Automatic enhancement of noisy images using evaluation of image quality

Abstract

A new method for automatic enhancement of noisy images is presented. In the proposed approach the smoothing and sharpening effects provided by nonlinear functions are controlled by resorting to a measure of image quality that takes into account the histograms of the edge gradients. Thus, a well balanced compromise between smoothing and sharpening can automatically be found and applied to the image data. Experimental results show that the performance of the automatic enhancement system is very satisfactory.