Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (34): 231-233.DOI: 10.3778/j.issn.1002-8331.2008.34.070

• 工程与应用 • Previous Articles     Next Articles

Research on optimization methods for intelligent vision surveillance algorithm on DSP

LI Bin,LI Gong-yan,XU Shi-yi,TIAN Yuan   

  1. Research Center of Integrated Information System,Institute of Automation,CAS,Beijing 100080,China
  • Received:2007-12-17 Revised:2008-03-18 Online:2008-12-01 Published:2008-12-01
  • Contact: LI Bin

DSP体系结构下视觉监控优化方法研究

李 斌,李功燕,许世颐,田 原   

  1. 中国科学院 自动化研究所 综合信息系统中心,北京 100080
  • 通讯作者: 李 斌

Abstract: First,a methodology for migrating intelligent visual monitoring algorithms onto the TI DSP platform is discussed.Then,the methods for optimization,algorithms of optimization in C language and in assembly directive,and the technologies of optimization in memory configuration are illuminated in detail.System of surveillance for multi-objects has been carried into effect in TMS320DM642 DSP.The performance has stepped up after optimizing as a result.

Key words: intelligent vision surveillance, Digital Signal Processor(DSP), TMS320DM642, software optimization

摘要: 探讨了智能视觉监控算法在TI DSP平台上的移植方法。详细介绍了算法优化、CCS编译器优化、C代码优化、汇编指令优化及存储器配置优化等优化技术。通过具体的监控实例,在TMS320DM642平台上实现了多个运动目标的实时检测、跟踪算法,结果表明,对DSP进行优化之后其性能获得了指数级的增长。

关键词: 智能视觉监控, 数字信号处理技术(DSP), TMS320DM642技术, 软件优化