摘要
基于各向异性扩散去噪模型(P-M)的图像恢复思想,提出了一种新的利用像素预判断的图像恢复并行高效算法模型。该模型充分利用图像中每一个像素的预判断知识,设计COW工作站机群以结合实现并行算法。实验结果表明,该算法既能提高P-M算法的运行效率,又能获得与其相当的恢复质量。
Based on the idea of anisotropic diffusion equation, a new efficient parallel algorithm model, which used the concept of pixels pre-judgment to restore images, was proposed. This model made full use of the pre-judgment knowledge in the image, and designed a workstation of COW cluster to achieve the parallel algorithms. Experimental results show that the algorithm can improve the operation efficiency of P-M algorithm and obtain comparable restoration quality.
出处
《计算机应用》
CSCD
北大核心
2009年第5期1353-1354,1358,共3页
journal of Computer Applications
基金
重庆市科委基金资助项目(CST2005BB0061KJ070514)
关键词
图像恢复
坏点检测
并行算法
各向异性扩散方程
image restoration
Impulse detection
parallel algorithm
anisotropic diffusion equation
