An efficient mixed noise removal technique from gray scale images using noisy pixel modification technique

摘要

Removing or reducing noises from image is a very active research area in image processing domain. This paper presents an efficient noise removal technique to restore digital images corrupted by mixed noise, preserving image contents optimally. The proposed filtering technique consists of two steps: the noisy pixel detection step using fuzzy technique and the mixed noise filtering step. Noises addressed in this method are a combination of salt and pepper noise and Gaussian noise. This method reduces mixed noises considerably without compromising on edge sharpness. Experimental results show that the proposed technique consistently outperforms many existing fuzzy based algorithms while balancing the tradeoff between noise reduction and detail preservation. Hence, this mixed noise removal technique finds application in various segments of image processing like digital television, medical image processing, digital camera, surveillance systems etc.