Image compression by morphological operators

Image compression techniques generally rely on the results of information theory. A decorrelation of the signal followed by quantization and entropy coding of the information to transmit achieves image compression. Mathematical morphology can be considered a shape-oriented approach to signal processing and some of its features are useful for compression. Classical linear signal processing tools are not well suited for a geometrical approach. Mathematical morphology has been developed as a geometrical approach to signal processing. This paper focuses on four morphological tools - connected operators, region-growing version of the watershed, the geodesic skeleton and a morphological interpolation technique, to be attractive for compression, and these tools cover the most important parts of a coding scheme.