一种基于局部特性的红外弱小目标检测方法

针对红外图像弱小目标检测程序复杂和虚警率高的问题,提出了一种基于局部特性检测红外弱小目标的方法.该方法主要根据待检测像素点局部灰度比、局部灰度差、局部能量比以及局部能量差值的大小进行目标检测,避免了传统方法需对红外图像进行背景预测、图像增强等处理,有效地改善了低信噪比下红外图像小目标的检测性能.通过仿真实验对该方法与2种传统方法进行了测试和对比,证明了该方法的有效性和优越性.

Method for Infrared Dim Target Detection Based on Local Quality

Aimed at the problem of complex program and high false alarm rate in the infrared dim target detection, a method for dim infrared target detection is proposed, based on local quality, in this paper. The target is detected, mainly according to the local grayscale ratio, local grayscale difference, local energy ratio and local energy difference of the undetected pixels, which avoids using the conventional method where the background prediction and image enhancement and so on are needed for the infrared image processing, and improve effectively the detection performance on infrared dim target in the case of lower SNR. By experiments on testing and comparing the proposed method with two conventional methods, it is shown that the method is of effectiveness and superiority.