多目标实时跟踪可编程片上系统的软件优化

叶有时; 赵保军; 唐林波; 蔡晓芳

doi:10.3788/OPE.20111903.0681

您当前的位置：

首页 >

文章列表页 >

多目标实时跟踪可编程片上系统的软件优化

信息科学 | 更新时间：2020-08-12

- 多目标实时跟踪可编程片上系统的软件优化
- Software optimization of multi-target real-time tracking SOPC system
- 光学精密工程 2011年19卷第3期页码：681-689
- 作者机构：
  
  北京理工大学电子工程系北京,100081
- 作者简介：
- 基金信息：
  
  国家863高技术研究发展计划资助项目(No.2008AA8012320B)()
- DOI：10.3788/OPE.20111903.0681
  中图分类号： TP311;V243.5
- 收稿日期：2010-04-19，
  
  修回日期：2010-07-09，
  
  网络出版日期：2011-03-22，
  
  纸质出版日期：2011-03-22
- 稿件说明：
移动端阅览
叶有时, 赵保军, 唐林波, 蔡晓芳. 多目标实时跟踪可编程片上系统的软件优化[J]. 光学精密工程, 2011,19(3): 681-689

YE You-shi, ZHAO Bao-jun, TANG Lin-bo, CAI Xiao-fang. Software optimization of multi-target real-time tracking SOPC system[J]. Editorial Office of Optics and Precision Engineering, 2011,19(3): 681-689
叶有时, 赵保军, 唐林波, 蔡晓芳. 多目标实时跟踪可编程片上系统的软件优化[J]. 光学精密工程, 2011,19(3): 681-689 DOI： 10.3788/OPE.20111903.0681.

YE You-shi, ZHAO Bao-jun, TANG Lin-bo, CAI Xiao-fang. Software optimization of multi-target real-time tracking SOPC system[J]. Editorial Office of Optics and Precision Engineering, 2011,19(3): 681-689 DOI： 10.3788/OPE.20111903.0681.

摘要

针对传统DSP+FPGA设计方案结构复杂、功耗高、体积大、抗干扰性差等问题

应用可编程片上(SOPC)技术建立了多目标实时跟踪系统并提出了基于NIOS II核及其数据链路特征的软件优化方法

对系统采用的迭代投影跟踪算法和通信制导控制算法进行了优化。研究了算法结构优化

链路配置优化

代码优化

自定义指令和C2H技术

并将这些技术应用于多目标跟踪SOPC系统来优化算法结构

减少代码量

去除程序相关性并进行链路层硬件转换

提升软件并行流水性能。实验结果表明

该优化方法可使软件效率提升9~18倍。在图像序列100 frame/s

分辨率256 pixel256 pixel的条件下

在1.3 ms内完成了单帧图像的多目标跟踪处理

470 s内完成了单周期导航运算处理

使系统运算满足了实时处理的要求并在节约硬件成本的同时了简化结构

提高了运行稳定性。

Abstract

For the problems of complex structure

high power consumption

large size and poor anti-interference in the traditional DSP+FPGA design

a real-time multi-target tracking system based on a System on Programmble Chip(SOPC) was established and a software optimization method based on NIOS II core and its data link was proposed. The optimization technologies for algorithm structure

data link configuration

code optimization

custom instruction and C2H were researched. Then

these technologies were applied to the multi-target tracking SOPC system to optimize algorithm structure

reduce the size of code

remove the correlation of code and implement link-layer hardware conversion to enhance the performance of software pipelining. Experimental results indicate that these optimizing technologies can improve the efficiency of software by 9-18 times. The computing time of multi-target tracking is 1.3 ms/frame and the navigation operation time is 470 s/cycle for the image sequence of 256 pixel256 pixel in 100 frame/s. The system can realize real-time processing with low hardware costs

simple structure and stable performance.

