目标跟踪系统在FPGA上的实时实现

doi:10.3969/j.issn.1001-2400.2012.05.028

J4 ›› 2012, Vol. 39 ›› Issue (5): 168-173.doi: 10.3969/j.issn.1001-2400.2012.05.028

目标跟踪系统在FPGA上的实时实现

刘速¹;孙伟峰²;王洪君¹

(1. 山东大学信息学院，山东济南 250014；
2. 中国石油大学(华东) 信息与控制工程学院，山东青岛 266555)

收稿日期:2011-10-31 出版日期:2012-10-20 发布日期:2012-12-13
作者简介:刘速(1984-)，女，山东大学博士研究生，E-mail: sueliu84@gmail.com．

Real-time object tracking system on the FPGAs

LIU Su¹;SUN Weifeng²;WANG Hongjun¹

(1. School of Information Science and Engineering, Shandong Univ., Jinan 250014， China;
2. College of Information and Control Engineering, China Univ. of Petroleum, Qingdao 266555， China)

Received:2011-10-31 Online:2012-10-20 Published:2012-12-13

摘要/Abstract

摘要：

目标跟踪是计算机视觉领域中的重要技术，它的难点之一在于其对实时性处理速度的要求．提出并应用高效的基于并行搜索的硬件架构，使得实时的目标跟踪系统在可配置器件上得以实现．采用基于减背景的算法，设计的目标跟踪系统采用并行架构来实现高速处理，并提出一种复合目标区域搜索技术来进一步加速系统在复杂跟踪环境下的运行速度．选用Altera Stratix Ⅲ EP3SL340H1152C2 FPGA器件作为硬件平台，将该目标跟踪算法基于CPU平台的软件实现(2.2GHz的处理器)与基于FPGA平台实现的处理速度进行了比较，速度提升最高可达100倍以上．

关键词: 目标跟踪, FPGA, 加速算法

Abstract:

Object tracking is an important task in computer vision applications. One of the crucial challenges is the real-time speed requirement. In this paper we implement an object tracking system in reconfigurable hardware using an efficient parallel architecture. In our implementation, we adopt a background subtraction based algorithm. The designed object tracker exploits hardware parallelism to achieve a high system speed. We also propose a dual object region search technique for further boosting the performance of our system under complex tracking conditions. For our hardware implementation we use the Altera Stratix III EP3SL340H1152C2 FPGA device. We compare the proposed FPGA-based implementation with the software implementation running on a 2.2GHz processor. It has been observed that the speed can be increased by more than 100 times for complex video inputs.

Key words: object tracking, FPGA, accelerating algorithm

刘速;孙伟峰;王洪君. 目标跟踪系统在FPGA上的实时实现[J]. J4, 2012, 39(5): 168-173.

LIU Su;SUN Weifeng;WANG Hongjun. Real-time object tracking system on the FPGAs[J]. J4, 2012, 39(5): 168-173.

参考文献

［1］ Yilmaz A, Javed O, Shah M. Object Tracking: a Survey［J/OL］.［2011-08-30］. http://dl.acm.org/citation.cfm?id=1177355.
［2］ Yuriy C. Real-time Object Tracker in C++［R/OL］.［2011-08-30］. http://cambridge.academia.edu/YuriyChesnokov.
［3］ Cho J U, Jin S H, Pham X D, et al. Multiple Objects Tracking Circuit using Particle Filters with Multiple Features［C/OL］.［2011-09-01］. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4209812.
［4］ Ali U, Malik M B, Munawar K. FPGA Soft-processor Based Real-time Object Tracking System［C/OL］. ［2011-09-01］. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4914888＆tag=1.
［5］ Xu J, Dou Y, Li J,et al. FPGA Accelerating Algorithms of Active Shape Model in People Tracking Applications［C/OL］. ［2011-09-05］. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4341504.
［6］ Leephokhanon S, Wiangtong T. Object Tracking and Motion Capturing in Hardware-Accelerated Multi-camera System ［J/OL］.［2011-09-05］. http://www.springerlink.com/content/e78t5332205q5n17/.
［7］ Mikolajczyk K, Schmid C. A Performance Evaluation of Local Descriptors［C/OL］.［2011-10-15］. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1498756.
［8］ Wren C, Azarbayejani A, Pentland A. Pfinder: Real-time Tracking of the Human body ［J］. IEEE Trans on Pattern Analysis and Machine Intelligence, 1997, 19(7): 780-785.
［9］ Rowley H, Baluja S, Kanade T. Neural Network-based Face Detection ［J］. IEEE Trans on Pattern Analysis and Machine Intelligence, 1998, 20(1): 23-38.
［10］ Viola P, Jones M, Snow D. Detecting Pedestrians Using Patterns of Motion and Appearance ［C/OL］.［2011-09-19］. http://www.springerlink.com/content/t61k38u53j531344/.
［11］ Papageorgiou C, Oren M, Poggio T. A General Framework for Object Detection ［C/OL］.［2011-09-15］. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=710772
［12］ Veenman C, Reinders M, Backer E. Resolving Motion Correspondence for Densely Moving Points ［J］. IEEE Trans on Pattern Analysis and Machine Intelligence, 2001, 23(1): 54-72.
［13］ Fieguth P, Terzopoulos D. Color-based Tracking of Heads and Other Mobile Objects at Video Frame Rates ［C/OL］.［2011-09-19］. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=609292.
［14］ Haritaoglu I, Harwood D, Davis L. W4: Real-time Surveillance of People and Their Activities ［J］. IEEE Trans on Pattern Analysis and Machine Intelligence, 2000, 22(8): 809-830.
［15］ Tanizaki H. Non-gaussian State-space Modeling of Nonstationary Time Series ［J］. Journal of the American Statististical Association, 1987, 82(400): 1032-1063.

