Using Sample Mean and Sample Mean Absolute Difference for Noise Detection and Reduction

DOI: 10.5176/2251-1938_ORS17.14

Authors: Mathee Pongkitwitoon and Watcharin Klongdee

Abstract:

Noise detection and reduction techniques are an important methodology in the image processing. This research uses one of the simple statistical methods to detect and reduce noisy pixels in the specified images. We introduce the approach based on sample mean and sample mean absolute difference that is easy to calculate for the research purpose. Although it is a simple method, the results show that this approach method can simply detect noises, both salt and pepper and general noises, and reduce the noisy pixels that are better quality than the median filter.

Keywords: Noise Detection, Noise Restoration, Sample Mean Absolute Difference