期刊文献+

基于CUDA加速的目标检测在桥梁中的应用

下载PDF
导出
摘要 为提高高速公路桥梁安全智能监控水平,该文以锡张高速公路全程监控系统构建为基础,对GMM背景建模提出了参数递推更新、阴影检测和每个像素高斯模型个数自适应选择的策略针对复杂条件下背景建模这一视频检测难点,该文利用常数内存来存储所有只读常量参数来解决存访速度;利用内存优化来对内存数据按段对齐、存储器合并访问、按SoA排序,以解决存访合并和IO延迟隐藏;对在CPU端和GPU端利用页锁定内存和CUDA流来实现代码异步并行,以从根本上来提高数据访问效率、降低算法复杂度。经测试结果表明,相比CPU实现,GPU方式能够明显提高算法的实时性能。
作者 王昶 张强
出处 《中国交通信息化》 2014年第10期137-142,共6页 China ITS Journal
  • 相关文献

参考文献10

  • 1Lipton A, FujJyoshi H, Patil P~ Moving target classification and tracking from reaI-time video[C].IEEE Workshop on Applications of Computer Vision, Princeton: IEEE Press, 1998= 8-14.
  • 2Collins Eetal. A system for video surveillance arid monitoring: VSAM final report[E]. Pittsburgh, USA= Oarnegie Mellon Universfty, 2000.
  • 3Arseneau S, Cooperstock J. Real-time image segmentation for action recognition[C].Proc. IEEE Pacific Rim Conyerence on Communications, Computers and 9ignal Processing, Canada= Victoria, 1999:86-89.
  • 4无锡至张家港高速公路监控系统竣工图(XZJDS001标)[R].北京北京公科飞达交通工程发展有限公司,2010.
  • 58tauffer C. Orimson W E k. Adaptive baekground mixture modeJs for realtime tracking[C].Oomputer Vision a, nd Pattern Eecognition. Co. USA. IEEE. 1999. 246-250.
  • 6Zivkovic. Improved adaptive Gausian mixture model for background subtraction, In Proceedings. lOPE, 2004.
  • 7姚会,苏松志,王丽,李绍滋.基于改进的混合高斯模型的运动目标检测方法[J].厦门大学学报(自然科学版),2008,47(4):505-510. 被引量:22
  • 8Harish P, Narayanan P J. Accelorating large graph algrithrnsOll the GPU using CUIA[J]. Computer Science, 2007(4875): 197-208.
  • 9NVIDIA Corporation.NVlDIA CUDA C Programming Guide Version 4.2[S].2012.
  • 10Janaka Liyanage, GMM based Background Subtraction on GPU[R]. CDA 6958: Project TechnicaJ Ieport. 2008.

二级参考文献13

  • 1代科学,李国辉,涂丹,袁见.监控视频运动目标检测减背景技术的研究现状和展望[J].中国图象图形学报,2006,11(7):919-927. 被引量:169
  • 2Wren C R, Azarbaye J A, Darrell T P. Real-time tracking of the human body [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997,19(7):780-785.
  • 3Stauffer C, Grimson W E L. Adaptive background mixture models for real-time tracking [C]//Computer Vision and Pattern Recognition. CO, USA: IEEE, 1999:246-250.
  • 4Stauffer C, Grimson W E L. Learning patterns of activity using real-time tracking [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22 (8):747-757.
  • 5KaewTraKulPong P, Bowden R. An improved adaptive background mixture model for real-time tracking with shadow detection [C]//The 2nd European Workshop on Advanced Video-based Surveillance Systems. Kingston: Kluwer Academic Publishers, 2001 : --5.
  • 6Salvador E, Cavallaro A, Ebrahim T I. Cast shadow segmentation using invariant color features [J]. Computer Vision and Image Understanding, 2004, 95 (2):238 - 259.
  • 7Horprasert T, Harwood D, Davis L S. A statistical approach for real-time robust background subtraction and shadow detection [C]//IEEE ICCV 1999 Frame-rate Workshop. Greece : Internatioanl Conference on Computer Vision, 1999: 1 - 19 .
  • 8Cuccbiara R, Grana C, Piccardi M,et al. Improving shadow suppression in moving object detection with HSV information [C]//IEEE Conference on Intelligent Transportation Systems. Oakland, CA, USA: IEEE, 2001:334 - 339.
  • 9Prati A, Mikic I, Trivedi M M,et al. Detecting moving shadows: algorithms and evaluation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(7) :918-923.
  • 10Adams R, Bischof L. Seeded region growing [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994, 16(6):641-647.

共引文献21

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