A Novel Video Compression Algorithm Using Wavelet Transform and neural network

Videos are made up of a temporal sequence of frames and are projected at a proper rate to create the illusion of motion. This means that there exists a high correlation between adjacent temporal frames so that when projected at a proper rate, smooth motion is seen. Correlation between adjacent temporal frames is called interframe correlation. In order to decode compressed video bit stream uniformly by various platforms and devices, the bit stream format must be predefined. Thus, there must be a standard for a video compressor, which will enable all standard-compliant compressed video data to be decoded anywhere. The goal is to propose a new video compression algorithm based on wavelet transform and neural networks. Using wavelet transform leads to factorization in temporal as well as spatial domain. The goal in this paper is to achieve a compression algorithm which would be faster and has more compression ratio. Neural networks are used for prediction which is one of the most important functions in any video compression scheme. Furthermore, the proposed algorithm is compared with MPEG standard. Simulation results show the befits of using wavelet transform which reveal that the proposed algorithm is faster and has better performance in some aspects compared to MPEG standard. The video which obtained from proposed algorithm has acceptable in human visual and since it needs less than space for storing, it is suitable for portable devices.