| 摘要: |
| 为了改进分布式视频压缩感知方案的性能，提出了一种基于残差重构的分布式视频压
缩感知方案。该方案在编码端逐帧独立进行测量，在解码端依靠视频信号的时域相关性提升
重构信号质量。首先，对关键帧独立进行重构；其次，利用已重构关键帧做运动估计/运动
补偿以生成非关键帧的边信息；接下来，对边信息采用与编码端相同的测量矩阵进行测量并
计算测量残差值；最后，采用全变分最小化重构残差信号值并将其与边信息相加生成最终的
重构图像。实验结果表明，在相同采样率下，与已有的分布式视频压缩感知方案相比，提出
的方案可获得2.8 dB以上的峰值信噪比增益。 |
| 关键词: 压缩感知 分布式视频压缩感知 残差重构 全变分最小化 边信息 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目（61261023）；广西自然科学基金资助项目（2011GXNSFD018024）；广西教育厅科研项目资助（201203YB001） |
|
| Residual reconstruction based distributed compressed video sensing |
| CHANG Kan,QIN Tuan-fa,TANG Zhen-hua |
| () |
| Abstract: |
| To improve the performance of Distributed Compressed Video Sensing(DCVS), a resi
dual reconstruction based DCVS framework is proposed. The proposed fr
amework samples each video frame independently at the encoder. However, it recov
ers frames at the decoder by exploiting inter frame correlation. Firstly, the ke
y frame of a Group of Pictures(GOP) is independently reconstructed. Secondly, S
ide Information(SI) is generated by performing bidirectional Motion Estimatio
n(ME) and Motion Compensation(MC) through the reconstructed key frames. Afterwa
rds, the generated SI frame is sampled by the same matrix as the one at the enco
der, and the measurement of SI is used to calculate the residual of measurement.
Finally, total variation minimization is applied to reconstruct the residual si
gnal, and the output frame is formed by adding SI to the residual signal. Experi
mental results show that compared with the existing DCVS method, the proposed on
e can get more than 2.8 dB Peak Signal to Noise Ratio(PSNR) increment at th
e same sampling rate. |
| Key words: compressed sensing(CS) distributed compressed video sensing(DCVS) residual rec
onstruction total variation minimization side information(SI) |
《电讯技术》编辑部 