Pyramidal predictive image coding with polynomial transforms
Boris Escalante Ramírez, Santiago Venegas-Martinez, Francisco Javier Garcia-Ugalde · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1995
We present a predictive image coding technique based on the polynomial transform. This is an image representation model that analyzes an image by locally expanding it into a weighted sum of orthogonal polynomials. The scheme proposed in this paper is a Laplacian-like pyramidal structure. Based on interpolation and deblurring algorithms implemented by means of the polynomial transform, our method improves the prediction of an image at a certain spatial scale from its representation at a lower scale, thereby reducing the entropy of the prediction error image, in comparison with the Laplacian pyramid and the scale-space image coding schemes.