In the frame of compressed sensing distributed video coding, the design of the quantization matrix directly affects the reconstruction quality of the receiving terminal of the video. In this article, we present a new ...In the frame of compressed sensing distributed video coding, the design of the quantization matrix directly affects the reconstruction quality of the receiving terminal of the video. In this article, we present a new design method of the Gaussian quantization matrix adapting to the compressed sensing coding, for that the distribution of the parameters of the image is featured of the characteristic of approximately normal distribution after measured by compressive sensing. By this way, the parameters of a certain quantity of the image frames depending on the video sequences generated by the Gaussian quantization matrix possess certain adaptive capacity. By comparison with the plan of the traditional quantization, the quantization matrix presented in this article would improve the reconstruction quality of the video.展开更多
In this paper,a video compressed sensing reconstruction algorithm based on multidimensional reference frames is proposed using the sparse characteristics of video signals in different sparse representation domains.Fir...In this paper,a video compressed sensing reconstruction algorithm based on multidimensional reference frames is proposed using the sparse characteristics of video signals in different sparse representation domains.First,the overall structure of the proposed video compressed sensing algorithm is introduced in this paper.The paper adopts a multi-reference frame bidirectional prediction hypothesis optimization algorithm.Then,the paper proposes a reconstruction method for CS frames at the re-decoding end.In addition to using key frames of each GOP reconstructed in the time domain as reference frames for reconstructing CS frames,half-pixel reference frames and scaled reference frames in the pixel domain are also used as CS frames.Reference frames of CS frames are used to obtain higher quality assumptions.Themethod of obtaining reference frames in the pixel domain is also discussed in detail in this paper.Finally,the reconstruction algorithm proposed in this paper is compared with video compression algorithms in the literature that have better reconstruction results.Experiments show that the algorithm has better performance than the best multi-reference frame video compression sensing algorithm and can effectively improve the quality of slowmotion video reconstruction.展开更多
为改善分布式压缩视频感知（distributed compressive video sensing，DCVS）系统的视频帧图像重构质量，以实时视频传输为应用场景，提出了一种基于双重稀疏模型的图像解码算法。解码端由相邻的已重构关键帧产生边信息（sideinformatio...为改善分布式压缩视频感知（distributed compressive video sensing，DCVS）系统的视频帧图像重构质量，以实时视频传输为应用场景，提出了一种基于双重稀疏模型的图像解码算法。解码端由相邻的已重构关键帧产生边信息（sideinformation，SI）；根据双重稀疏模型思想，分离样本图像小波域下不同尺度的子带，分别使用K均值奇异值分解（K-means singular value decomposition，K—SVD）算法得到具有多尺度特性的冗余字典，结合梯度投影稀疏重建（gradient pursuit for sparsereconstruction，GPSR）算法，完成对非关键帧的重构。仿真结果表明，在相同压缩率下，相比传统K—SVD字典训练方法，本文所提出的方法对应的视频帧图像重构峰值信噪比（peak signal to noise ratio，PSNR）可获得0．5～1．5dB以上的增益。展开更多
Although compressive measurements save data storage and bandwidth usage, they are difficult to be used directly for target tracking and classification without pixel reconstruction. This is because the Gaussian random ...Although compressive measurements save data storage and bandwidth usage, they are difficult to be used directly for target tracking and classification without pixel reconstruction. This is because the Gaussian random matrix destroys the target location information in the original video frames. This paper summarizes our research effort on target tracking and classification directly in the compressive measurement domain. We focus on one particular type of compressive measurement using pixel subsampling. That is, original pixels in video frames are randomly subsampled. Even in such a special compressive sensing setting, conventional trackers do not work in a satisfactory manner. We propose a deep learning approach that integrates YOLO (You Only Look Once) and ResNet (residual network) for multiple target tracking and classification. YOLO is used for multiple target tracking and ResNet is for target classification. Extensive experiments using short wave infrared (SWIR), mid-wave infrared (MWIR), and long-wave infrared (LWIR) videos demonstrated the efficacy of the proposed approach even though the training data are very scarce.展开更多
