Abstract
Multiresolution image decomposition based on linear filters has been widely used in image compression. In this research, we investigate a nonlinear multiresolution image decomposition based on mathematical morphology and obtain a morphological image pyramid. We propose to use opening and closing operations with a certain structure element to achieve image decomposition. The entropy and histogram of the difference images in the image pyramid are them examined. We use a numerical example to demonstrate that the morphological filtering approach provides some potential advantages over the conventional linear filtering approach in effective image coding.