[1]	汪晋, 苏洪涛, 汪圣利, 陆超. 基于SLSTM网络的两级修正机动目标跟踪方法[J]. 西安电子科技大学学报, 2025, 52(1): 37-49.
[2]	严俊坤, 张聪睿, 李婉萍, 戴金辉, 张鹏, 刘宏伟. 面向编队突防的多干扰机协同资源分配方法[J]. 西安电子科技大学学报, 2024, 51(5): 24-34.
[3]	张永权, 李志彬, 张文博, 苏镇镇. 多源传感器箱粒子LMB滤波算法[J]. 西安电子科技大学学报, 2024, 51(4): 51-66.
[4]	贺王鹏, 胡德顺, 李诚, 周悦, 郭宝龙. 结合模板更新与轨迹预测的孪生网络跟踪算法[J]. 西安电子科技大学学报, 2024, 51(3): 46-54.
[5]	张逸宸,水鹏朗,廖沫. 目标回波特征辅助的海面多目标跟踪方法[J]. 西安电子科技大学学报, 2023, 50(5): 1-10.
[6]	李翠芸,衡博文,谢金池. 模糊推理优化的抗遮挡PMBM跟踪算法[J]. 西安电子科技大学学报, 2023, 50(5): 54-64.
[7]	陈振,李翠芸,李想. B样条曲面三维扩展目标跟踪算法[J]. 西安电子科技大学学报, 2023, 50(2): 101-111.
[8]	张兆宇,田春娜,周恒,田西兰. 联合在线分类的双注意力RGBT孪生网络跟踪[J]. 西安电子科技大学学报, 2022, 49(6): 76-85.
[9]	张昀普,单甘霖,付强. 多传感器协同地面目标跟踪的调度方法[J]. 西安电子科技大学学报, 2022, 49(5): 100-108.
[10]	孙雅媚,肖嵩,曲家慧,董文倩. 自适应上下文感知相关滤波类目标跟踪算法[J]. 西安电子科技大学学报, 2022, 49(3): 21-27.
[11]	宋建锋,苗启广,王崇晓,徐浩,杨瑾. 注意力机制的多尺度单目标跟踪算法[J]. 西安电子科技大学学报, 2021, 48(5): 110-116.
[12]	顾兆军,陈辉,王家亮,高冰. 一种小型四轴飞行器目标跟踪控制算法[J]. 西安电子科技大学学报, 2021, 48(5): 117-127.
[13]	王海军,张圣燕,杜玉杰. 响应差异约束的相关滤波无人机目标跟踪算法[J]. 西安电子科技大学学报, 2021, 48(5): 149-155.
[14]	毛昭勇,王亦晨,王鑫,沈钧戈. 面向高速公路的车辆视频监控分析系统[J]. 西安电子科技大学学报, 2021, 48(5): 178-189.
[15]	逯彦,廖桂生,黄庆享. 结合稀疏表示的跨摄像头运动目标跟踪算法[J]. 西安电子科技大学学报, 2021, 48(2): 197-204.

目标跟踪系统在FPGA上的实时实现

Real-time object tracking system on the FPGAs

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0