采用压缩传感的鲁棒的视频指纹方案
|
摘要
视频指纹技术在视频检索、识别、安全等领域有着广泛的应用,提出一种基于压缩传感理论的鲁棒性视频指纹方法,该方法采用压缩传感的稀疏性和安全性对提取的视频关键帧进行采样,再对采样矩阵分块与分类,提取能量值大的一些子块构成新的特征矩阵。对特征矩阵使用奇异值分解,对较大奇异值量化编码生成指纹。同时,也提出了高效的两步匹配方案,通过粗精两步搜索对视频进行检索,提高了视频搜索速度,实验结果表明,能准确检测视频片段,对通常的视频处理具有较强鲁棒性,满足视频检索的实时要求。
Video fingerprinting techniques have many applications in video retrieval,identification,and security.A robust video fingerprinting based on compressed sensing is proposed.In video fingerprinting,video key clips extracted are sampled by using the sparse and safety of compressed sensing.Each matrix sample is made into blocks,from which several big energy blocks are made into a new feature matrix.The singular value is used as code for the fingerprinting by SVD from this new feature matrix.Furthermore,an efficient two step match algorithm is proposed to using a search and match approximation,which improves video searching speed.The experimental results show that the proposed video fingerprinting is accurate in identifying different video clips,robust against common video processing,and can retrieve videos in real-time.
Video fingerprinting techniques have many applications in video retrieval,identification,and security.A robust video fingerprinting based on compressed sensing is proposed.In video fingerprinting,video key clips extracted are sampled by using the sparse and safety of compressed sensing.Each matrix sample is made into blocks,from which several big energy blocks are made into a new feature matrix.The singular value is used as code for the fingerprinting by SVD from this new feature matrix.Furthermore,an efficient two step match algorithm is proposed to using a search and match approximation,which improves video searching speed.The experimental results show that the proposed video fingerprinting is accurate in identifying different video clips,robust against common video processing,and can retrieve videos in real-time.
引文
[1]Liu Y,Yao L.Research of robust video fingerprinting[C]//Proceedings of International Conference on Computer Applicationand System Modeling(ICCASM).Taiyuan,China:IEEE,2010:43-46.
[2]Zhuang X Y.A video retrieval system based on video fingerprints[D].Beijing:Beijing University of Technology,2005.[庄新月.基于视频指纹的视频检索技术研究[D].北京:北京工业大学.2005.]
[3]Sunil L,Chang D Y.Robust video fingerprinting for content-based video identification[J].IEEE Transactions on Circuits andSystems for Video Technology,2008,18(7):983-988.
[4]Cheung C S,Zakhor A.Efficient video similarity measurementwith video signature[J].IEEE Trans.on Circuits Syst.VideoTechnol,2003,13(1):59-74.
[5]Bhat N D,Nayar K S.Ordinal measures for image correspon-dence[J].IEEE Trans.on Pattern Anal.Mach.Intell.,1998,20(4):415-429.
[6]Kim C,Vasudev B.Spatiotemporal sequence matching for effi-cient video copy detection[J].IEEE Trans.on Circuits Syst.Video Technol,2005,15(1):127-132.
[7]Fridrich J,Goljan M.Robust hash functions for digital water-marking[C]//Proceedings of IEEE International Conference onInformation Technology:Coding and Computing.Las Vegas,Ne-vada,USA:IEEE,2000:178-183.
[8]Mihcak M K,Venkatesan R.New iterative geometric techniquefor robust image hashing[C]//Proceedings of ACM workshop onSecurity and Privacy in Digital Rights Management.Vancover,BC,Canada:ACM,2001:13-21.
[9]Swaminathan A,Mao Y,Wu M.Robust and secure imagehashing[J].IEEE Trans.on Information Forensics and Security,2006,1(2):215-230.
[10]Lee S,Yoo D C.Robust video fingerprinting for content-basedvideo identification[J].IEEE Trans.on Circuits Syst.VideoTechnol,2008,18(7):983-988.
[11]Law-To L,Chen L,Joly A,et al.Video copy detection:a com-parative study[C]//Proc.of Conf.Image Video Retrieval(CIVR).Amsterdam,Netherlands:IEEE,2007:371-378.
[12]Esmaeili M M,Fatourechi M,Ward R K.A Robust and fast videocopy detection system using content-based fingerprinting[J].IEEE Trans.on Information forensics and Security,2011,6(1):213-226.
[13]Shao W Z,Wei Z H.Advances and perspectives on compressedsensing theory[J].Journal of Image and Draphics,2012,17(1):1-12.[邵文哲,韦志辉.压缩感知基本理论:回顾与展望[J].中国图象图形学报,2012,17(1):1-12.]
[14]Donoho D L.Compressed sensing[J].IEEE Transactions on In-formation Theory,2006,52(4):1289-1306.
[15]Wang K,Liu Y L,Zhang J X.RIP analysis for quasi-Toeplitz CSmatrices[C]//Proceedings of International Conference on FutureInformation Technology and Management Engineering.Chongqing,China:IEEE,2010:1512-1514.
[16]Duarte M F,Wakin M B,Baraniuk R G.Fast reconstruction ofpiecewise smooth signals from random projections[C]//Proc.ofSignal Processing with Adaptive Sparse Sparse Structure Repre-sentation.Rennes,France:[s.n.],2005.
[17]Liu G F,Debrunner V.Matching pursuits may yield superior re-sults to orthogonal matching pursuits when secondary informationis estimated from the signal model[C]//Proceedings of 2010Conference Record of the Forty Fourth Asilomar Conference onSignals,Systems and Computers.Califomia,USA:IEEE,2010:2013-2016.