关键词

Keywords

references

漆长松,刘恩海,钟建勇. 天基平台深空运动小目标检测[J]. 光电工程,2008,35(5):28-32. QI CH S, LIU EN H, ZHONG J Y. Discussion on space-based deep space small moving target detection [J]. Opto-Electronic Engineering, 2008,35(5):28-32. (in Chinese)[2] CLARK F O, DANIEL P H. Automatic target recognition by matching oriented edge pixels [J]. IEEE Transactions on Image Processing, 1997,6(1):103-113.[3] YIN CH F, ZHAO D. The research on target recognition and image tracking system based on high-speed DSP . Proceedings of the 4th International Conference on Computer Science & Education, Nanning, P.R. China: ICCSE,2009:697-700.[4] 赵鹏,沈庭芝,单宝堂. 一种基于粒子滤波的无人机电视导引系统目标跟踪算法[J]. 光学精密工程, 2008,16(1):134-140. ZHAO P, SHEN T ZH, SHAN B T. An object tracking algorithm for TV guiding system of UAV based on particle filter [J]. Opt. Precision Eng., 2008,16(1):134-140. (in Chinese)[5] 王广平,许廷发,倪国强,等. 多ADSP-TS201红外弱小目标实时检测跟踪系统的硬件设计[J]. 光学精密工程,2007,15(6):941-944. WANG G P, XU T F, NI G Q, et al.. Hardware design of infrared dim target real time detection and tracking system based on multiple ADSP-TS201S [J]. Opt. Precision Eng., 2007,15(6):941-944. (in Chinese)[6] 许廷发,赵思宏,周生兵,等. DSP并行系统的并行粒子群优化目标跟踪[J]. 光学精密工程,2007,17(9):2236-2240. XU T F, ZHAO S H, ZHOU SH B, et al.. Particle swarm optimizer tracking based on DSP parallel system [J]. Opt. Precision Eng., 2007,17(9):2236-2240. (in Chinese)[7] ALTERA.NIOS II processor reference handbook [M]. Published by Altera, 2009.[8] XU H, WEI SH SH. An embedded SOPC system using automation design . Proceedings of the International Conference on Parallel Processing Workshops, Oslo, Norway: ICPP,2005:232-239.[9] 叶有时,唐林波,赵保军,等. 基于SOPC的深空目标实时跟踪系统[J]. 系统工程与电子技术,2009,31(12):3002-3006. YE Y SH, TANG L B, ZHAO B J, et al.. Real-time deep space target track system based on SOPC [J]. Systems Engineering and Electronics, 2009,31(12):3002-3006.(in Chinese)[10] 丁瑞,史彩成,何佩琨. 基于FPGA+DSP系统的红外图像非均匀校正的实现[J]. 光学技术,2006,32(8):121-123. DING R, SHI C CH, HE P K. The realization of two-point non-uniformity correction of infrared image based on FPGA+DSP [J]. Optical Technique, 2006,32(8):121-123.(in Chinese)[11] 王勇,赵保军. 复杂背景下低信噪比点目标的实时检测算法及实现[J]. 系统工程与电子技术,2002,24(12):18-21. WANG Y, ZHAO B J. A real-time detection algorithm for low SNR point targets in a complex background [J]. Systems Engineering and Electronics, 2002,24(12):18-21.(in Chinese)[12] Bruno Pimentel. Recursion in hardware: applicability and implementation strategies . Proceedings of the Second International Conference on Advances in Circuits, Electronics and Micro-electronics. Sliema, Malta: CENICS, 2009:94-100.[13] 孟占红,赵保军. 基于DSP的实时图像压缩软件优化技术研究[J]. 电子学报,2006,34(9):1558-1561. MENG ZH H, ZHAO B J. The study on software optimizing technology of real-time image compression system based on DSP [J]. Acta Electronica Sinica, 2006,34(9):1558-1561.(in Chinese)[14] 虞响,张笑微,潘泽友. 实时目标跟踪中TMS320 DM642的软件优化研究[J]. 兵工自动化,2009,28(6):86-91. YU X, ZHANG X W, PAN Z Y. Software optimization in real-time tracking system of moving object based TMS320DM642 [J]. Ordnance Industry Automation, 2009,28(6):86-91. (in Chinese)[15] ALTERA.Embedded Design Handbook [M]. Published by Altera, 2010.

浏览量

517

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

目标运动轨迹匹配式的红外-可见光视频自动配准

红外成像系统的非均匀性实时校正

星上时间延迟积分CCD拼接相机图像的实时处理

基于TMS320C6416T的MPEG-4编码优化

红外图像实时显示增强系统设计