Method and apparatus for compressing image data by mathematical transformation

摘要

The aim is to provide a method which is economical in "silicon area/processing speed" cost, for implementing a DCT type mathematical transformation on matrices of pixels, and which is appropriate to image compression. <??>According to the invention, the method consists essentially in combining a permutation (1) of each block of pixels with a set of one-dimensional RFTs (211, 213). A series of variants is obtained by appending or otherwise steps of grid computation (22) for reconstruction of a complex 2D-RFT, of rotations (3) for reconstruction of the DCT, and/or of suppression of the imaginary coefficients.