期刊文献+

SIFT算法在CUDA加速下的实时人物识别与定位 被引量:11

SIFT Algorithm on the CUDA-accelerated Real-time Character Recognition and Positioning
在线阅读 下载PDF
导出
摘要 首先采用Directshow技术对视频图像中的运动目标进行差分检测,然后在此基础上对视频图像进行分割和识别,最后实现人体图像及头部图像的识别和定位。视频图像的处理和识别主要采用差分法和SIFT特征的模式匹配来完成人体和头部的识别和定位。经实验验证,SIFT算法在CUDA加速下具有较好的识别速度和准确率,解决了图像识别技术上的一些问题。 首先采用Directshow技术对视频图像中的运动目标进行差分检测,然后在此基础上对视频图像进行分割和识别,最后实现人体图像及头部图像的识别和定位。视频图像的处理和识别主要采用差分法和SIFT特征的模式匹配来完成人体和头部的识别和定位。经实验验证,SIFT算法在CUDA加速下具有较好的识别速度和准确率,解决了图像识别技术上的一些问题。
出处 《计算机科学》 CSCD 北大核心 2012年第S3期391-394,共4页 Computer Science
关键词 SIFT 模式匹配 图像识别 尺度 特征 SIFT Pattern matching Image recognition Scale Feature
  • 相关文献

参考文献14

  • 1Volkov V,Lu D J.QR and Cholesky factorizations using vectorcapabilities of GPUs. . 2008
  • 2Volkov V,Demmel J W.Benchmarking GPUs to tune dense li-near algebra. SC’’08:Proceedings of the 2008ACM/IEEEconference on Supercomputing . 2008
  • 3Bell N,Garland M.Efficient sparse matrix-vector multiplicationon CUDA. NVIDIA Technical Report NVR-2008-004 . 2008
  • 4Chen Jian,Smith R T,Tian Jie,et al.A novel registration me-thod for retinal images based on local features. . 2009
  • 5Jin Hyuk Jung,Dianne P O’’Leary.Cholesky decomposition and linear programming on a GPU. . 2006
  • 6C. Nvidia.Compute unified device architecture programming guide. . 2007
  • 7M. T. Do Dinh.GPUs-Graphics Processing Units. . 2008
  • 8Volkov V,Lu D J.QR and Cholesky factorizations using vectorcapabilities of GPUs. . 2008
  • 9Volkov V,Demmel J W.Benchmarking GPUs to tune dense li-near algebra. SC’’08:Proceedings of the 2008ACM/IEEEconference on Supercomputing . 2008
  • 10Bell N,Garland M.Efficient sparse matrix-vector multiplicationon CUDA. NVIDIA Technical Report NVR-2008-004 . 2008

同被引文献50

  • 1Chang Y L,Lin T L: Network- based H.264/AVC wholeframe loss visibility model and frame dropping methods[J].IEEE Transactions on Image Processing,2012,21(8):3353-3363.
  • 2CRISTANI M,RAGHAVENDRA R,BUE D A.Human behavior analysis in video surveillance:A Social Signal Processing perspective[J].Neurocomputing,2013,100(1):86-97.
  • 3Liu YY,Liu HP,et al. GPU accelerate the online k-meansclustering particle filter tracking alorithm [J].JournalofCentral South University supplement,2011,42(9):1-7.
  • 4Kim Z K. Real time object trcking based on dynamic fea-ture grouping with background subtraction [C]// Proceed-ings of the 2008 IEEE Conference on Com- pter Visionand Pattern Recognition, Anchorage,AK: IEEE,2008: 1-8.
  • 5Tsai H C, Wang W C, Wang J C,et al. Long distanceperson identification using height measurement and facerecognition[C]//Region 10 Conference TENCON. USAIEEE, 2009: 1-4.
  • 6Y Yi, B Abidi, N D Kalka, et al. High magnification andlong distance face recognition[C]//Biometrics Symposium:Special Session on Research at the Biometric ConsortiumConference, Baltimore, MD, 2006: 1-6.
  • 7Wang Qunming, Shi Wenzhong. Utilizing multiplesubpixel shifted images in subpixel mapping with imageinterpolation [J]. IEEE Geoscience and remote sensingletter (S1545-598X). 2014,11(4): 798-802.
  • 8Chai Zhenhua, Sun Zhenan, Heydi Vazquez, et al. Gaborordinal measures for face recognition [J]. IEEETransactions on information forensics and security(S1556-6013), 2014, 9(1): 14-26.
  • 9Shantanu Rane, Petros Boufounos. Privacy-preservingnearest neighbor methods[J].IEEE Signal ProcessingMagazine (S1053-5888),2013, 30(2): 20-29.
  • 10丁伟杰,周国民,任文华.基于小波分解和颜色熵的浮游生物图像识别[J].计算机工程,2008,34(14):188-190. 被引量:2

引证文献11

二级引证文献24

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部