Patlaenko M. Increasing the compression ratio of TV images by recurrent transformations and wave algorithms

Object of research: the process of compression of TV images and video sequences of high and ultra-high definition in television systems. The subject of research - methods of spectral transformations of high and ultra-high definition television images. Research methods: spectral transformations, wavelet analysis, interpolation methods, affine transformations, recurrent transformations, methods of mathematical analysis and modeling in SciLab, MATLAB, optimization methods, approximation methods, probability theory, methods of mathematical interpretation of scientific research. The main scientific result of the dissertation is the first proposed complex method of image compression. Within the main scientific result the following is obtained: for the first time the method of adaptive threshold processing of signal contours is proposed, for the first time the method of adaptive frequency-dependent quantization of wavelet decomposition levels is proposed, the method of prediction of wavelet decomposition orders is proposed for the first time. parametric method of fractal compression by splitting the image into rank and domain blocks.