针对煤矿井下视频监控图像因数据量大而导致传输和存储困难等问题,引入压缩感知理论对视频图像进行编解码,提出一种新的分布式视频编解码(distributed compressed video sensing, DCVS)方案.为了获得更稀疏的表示和更普遍的适用性,提出...针对煤矿井下视频监控图像因数据量大而导致传输和存储困难等问题,引入压缩感知理论对视频图像进行编解码,提出一种新的分布式视频编解码(distributed compressed video sensing, DCVS)方案.为了获得更稀疏的表示和更普遍的适用性,提出一种基于块的自适应混合稀疏基方案.针对边信息获取过程中通常采用固定权值合成边信息而忽略不同图像块之间相关性的问题,提出一种块分类加权边信息生成方案.结果表明:与传统的视频编解码方案相比,基于块的分类编解码方案能充分利用帧间相关性,在不同采样率下视频重构的峰值信噪比均有所提高,视频帧重构的质量也得到有效提升.展开更多
现有优秀的基于深度学习的分布式视频压缩感知(Distributed Compressed Video Sensing,DCVS)重构算法利用测量值和参考帧顺序更新非关键帧,获得了较好的重构性能,但由于缺乏较严格的理论指导,无法充分结合这两类信息,限制了非关键帧重...现有优秀的基于深度学习的分布式视频压缩感知(Distributed Compressed Video Sensing,DCVS)重构算法利用测量值和参考帧顺序更新非关键帧,获得了较好的重构性能,但由于缺乏较严格的理论指导,无法充分结合这两类信息,限制了非关键帧重构质量的进一步提升.针对该问题,本文首先利用贝叶斯理论及最大后验概率(Maximum A Posteriori,MAP)估计推导出DCVS中非关键帧重构的优化方程,再基于近端梯度算法推导出优化方程的求解框架,包含多信息流梯度更新聚合方程.基于此,本文设计了多信息流梯度更新及聚合模块(Multi-Information flow Gradient update and Aggregation,MIGA),并构建了深度多信息流梯度更新与聚合网络(Deep Multi-Information flow Gradient update and Aggregation Network,DMIGAN)用于DCVS非关键帧重构.MIGA利用测量值与多参考帧对当前非关键帧进行并行梯度更新,再做信息交互融合,从而充分结合多种信息流更新重构帧.本文级联MIGA与去噪子网络用于模拟近端梯度算法的单次迭代,作为基础模块(phase),并通过级联多个phase构造深度重构网络DMIGAN,实现帧重构的深度优化过程.实验表明,DMIGAN与具代表性的传统迭代优化算法结构相似的帧间组稀疏表示重构算法(Structural SIMilarity based Inter-Frame Group Sparse Representation,SSIM-Inter F-GSR)相比,在低采样率与高采样率下性能分别提升了8.8 dB和7.36 dB;和具有代表性的深度学习重构算法VCSNet-2相比,在低采样率和高采样率下性能分别提升了7.09 dB和8.78 dB.展开更多
文摘In the frame of compressed sensing distributed video coding, the design of the quantization matrix directly affects the reconstruction quality of the receiving terminal of the video. In this article, we present a new design method of the Gaussian quantization matrix adapting to the compressed sensing coding, for that the distribution of the parameters of the image is featured of the characteristic of approximately normal distribution after measured by compressive sensing. By this way, the parameters of a certain quantity of the image frames depending on the video sequences generated by the Gaussian quantization matrix possess certain adaptive capacity. By comparison with the plan of the traditional quantization, the quantization matrix presented in this article would improve the reconstruction quality of the video.
文摘In this paper,a video compressed sensing reconstruction algorithm based on multidimensional reference frames is proposed using the sparse characteristics of video signals in different sparse representation domains.First,the overall structure of the proposed video compressed sensing algorithm is introduced in this paper.The paper adopts a multi-reference frame bidirectional prediction hypothesis optimization algorithm.Then,the paper proposes a reconstruction method for CS frames at the re-decoding end.In addition to using key frames of each GOP reconstructed in the time domain as reference frames for reconstructing CS frames,half-pixel reference frames and scaled reference frames in the pixel domain are also used as CS frames.Reference frames of CS frames are used to obtain higher quality assumptions.Themethod of obtaining reference frames in the pixel domain is also discussed in detail in this paper.Finally,the reconstruction algorithm proposed in this paper is compared with video compression algorithms in the literature that have better reconstruction results.Experiments show that the algorithm has better performance than the best multi-reference frame video compression sensing algorithm and can effectively improve the quality of slowmotion video reconstruction.
