摘要
运用马尔科夫随机场(MRF)进行图像处理时,对图像平滑区域与边缘区域所采用的处理方法不加区别,会导致大量冗余程序且运行时间过长。针对该问题,提出基于模糊分类的MRF图像恢复方法,根据图像子块内服从不同分布的像素统计特征,对图像子块进行模糊分类,在分类基础上应用MRF进行图像恢复。对退化的二值图像进行恢复实验,结果表明,与MRF方法相比,基于模糊分类的MRF方法能减少程序运行时间,改善去噪效果。
The processing of smooth region and edge region is identity using Markov Random Fields（MRF）,which leads to a lot of redundancy program and too long operation time.Aiming at this problem,this paper presents MRF image restoration method based on fuzzy classification.Image subblock can be fuzzy classified according to its statistical characteristics obeying different distribution,the image can be restored by MRF.The blurred binary image is restored by this method and the experimental results show that comparing with the MRF method,the method based on fuzzy classification can reduce operation time and improve denoising effect.
出处
《计算机工程》
CAS
CSCD
北大核心
2011年第10期198-199,221,共3页
Computer Engineering
基金
陕西省教育厅自然科学专项基金资助项目(2010JK563)
关键词
马尔科夫随机场
图像恢复
模糊分类
最大后验估计
先验概率
Markov Random Fields（MRF）
image restoration
fuzzy classification
maximum a posteriori estimation
prior probability
