基于RX算子的背景杂波抑制  

Background Clutter Suppression Using RX Operator

作　　者：李世杰[1] 秦翰林[2] 周慧鑫[2] 

机构地区：[1]中国空空导弹研究院,河南洛阳471009 [2]西安电子科技大学技术物理学院,西安710071

出　　处：《半导体光电》2012年第4期575-578,共4页Semiconductor Optoelectronics

基　　金：中央高校基本科研业务费专项资金资助项目(72005623);国家部委科技项目资助项目(625010214);教育部科学技术研究重点项目(108114)

摘　　要：红外图像复杂背景杂波抑制是红外监视告警系统发现远距离弱小目标的难题。文章根据红外图像中目标和背景杂波的特性,提出了一种将RX算子与非线性扩散方程相结合的弱小目标背景抑制新方法。该方法首先采用非线性扩散方程对图像进行多尺度分解,获得图像的多尺度特征,然后,根据目标和背景杂波信号系数在不同尺度之间的差异,通过应用RX算子进行处理,从而将红外图像中弱小目标和背景杂波分离,达到抑制背景的目的。实验结果显示,与二维最小均方误差滤波方法相比较,该方法能有效地检测出信杂比在1.6以上的目标。Complex background clutter suppression is a key problem for finding long distance dim and small targets in infrared （IR） warning system. Based on analyzing the characteristics of IR image target and background clutter, a new method of background suppression for dim and small targets was proposed by combing RX operator with nonlinear diffusion equation. Firstly, the nonlinear diffusion equation was adapted to decompose the input infrared image and extract multi-scale detailed features of the image. Then, according to difference between target and the background clutter signal, RX operator is introduced to suppress the background details and enhance the target information. Experimental results demonstrate that, compared with two dimensional least mean square （TDLMS） method, the proposed method can segment the infrared target images with SCR higher than 1.6 effectively.

关 键 词：目标检测 背景抑制 RX算子 非线性扩散方程 

分 类 号：TN216[电子电信—物理电子学]