Research On Compressed Video Sensing At Low Sampling Rate

Compressed sensing (CS) breaks the limits of Nyquist sampling rate and achievesa direct sampling for information. Compressed video sensing (CVS) introduces CS intovideo encoding and thus realizes a low-complexity method of encoding.The existed compressed video sensing schemes are mostly of great decodingquality at high sampling rate. However, in wireless networks whose bottom node withlimited energy, enhancing sampling rate can increase the consumption of energy,leading to reducing the lifetime of the network. In this paper, a novel compressed videosensing system which is exactly suitable for wireless multimedia sensor networks isproposed by combining compressed sensing and distributed video coding. This systemimproves the performance at low sampling rate by enhancing the decoding quality ofboth Key and non-key frames. At the decoder, the amount of measurement is increasedby resampling and thus improving the reconstruction quality of non-key frames. By thethought of using side information for assisted decoding to enhance the decodingperformance of non-key frames, the quality of Key frames is enhanced byre-reconstruction. Simulation results show that the system has a high-quality decodingperformance, especially in the case of low sampling rate. Both the quality of key framesand that of non-key frames can be effectively improved.