Local adaptive contrast enchancement in digital images
George Apostolopoulos, Evangelos Dermatas · International Conference on Mobile Multimedia Communications · 2007
In this paper an algorithm for optical quality enhancement in low contrast images is proposed. The brightness dynamic region is increased using a non-linear transformation function, which is adapted from the local contrast. In a new approach, the local contrast is defined in every central pixel, using a four direction masking method and a statistical estimation of the local contrast value. Two adaptive thresholds locate the exact pixel positions where the proposed contrast adjustment algorithm is used to improve the image quality. In the evaluation process, several artificially distorted images are enhanced and the mean square error and peak noise ratio between the restored and the original images are estimated. The experimental results show both objective and subjective improvements in the image quality.