ADAPTIVE COMPRESSION CODING

A compression technique which preserves edges in compressed pictures is developed. It is desirable to build edge preservation characteristics in compression methods since many applications in engineering and vision depend on edge information. In this paper we present a compression algorithm which adapts itself to the local nature of the image. Smooth regions are represented by their averages and edges are preserved using quad trees. Textured regions are encoded using BTC (block truncation coding) and a modification of BTC using look-up tables. We developed the latter (BTC with look-up tables) so as 1) to improve on the compression ratio of BTC and 2) to leave the visual quality of compression exactly the same as that of BTC. A threshold using a range which is the difference between the maximum and the minimum grey levels in a 4 x 4 pixel quadrant is used. At the recommended value of the threshold (equal to 18), the quality of the compressed textured regions is very high, the same as that of AMBTC (absolute moment block truncation coding) but the edge preservation quality is far superior to that of AMBTC. This significant improvement is achieved at compression levels (1.1-1.2 b/pixel) which are better than that of AMBTC (1.63). Compression levels below (0.5-0.8) b/pixel may be achieved. The high quality of edge preservation of this method does not change at these low compression levels or as the threshold value changes. However, a postfilter is needed to improve the blocky appearance in the smooth and textured regions.