Morphology-Based Shape Adaptive Compression

In this paper, we use morphology to improve the performance of the object oriented coding scheme. The object oriented compression algorithm is popular in recent years. However, due to the high frequency components of the edges, the reconstructed image is always seriously disturbed by the object edges (i.e., the ring effect) when using the algorithm. In this paper, we use the morphological operation to separate an object into the interior and the contour sub-regions before applying the shape-adaptive DCT. Since the contour subregion is separated apart, the high frequency components in it will not disturb the interior region and the reconstructed image will have much better quality. Simulation results show that the proposed algorithm outperforms the conventional object oriented compression algorithm.