Method of data reduction by means of a fractal image coding and encoder and decoder for performing the method

摘要

In a method of data reduction of a luminance and/or chrominance signal of a digital picture signal by means of fractal image coding, in which method each image of the luminance/chrominance signal is divided into range blocks of nxn pixels each, in which a domain block is searched for each range block, which domain block is imageable on the range block with a minimal deviation while using a transformation function, which domain blocks have twice the size of the range blocks, and in which information on the transformation functions is transmitted, from which information the image data are regained at the receiver end in an iterative process, the search for appropriate domain blocks is facilitated in that the data of each range block are examined on geometrical basis functions which are possibly satisfactory approximations of the data structure of the range block, in that a difference value is computed from the basis function found for the best approximation of the data structure of the range block and the data of the range block, which difference value represents a residual roughness, in that the data of the domain blocks are examined on geometrical basis functions which are possibly satisfactory approximations of the data structure of the domain block, in that a difference value is computed from the basis function found for the best approximation of the data structure of the domain block and the data of the domain block, which difference value represents a residual roughness, in that the search for the domain block which is imageable on a range block with a minimal deviation, and the associated transformation function is performed with reference to the computed difference values of the range block and the computed difference values of the domain blocks, and in that information on the mutually balanced basis functions of this block and of the most favorable domain block, as well as the transformation function, by means of which the most favorable domain block is imageable on the range block, is transmitted for each range block as a data-reduced data current instead of the luminance/chrominance signal.