采用方向信息测度和时间域迭代技术的红外图像扩散去噪算法  

Diffusion denoising algorithm for infrared image based on direction information measure and time-domain iterative technology

作　　者：钱惟贤[1] 陈钱[1] 顾国华[1] 管志强[1] 

机构地区：[1]南京理工大学光电工程国防重点学科实验室,江苏南京210094

出　　处：《红外与激光工程》2010年第3期561-565,共5页Infrared and Laser Engineering

基　　金：南京理工大学自主科研专项计划资助项目(2010ZDJH12)

摘　　要：噪声抑制是红外图像处理中的一个重要研究课题,但常用的去噪算法会造成细节的损失。为了有效地抑制噪声,同时保护边缘,文中在P-M扩散方程的基础上,提出了一种新的采用方向信息测度和时间域迭代技术的各项异性扩散滤波算法。方向信息测度的引入能够在抑制噪声的同时有效保护边缘;时间域迭代技术通过时空混合滤波来同时抑制噪声和保护边缘,而且取消了常规扩散算法几十次的迭代运算,使得实时各向异性扩散算法成为了可能。仿真结果表明:该算法的性能优于常规算法,具有良好的前景和实用价值。Noise suppression is an important research topic in the field of infrared image processing. The general methods of noise suppression cause the loss of the details. In order to effectively suppress noise and protect edges, an anisotropic diffusion filter algorithm was proposed based on the direction information measure and the time-domain iterative technology. The nonlinear diffusion equation based on the direction information measure was used to suppress noise and enhance edges. The time-domain iterative technology was applied to avoid dozens of iterations used in the common diffusion algorithm such that the real-time anisotropic diffusion algorithm could be implemented. The simulation results show that the proposed algorithm is better than the traditional algorithms. And this algorithm has a good prospect of development and application value.

关 键 词：噪声抑制 各项异性扩散 方向信息测度 时间域迭代 

分 类 号：TN911.73[电子电信—通信与信息系统]