Images compression using moments method of orthogonal polynomials

Image compression is one absolute necessity for applications such as transmission in databases. In this paper, we propose a scheme for image compression. We use a moments method in order to obtain a set of orthogonal moments of the image and from these moments we obtain the reconstructed version of the image. The properties of this method are examined in detail, with both theoretical and experimental results presented. The maximal entropy principle is used to determine at which moments order we have to stop. Finally, in order to allow the receiver to recognise a picture as quickly as possible, and at minimum cost, we present a progressive transmission scheme. It is shown that the moments method is particularly well adapted to progressive transmission.