AN OPTIMUM SWITCHING BILATERAL FILTER FOR NOISE REMOVAL

摘要

The switching bilateral filter (SBF) is one of the most effective filters that is proposed for the mixed noise removal in images. The performance of the SBF is controlled by two parameters, the values of the radiometric weight and the spatial weight. The proportional values of these parameters depend on three important factors which are level of noise, the type of noise and the type of image. In image communication, the image (or information) is always change. Therefore, in this paper, the optimum SBF (OSBF) was studied and the values of radiometric weight and spatial weight (radiometric weight _OP and spatial weight_OP) were optimized. The proportional value of radiometric weight_OP and spatial weight_OP can be applied for all image types and the high PSNR can be obtained. The proposed technique was compared with three conventional filters; Bilateral filter (BF), Median filter (MF) and switching bilateral filter (SBF). In order to confirm that OSBF is compatible with all types of images, two cases of the image including database and non-database are considered. The simulation results show that the proposed technique performs a good result than other techniques.