Mixed Noise Removal Method Based on Robust Switching Vector Median Filter and ε-Filter

Abstract

In this report, a new noise removal method in color images is presented. In the conventional noise removal methods for color images, the monochrome image processing techniques have been often performed for each color component separately. However, the component-wise noise removal does not give desirable results because the output values may be with color distortion. Therefore, it is desirable to be processed by the vector-type filters in general. In the proposed method, firstly, a tentative output image, in which the noise is removed almost perfectly, is obtained by using the iterative robust switching vector median-based vector $\varepsilon$-filter. Then the residual components between the input and the tentative output images are calculated, and image components consisting of edges, corner and other image details are extracted from the residual components by using the correlation characteristic in RGB components. Consequently, a final output is obtained by adding the extracted image components into the tentative output image. The effectiveness of the proposed method is illustrated through some experiments.