A Mean Filter With Adaptive Window Size
Download as PDF
DOI: 10.23977/icmit2021.010
Corresponding Author
Xinyue Ma
ABSTRACT
Removing the salt and pepper noise of an image is an essential step in digital image processing. In order to improve the effect of image denoising and obtain higher quality denoised images, this paper proposes a mean filter with adaptive window size. The filter can determine the appropriate window size according to the actual situation and restore the noise point's pixel value through the mean filtering method. Meanwhile, the filter uses a threshold comparison method to distinguish texture points and noise points more accurately. By testing on the data set, the superiority of the denoising method is verified. Furthermore, the value range of each parameter in the proposed method is determined through the experimental analysis.
KEYWORDS
A mean filter, image denoisingsalt, and pepper