期刊文献+

区域法立体匹配的实时性问题研究 被引量:4

PC based real-time implementation for regiond matching in stereo vision
下载PDF
导出
摘要 立体匹配的一个关键问题是计算量大,实时性难以保证.作者从两个方面着手对立体匹配算法进行优化.首先是在算法结构上,应用盒滤波技术消除了冗余计算;其次在实现上,针对IntelPentium4处理器的特性,从存储器组织和高速缓存的利用,SSE2多媒体扩展指令的应用以及基于超线程技术的并行处理等三个方面进行优化实现.实验结果表明,所提出的算法在实时性上有了很大的提高. The requirement of mass computation power was one of the main obstacles to the applications based on stereo matching. This paper tried to solve this problem in two ways. First, the computation redundancies were reduced using the box-filtering technique. Then further optimizations related to memory organization and cache efficiency, Streaming SIMD Instructions, and the parallel process using Hyper-threading technology have been efficiently implemented on Intel Pentium 4 based computer. The test result proves that the approach proposed in this paper can reduce the process time effectively.
出处 《浙江大学学报(理学版)》 CAS CSCD 2003年第6期628-631,共4页 Journal of Zhejiang University(Science Edition)
基金 国家自然科学基金资助项目(60105003) 浙江省自然科学基金资助项目(600025).
关键词 区域法立体匹配 实时性 视差 优化 IA架构 算法 盒滤波技术 IntelPentium4处理器 图像 stereo matching real-time disparity optimization intel architecture
  • 相关文献

参考文献8

  • 1HARIYAMA M, TAKEUCHI T, KAMEYAMA M.VLSI processor for reliable stereo matching based on adaptive window-size selection [A]. Proceedings ofIEEE International Conference on Robotics and Automation[C]. Korea: Seoul, 2001. 1168-1173.
  • 2KONOLIGE K. Small vision systems: hardware and implementation [A]. In 8th Int Symposium on Robotics Research[C]. Japan: Hayama, 1997.
  • 3CORKE P, DUNN P. Real-time stereopsis using FPGAs [J]. Proceedings of IEEE Region 10 Annual Conference Speech and Image Technologies for Computing and Telecommunications, 1997, (1) : 235- 238.
  • 4FAUGERASO, HOTZ B, MATHIEU H, et al.Stereo: algorithm, implementations and applications[EB/OL]. http://www.inria. fr/rrrt/rr-2013. html,1993.
  • 5MCDONNEL M J. Box-filtering techniques [ J ].Computer Graphics and Image Processing, 1981,(17):65-70.
  • 6RICHARD G. The Software Optimization Cookbook[M]. Norway: Intel Press,2002.
  • 7SATOSHI K, KEI O, MaASAYUKI I, et al. Design and implementation of onbody real-time depthmap generation system [A]. Proceedings of the IEEE International Conference on Robotics and Automation[C]. San Francisco: CA,2000. 1441-1445.
  • 8SEBOT J, NATHALIE D. Memory bandwidth: the true bottleneck of {SIMD} multimedia performance on a superscalar processor [J]. LNCS, 2001, 2150: 439-447.

同被引文献37

  • 1刘钢,彭群生,鲍虎军.基于图像建模技术研究综述与展望[J].计算机辅助设计与图形学学报,2005,17(1):18-27. 被引量:57
  • 2陈华华,杜歆,顾伟康.基于神经网络和遗传算法的机器人动态避障路径规划[J].传感技术学报,2004,17(4):551-555. 被引量:23
  • 3Haan G de, Bellers E B. Deinterlacing-An Overview[ C]//Proceedings of the IEEE, 1998,86 (9) : 1839 - 1857.
  • 4Ronald H Y, Chung Kwan-Yee K. Wong, etc. Generalized Motion and Edge Adaptive Interpolation De-interlacing Algorithm [ J ]. WSEAS Transactions on Computers, 2006 : 1 - 8.
  • 5Brox P, Sanchez-Solano S, Baturone I, et al. FPGA Implementation of a Fuzzy Based Video De-interlacing Algorithm[ C]//Proc. of SPIE, 2005, 5837 : 650 - 660.
  • 6TSAI Tsung-Han. Platform-Based Design for the Low Complexity and High Performance De-interlacing System [ C ]//IEICE Trans. Information and Systems, 2008, E91 -D( 13 ) :2784 - 2792.
  • 7NVIDIA. http ://cuda. csdn. net/[ EB/OL ].
  • 8Lin Shyh-Feng, Chang Yu-Lin, and Chen Liang-Gee. Motion Adaptive De-interlacing by Horizontal Motion Detection and En- hanced ELA Processing[ C]//IEEE International Symposium on Circuits and Synstems, 2003:696 - 699.
  • 9Hsiao H H, Jeng J I4. Modified De-interlacing Method Based on Edge Direction[ C ]//The 23rd Workshop on Combinatorial Mathematics and Computation Theory :225 - 231.
  • 10Shahinfard, E. Sid-Ahmed, M. A. Ahmadi, M. A Motion Adaptive Deinterlacing Method with Hierarchical Motion Detection Algorithm [ C ]//15th IEEE International Conference on Image Processing, 2008:889 -892.

引证文献4

二级引证文献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