基金Supported by National Natural Science Foundation of China（61170147） Major Cooperation Project of Production and College in Fujian Province（2012H61010016） Natural Science Foundation of Fujian Province（2013J01234）
文摘为改善分布式压缩视频感知（distributed compressive video sensing，DCVS）系统的视频帧图像重构质量，以实时视频传输为应用场景，提出了一种基于双重稀疏模型的图像解码算法。解码端由相邻的已重构关键帧产生边信息（sideinformation，SI）；根据双重稀疏模型思想，分离样本图像小波域下不同尺度的子带，分别使用K均值奇异值分解（K-means singular value decomposition，K—SVD）算法得到具有多尺度特性的冗余字典，结合梯度投影稀疏重建（gradient pursuit for sparsereconstruction，GPSR）算法，完成对非关键帧的重构。仿真结果表明，在相同压缩率下，相比传统K—SVD字典训练方法，本文所提出的方法对应的视频帧图像重构峰值信噪比（peak signal to noise ratio，PSNR）可获得0．5～1．5dB以上的增益。
文摘Although compressive measurements save data storage and bandwidth usage, they are difficult to be used directly for target tracking and classification without pixel reconstruction. This is because the Gaussian random matrix destroys the target location information in the original video frames. This paper summarizes our research effort on target tracking and classification directly in the compressive measurement domain. We focus on one particular type of compressive measurement using pixel subsampling. That is, original pixels in video frames are randomly subsampled. Even in such a special compressive sensing setting, conventional trackers do not work in a satisfactory manner. We propose a deep learning approach that integrates YOLO (You Only Look Once) and ResNet (residual network) for multiple target tracking and classification. YOLO is used for multiple target tracking and ResNet is for target classification. Extensive experiments using short wave infrared (SWIR), mid-wave infrared (MWIR), and long-wave infrared (LWIR) videos demonstrated the efficacy of the proposed approach even though the training data are very scarce.
文摘针对煤矿井下视频监控图像因数据量大而导致传输和存储困难等问题,引入压缩感知理论对视频图像进行编解码,提出一种新的分布式视频编解码(distributed compressed video sensing, DCVS)方案.为了获得更稀疏的表示和更普遍的适用性,提出一种基于块的自适应混合稀疏基方案.针对边信息获取过程中通常采用固定权值合成边信息而忽略不同图像块之间相关性的问题,提出一种块分类加权边信息生成方案.结果表明:与传统的视频编解码方案相比,基于块的分类编解码方案能充分利用帧间相关性,在不同采样率下视频重构的峰值信噪比均有所提高,视频帧重构的质量也得到有效提升.
文摘现有优秀的基于深度学习的分布式视频压缩感知(Distributed Compressed Video Sensing,DCVS)重构算法利用测量值和参考帧顺序更新非关键帧,获得了较好的重构性能,但由于缺乏较严格的理论指导,无法充分结合这两类信息,限制了非关键帧重构质量的进一步提升.针对该问题,本文首先利用贝叶斯理论及最大后验概率(Maximum A Posteriori,MAP)估计推导出DCVS中非关键帧重构的优化方程,再基于近端梯度算法推导出优化方程的求解框架,包含多信息流梯度更新聚合方程.基于此,本文设计了多信息流梯度更新及聚合模块(Multi-Information flow Gradient update and Aggregation,MIGA),并构建了深度多信息流梯度更新与聚合网络(Deep Multi-Information flow Gradient update and Aggregation Network,DMIGAN)用于DCVS非关键帧重构.MIGA利用测量值与多参考帧对当前非关键帧进行并行梯度更新,再做信息交互融合,从而充分结合多种信息流更新重构帧.本文级联MIGA与去噪子网络用于模拟近端梯度算法的单次迭代,作为基础模块(phase),并通过级联多个phase构造深度重构网络DMIGAN,实现帧重构的深度优化过程.实验表明,DMIGAN与具代表性的传统迭代优化算法结构相似的帧间组稀疏表示重构算法(Structural SIMilarity based Inter-Frame Group Sparse Representation,SSIM-Inter F-GSR)相比,在低采样率与高采样率下性能分别提升了8.8 dB和7.36 dB;和具有代表性的深度学习重构算法VCSNet-2相比,在低采样率和高采样率下性能分别提升了7.09 dB和8.78 dB.
