An Iterative Bilateral Weighted Median filter for the removal of High-Density impulse noise

Abstract

For removing the impulse noise with better edge and fine detail preservation, we propose an iterative bilateral weighted median (IBWM) filter. Instead of the specified weight of WM filters, we propose a novel weight fuction, bilateral weight (BW) where is the combination of domain weight (DW) in spatial domain and range weight (RW) in intensity domain. The DW is designed to give higher weight to pixels that are spatially close to a center pixel. And the RW is designed to give higher weight to pixels that are similar to the mean value of pixels in the filtering window. Finally, the rates of DW and RW are controlled according to noise distributions in a filtering window. For ensuring high accuracy of noise detection, the IBWM filter uses iterative process using increasing filtering window size. Simulation results indicate that our algorithm provides a significant improvement than any other existing techniques.