一种基于区域信息的图像高斯噪声滤除方法

Regional information-based approach for removing Gaussian noise on images

提出了一种滤除加性脉冲噪声的模糊加权平均滤波器 ,其加权系数是一个携带了原图像信息的隶属度函数 ,同时该滤波器还根据图像区域特性差异采用了不同的滤波方法 .新的滤波器优点是简单快速 ,和 neuro-fuzzy滤波器相比又无需训练学习时间 .实验表明该方法滤波效果优于传统的滤波器和其他模糊滤波器 .

A weighted fuzzy mean filter for the removal of additive impulse noise is proposed in this paper.The weighted coefficients are membership function that carries information of the original image.Meanwhile,the filter adopts different method of image restoration in the light of divergence of image regional character.Similar to conventional filters,the new filter has the benefits of being simple and quick.Relative to neuron-fuzzy filters,it doesn't need training and learning time.Experimental result shows that the filter gives superior performance compared with conventional filters and other fuzzy filters.