X-ray digital image improvement by maximum entropy

Maximum entropy (ME) as prior probability for positive, additive distributions in Bayesian statistics has been investigated in relation with digital image improvement. An iterative algorithm has been developed to minimise a specific function associated to an image with Gaussian noise. Scanned X-ray mammographs have been subject of comparative analyses and have proved to be of a superior quality as compared with their input counterparts in terms of detail contrast and visibility.