Fractal Image Compression Using Dynamically Pipelined GPU Clusters

摘要: The main advantage of image compression is the rapid transmission data. conventional techniques exploit redundancy in images that can be encoded. idea to remove redundancies when stored and replace it back reconstructed. But ratio this technique quite insignificant, hence not a suitable candidate for an efficient encoding technique. Other methods involve removing high frequency Fourier coefficients retaining low ones. This method uses discrete cosine transforms(DCT) used extensively different flavors pertaining JPEG standards. Fractal provides resolution-independent based on contractive function concept. concept implemented using attractors (seed) are encoded/copied affine transformations plane. transformation allows operations such as, skew, rotate, scale, translate input which turn extremely difficult or impossible perform without having problem pixelization. Further, while decoding fractal image, there exist no natural size, thus decoded scaled any output size losing detail. A few years was purely mathematical but with availability cheap computing power like graphical processor units (GPUs) from Nvidia Corporation its realization now possible graphically. MatLab programming interface runs GPU clusters. GPUs consist many cores together give very speed over 24 GFLOPS. have varied usage satellite surveillance reconnaissance, medical imaging, meteorology, oceanography, flight simulators, extra-terrestrial planets terrain mapping, aircraft body frame design testing, film, gaming animation media, besides other allied areas.