of small target (e.g., enemy airplane, missile, and ship in the distance) with unknown position and velocity is conducive to warn potential threat and take countermeasures [5]–[7]. However, infrared images might be contaminated by the noise and nonuniformity of gray distributions in the imaging process. The images tend to have poor contrast. Besides, the projected targets at a far distance are usually small and dim without concrete shapes, textures, and structure information [2], [8], [9]. Therefore, it is a difficult and challenging task to detect small target under such low signal-to-clutter ratio (SCR) and complex infrared backgrounds. To solve the detection problem, various methods have been proposed in the past few decades.