基于视觉显著性的红外图像弱小目标检测方法  

Detection Algorithm of Small Dim Target Based on Visual Saliency Infrared Image

作　　者：徐凌[1] 谭贤四[1] 

机构地区：[1]空军预警学院,武汉430019

出　　处：《空军预警学院学报》2013年第6期397-401,406,共6页Journal of Air Force Early Warning Academy

摘　　要：针对复杂背景下红外图像弱小目标检测难题,提出了一种基于视觉显著性的目标检测方法.首先利用基于视觉注意机制的显著性模型计算图像的显著度图,然后根据图像的显著度图确定目标的区域,最后在原图像上确定目标的精确位置.仿真实验结果表明,本文方法能够很好地检测出红外图像中的弱小目标;与常用背景抑制法相比,该方法不仅较好地突显了目标,而且极大地消除了虚警干扰.Aimed at the problem of detecting a small dim target under the infrared image of complex background, this paper proposes a method for the target detection based on the visual saliency. Firstly, the saliency map of the image is computed by using the visual attention mechanism-based saliency model, next, the target＇ s region can be determined in terms of the image saliency, and finally, the accurate position is located on the original image. Simulation experiments show that this proposed method is of excellent performance on detecting the small dim targets in the infrared image, and can not only feature the target, but also eliminate the interference of false alarms remarkably, in contrast with the conventional method of background suppression.

关 键 词：视觉显著性 红外图像 弱小目标检测 

分 类 号：TN958[电子电信—信号与信息处理] TP391[电子电信—信息与通信工程]