[18]Cormode G,Muthukrishnan S.Towards an algorithmic theory ofcompressed sensing,DIMACS TR:2005-40[R].Newark,NewJersey,USA:Rutgers University,2005.
[19]Sun Z H,Wang C,Kebin J.Near-duplicate video clips detectionwith motion based video fingerprinting[C]//Proceedings of the4th International Congress on Image and Signal Processing.Bei-jing,China:IEEE,2011:456-459.
[20]Wang R,Ping X J.Detection of resampling based on texturecomplexity and singular value decomposition[J].Journal of Com-puter-Aided Design and Computer Graphics,2010,22(9):1606-1612.[汪然,平西建.基于图像纹理复杂度和奇异值分解的重采样检测[J].计算机辅助设计与图形学学报,2010,22(9):1606-1612.]
[21]Su X,Huang T J,Gao W.Robust video fingerprinting based onvisual attention regions[C]//Proceedings of ICASSP 2009.Bei-jing,China:IEEE,2009:1525-1528.
[2]Zhuang X Y.A video retrieval system based on video fingerprints[D].Beijing:Beijing University of Technology,2005.[庄新月.基于视频指纹的视频检索技术研究[D].北京:北京工业大学.2005.]
[3]Sunil L,Chang D Y.Robust video fingerprinting for content-based video identification[J].IEEE Transactions on Circuits andSystems for Video Technology,2008,18(7):983-988.
[4]Cheung C S,Zakhor A.Efficient video similarity measurementwith video signature[J].IEEE Trans.on Circuits Syst.VideoTechnol,2003,13(1):59-74.
[5]Bhat N D,Nayar K S.Ordinal measures for image correspon-dence[J].IEEE Trans.on Pattern Anal.Mach.Intell.,1998,20(4):415-429.
[6]Kim C,Vasudev B.Spatiotemporal sequence matching for effi-cient video copy detection[J].IEEE Trans.on Circuits Syst.Video Technol,2005,15(1):127-132.
[7]Fridrich J,Goljan M.Robust hash functions for digital water-marking[C]//Proceedings of IEEE International Conference onInformation Technology:Coding and Computing.Las Vegas,Ne-vada,USA:IEEE,2000:178-183.
[8]Mihcak M K,Venkatesan R.New iterative geometric techniquefor robust image hashing[C]//Proceedings of ACM workshop onSecurity and Privacy in Digital Rights Management.Vancover,BC,Canada:ACM,2001:13-21.
[9]Swaminathan A,Mao Y,Wu M.Robust and secure imagehashing[J].IEEE Trans.on Information Forensics and Security,2006,1(2):215-230.
[10]Lee S,Yoo D C.Robust video fingerprinting for content-basedvideo identification[J].IEEE Trans.on Circuits Syst.VideoTechnol,2008,18(7):983-988.
[11]Law-To L,Chen L,Joly A,et al.Video copy detection:a com-parative study[C]//Proc.of Conf.Image Video Retrieval(CIVR).Amsterdam,Netherlands:IEEE,2007:371-378.
[12]Esmaeili M M,Fatourechi M,Ward R K.A Robust and fast videocopy detection system using content-based fingerprinting[J].IEEE Trans.on Information forensics and Security,2011,6(1):213-226.
[13]Shao W Z,Wei Z H.Advances and perspectives on compressedsensing theory[J].Journal of Image and Draphics,2012,17(1):1-12.[邵文哲,韦志辉.压缩感知基本理论:回顾与展望[J].中国图象图形学报,2012,17(1):1-12.]
[14]Donoho D L.Compressed sensing[J].IEEE Transactions on In-formation Theory,2006,52(4):1289-1306.
[15]Wang K,Liu Y L,Zhang J X.RIP analysis for quasi-Toeplitz CSmatrices[C]//Proceedings of International Conference on FutureInformation Technology and Management Engineering.Chongqing,China:IEEE,2010:1512-1514.
[16]Duarte M F,Wakin M B,Baraniuk R G.Fast reconstruction ofpiecewise smooth signals from random projections[C]//Proc.ofSignal Processing with Adaptive Sparse Sparse Structure Repre-sentation.Rennes,France:[s.n.],2005.
[17]Liu G F,Debrunner V.Matching pursuits may yield superior re-sults to orthogonal matching pursuits when secondary informationis estimated from the signal model[C]//Proceedings of 2010Conference Record of the Forty Fourth Asilomar Conference onSignals,Systems and Computers.Califomia,USA:IEEE,2010:2013-2016.
[18]Cormode G,Muthukrishnan S.Towards an algorithmic theory ofcompressed sensing,DIMACS TR:2005-40[R].Newark,NewJersey,USA:Rutgers University,2005.
[19]Sun Z H,Wang C,Kebin J.Near-duplicate video clips detectionwith motion based video fingerprinting[C]//Proceedings of the4th International Congress on Image and Signal Processing.Bei-jing,China:IEEE,2011:456-459.
[20]Wang R,Ping X J.Detection of resampling based on texturecomplexity and singular value decomposition[J].Journal of Com-puter-Aided Design and Computer Graphics,2010,22(9):1606-1612.[汪然,平西建.基于图像纹理复杂度和奇异值分解的重采样检测[J].计算机辅助设计与图形学学报,2010,22(9):1606-1612.]
[21]Su X,Huang T J,Gao W.Robust video fingerprinting based onvisual attention regions[C]//Proceedings of ICASSP 2009.Bei-jing,China:IEEE,2009:1525-1528.