IMAGE COMPRESSION BY MOMENT-PRESERVING EDGE-DETECTION

Abstract

A new image compression algorithm based on a new edge detection technique is proposed. A given image is compressed by dividing the image into non-overlapping square blocks and coding the edge information in each non-uniform block. The edge feature in a given window is detected by applying the moment-preserving principle to the image data. The edge directions are approximated by multiples of 45-degrees to speed up the direction process without introducing obvious distortion. For a larger block size, a more accurate model of a two-step edge is employed in the detection process. The proposed algorithm offers excellent reconstructed image quality agreeing with human perception, high compression ratios, and greatly reduced coding complexity. The solution to the edge detection problem in a given block is also analytic. The algorithm can be performed very fast for real-time applications with no need for special hardware.