Adaptive-fuzzy filter algorithm
Yu Ding · Harbin Gongcheng Daxue Xuebao/Journal of Harbin Engineering University · 2006
There are simultaneous Gauss and impulse noises in the great mass of nature images,while it is difficult to get rid of Gauss noise and impulse noise only by Median filter or Mean filter at the same time.Based on analyzing traditional filter algorithm,an improved adaptive-fuzzy algorithm is given in this paper.Firstly,the algorithm demarcates impulse noise and separate two kinds of the noise in each part by noise detection.Then the adaptive median filter is applied to remove the impulse noise.Based on filtering the impulse noise,the concept of membership function in fuzzy control is introduced to improve the mean filter algorithm.And combining histogram,the threshold of the membership function is proposed,which overcame the fuzziness of picture caused by mean filter.Finally,a bran chylogy algorithm is proposed to get median in image filter so as to improve the practicability.In real application,the new algorithm shows better performance than traditional algorithm in an image with Gauss noise and impulse noise.