Peer Group Filtering Technique for Impulsive Noise Removal in Color Images

B. Smolka (Poland)


Impulse noise, color image processing, nonlinear filtering.


In this paper a new approach to the problem of impulsive noise removal in color images is presented. The proposed method is based on the evaluation of the statistical prop erties of a sorted sequence of the accumulated distances used for the calculation of the vector median of samples belonging to the filtering window. The statistical analysis is performed using the Fisher’s linear discriminant, which enables the detection of outliers introduced by the noise process. The proposed filter output switches between the vector median and the original central pixel. In order to increase the filter’s performance, two thresholds are intro duced, which enhance the detail preserving abilities of the proposed denposing method. The described filtering tech nique is robust, fast and therefore it can be used in real time denoising applications.

Important Links:

Go Back