The Efficient Video Codec Model Based On Compressed Sensing

Compressed sensing is a new coding technique that sampled the signal simultaneously compressed, greatly reducing the sampling rate, breaking the Nyquist sampling theorem, the sampling rate must be greater than twice the highest frequency signal limitations. Compressed sensing with a low sampling rate, low computational complexity at coding end, low hardware requirements and high reconstruction quality and other characteristics, with the demand for ultra-wideband wireless sensor network of video communication, It will show its superiority in future wireless video art.For the defects and shortcomings of common reconstruction algorithms, combined with smoothing 0 norm compressed sensing reconstruction algorithm SL0 algorithm,by improving iterative search algorithm, iterative search method and the loop termination conditions, a fast and efficient algorithm is put forward. The algorithm can improve the quality and reduce the reconstruction consuming, for all kinds of video sequences with universal adaptability.By analyzing the existing distributed video coding model and insufficient distributed compressed sensing video coding model, we propose a new distributed video codec compressed sensing model. The coding end using differential coding model, and the decoding end using the joint decoding model. Model reduces the maximum degree of computational complexity of encoding end and power, a large amount of storage is transferred to the arithmetic decoder, the proposed model is an efficient codec capability and strong anti-interference ability of the channel.Finally, in order to solve the compressed sensing inter and intra-frame redundancy, to improve the stability and robustness of the system, an adaptive compression perceptual video coding model is put forward. By sampling rate controller and adaptive calibration module two subsystems, effectively improves video coding efficiency, reducing the bit error rate generated in the process. At the same time presents a channel estimation method based on compressed sensing, experiments demonstrate the effectiveness of the proposed method.