Noise filtering for mammographic images

We have developed and implemented two locally adaptive image restoration filters to improve the signal to noise ratio of mammographic images. Previous efforts in restoration of digitized mammograms have assumed a stationary image with uncorrelated noise. In this work we considered the realistic case of a non-stationary image model and signal dependent noise of photonic and film-grain origins. The Baysian filter was found to outperform the adaptive Wiener filter. Both filters were implemented in real time as part of our mammographic image acquisition and analysis system.