Morphological Compensation for the Enhancement of Compressed Image

This paper describes a morphological compensation algorithm for the enhancement of grayscale image compression under the standard of JPEG and JPEG2000, which is more effective for remote sense image than for natural image. Morphological reconstruction is a nonlinear filter. When used for image enhancement it dilates some pixels while preserving the edge information. The algorithm has a close relationship with morphological reconstruction. When this compensation algorithm is applied to images, among which there are images compressed by standard JPEG2000 and JPEG, it can improve the quality of decompressed image.