S. V. Kuleshov

The digital video stream transmission is a resource costing task which cannot be solved without compression of data being transmitted. Herewith the increase in visual quality results in corresponding increase in demands for transmission channel bandwidth. The task of compression, transmision and backwards decompression is being solved by coders and decoders (codecs) which have software or hardware realization. Multistage process of compression allows as to independently choose the algorithm for every single stage but demands cross-stage consistency of parameters to achieve most efficient result in codec work. To achieve the normalization and secondary compression stages consistency estimation, the criterion reflecting integral energy efficiency of codec is proposed. This criterion is based on the assumption that the most efficient normalizing basis contains elements which minimal complexity in Kolmogorov's sense is fully corresponds to elements of real image of given class. The usage of proposed estimations allows optimization and simultaneous normalization basis selection which is most efficient with the given methods of secondary compression in development of image and video data compression algorithms.