Dempster-Shafer theory for image restoration
Ramesh K. Kulkarni, Saroj Kumar Meher, V. S. Lunge · 2010
This work proposes a new filter based on progressive decision using Dempster-Shafer theory, to suppress the impulse noise to preserve details of image and to restore image corrupted by random valued impulse noise. The new filter mechanism is composed of an efficient D-S impulse detector and a noise filter. The D-S evidence theory provides a way to deal with uncertainty in the evidence. The pixel detection and noise filtering are applied progressively through several iteration. If a pixel is noisy then the proposed filter will replace it with the central noise free ordered mean value otherwise is kept unchanged. The final image is further refined by using median filter which improves the image quality. Simulation results reveal that proposed algorithm outperforms other existing median filters for random noise.