Greens Function as a Transform for Image Compression

V. Sitnikov (Finland)

Keywords

Image Processing, Image Manipulation, Image Compression, Image Synthesis

Abstract

Digital visual data has step and pulse-like structure. Common compression methods mostly based on Fourier Transform cannot compress step and pulse-like data very well due to its infinite spectrum. This article presents a method, which is naturally not affected by this issue. The paper summarizes research results on a brand new approach to image representation and compression using methods of mathematical physics. Images are considered as solutions of differential equations. Data corresponding to image contours is taken to reconstruct the initial image by means of Green's Function utilized as a transform. The paper proposes a full transform and compression decompression schemes for the introduced technique.

Important Links:



Go Back