---

The need for higher compression combined with the shortcomings of block-oriented image compression techniques (such as conventional transform coding) have motivated the search for alternative compression method known as second generation image coding techniques. An image is the projection of 3-d scene onto a 2-d plane. Therefore, description of an image in terms of objects rather than individual pixels (or blocks of pixels which is the case in conventional coding methods) should result in a more compact description. Region-based approaches are also attractive because they allow graceful degradations in image quality at high compressions, whereas conventional transform-based approaches are limited by artifacts such as `blockiness'. Region-based approaches also lend themselves to post-processing tasks such as edge enhancement. In contrast, the conventional transform-based approaches do not allow such post-processing manipulations (due to blocking artifacts).