基于偏微分方程的非线性弱目标增强算法  

Nonlinear dim target enhancement algorithm based on partial differential equation

作　　者：王艳华[1] 刘伟宁[1] 陈爱华[1] 王勇[1] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所

出　　处：《大连海事大学学报》2008年第2期57-60,共4页Journal of Dalian Maritime University

基　　金：中国科学院二期创新基金资助项目(C04708Z)

摘　　要：为解决低对比度弱目标单帧检测率较低的问题,提出一种基于偏微分方程(PDE)的弱目标增强算法,对传统的P-M各向异性扩散模型进行改进,在扩散函数中引入锐化因子,通过选取适当的梯度门限和锐化强度参数,自适应地平滑背景和增强弱目标.该算法与中值滤波相结合,消除了传统模型在处理椒盐噪声方面的缺陷.仿真实验表明,该算法可有效克服传统滤波器的缺陷,在抑制干扰和噪声的同时增强目标.To solve the problem of dim target detection with lower signal to noise/clutter ratios, an image enhancement algorithm was developed by improving the traditional Perona ＆ Malik diffusion model based on partial differential equation （PDE）. Sharpening factor was introduced in the diffusion equation, which can adaptively smooth the background clutter and enhance the dim target through selecting the appropriate parameters. Moreover the algorithm was combined with median filter to avoid the interference of white-black noise. Experimental results show that the algorithm can overcome the traditional filter＇s drawbacks and enhance dim target while inhibiting interference and noise.

关 键 词：弱目标 图像增强 偏微分方程 扩散滤波 

分 类 号：TN911.73[电子电信—通信与信息系统]