基于颜色的粒子滤波算法的改进与全硬件实现

王一木; 潘赟; 严晓浪

doi:10.3724/SP.J.1146.2010.00294

基于颜色的粒子滤波算法的改进与全硬件实现

doi: 10.3724/SP.J.1146.2010.00294

基金项目:

国家自然科学基金 (60720106003)和国家863计划项目(2009AA011 706)资助课题

计量
- 文章访问数: 3821
- HTML全文浏览量: 97
- PDF下载量: 1881
- 被引次数: 0
出版历程
- 收稿日期: 2010-03-26
- 修回日期: 2010-10-18
- 刊出日期: 2011-02-19

An Improved Color-based Particle Filter and Its Full Hardware Implementation

摘要

摘要: 传统基于颜色的粒子滤波算法在硬件实现中存在着跟踪效果不理想、实时性差等问题。该文结合硬件电路需要对基于颜色的粒子滤波算法进行了改进，在传统SR重采样算法的基础上将剩余粒子撒向目标点附近，以提高其在硬件环境下跟踪的准确性与稳定性。文中给出了改进算法的全硬件实现的电路架构，并在FPGA上完成了目标跟踪系统的实现。实验表明提出改进算法与硬件实现方案对简单背景环境下运动目标有着良好的跟踪效果，系统实时处理能力可达72 FPS，而硬件消耗为7387个逻辑单元。为进一步适应复杂的应用环境，在此粒子滤波器基础上给出了可扩展的分布式粒子滤波系统的架构以及相应的重采样策略，良好的并行性与可扩展性使得系统能够完成复杂背景环境下多特征、多目标的跟踪任务。
- 目标跟踪 /
- 粒子滤波 /
- 实时性 /
- 可扩展性
Abstract: Nowadays, the research on the implementation of color-based particle filters is facing the problem of tracking accuracy and real-time processing speed. This paper presents a modified color-based particle filter, which basing on the traditional SR resample algorithm, scatters the left particles causing by hardware circuits around the target. Results show that proposed particle filters improve the accuracy and robustness of object tracking. Moreover, the architecture of its full hardware implementation is depicted in the paper. The experimental study on FPGA indicates that the proposed color-based particle filters perform robust tracking at 72 FPS and with 7387 LEs (Logic Element) hardware cost. In addition to that, the framework of scalable distributed particle filters and its resample scheme are presented to adapt more complex scenarios, carrying out multi-feature and multi-target tracking.
- Object tracking /
- Particle filter /
- Real-time /
- Scalability

HTML全文

参考文献(1)

Arulampalam M S, Maskell S, and Gordon N, et al.. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking[J].IEEE Transactions on Signal Processing.2002, 50(2):174-188[2]Doucet A, Gordon N J, and Krishnamurthy V. Particle filters for state estimation of jump Markov linear systems[J].IEEE Transactions on Signal Processing.2001, 49(3):613-624[3]洪少华, 史治国, 陈抗生. 用于纯方位跟踪的简化粒子滤波算法及其硬件实现[J].电子与信息学报.2009, 31(1):96-100浏览Hong Shao-hua, Shi Zhi-guo, and Chen Kang-sheng. Simplified algorithm and hardware implementation for particle filter applied to bearings-only tracking[J].Journal of Electronics Information Technology.2009, 31(1):96-100[4]Ye Bin-li and Zhang Yun-hua. Improved FPGA implementation of particle filter for radar tracking applications. Asian-Pacific Conference on Synthetic Aperture Radar, Xi'an, China, 2009: 943-946.[5]Perez P, Hue C, and Vermaak J, et al.. Color-based probabilistic tracking. Proceedings of the 7th European Conference on Computer Vision-Part I, Copenhagen, Denmark, 2002: 661-675.[6]Medeiros H, Park J, and Kak A. A parallel color-based particle filter for object tracking. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, Alaska, 2008: 1-8.[7]Wang Tzu-Heng, Chang Jing-Ying, and Chen Liang-Gee. Algorithm and architecture for object tracking using particle filter. IEEE International Conference on Multimedia and Expo, New York, 2009: 1374-1377.[8]Li Gong-yan, Li Bin, and Liu Zhou, et al.. Implementation and optimization of particle filter tracking algorithm on Multi-DSPs system. IEEE Conference on Cybernetics and Intelligent Systems, Chengdu, China, 2008: 52-157.[9]Doucet A, Freitas N D, and Gordon N E. Sequential Monte Carlo methods in practice. New York: Springer Verlag, 2001, Chapter 15-26.[10]BoliM, Athalye A, and Hong S, et al.. Generic hardware architectures for sampling and resampling in particle filters. EURASIP Journal of Applied Signal Processing, 2005, (17): 2888-2902.[11]Boli M, Athalye A, and Hong S, et al.. Study of algorithmic and architectural characteristics of Gaussian particle filters[J].Journal of Signal Processing Systems.2010, 61(2):205-218[12]Lehmann E A and Williamson R C. Experimental comparison of particle filtering algorithms for acoustic source localization in a reverberant room. IEEE International Conference on Acoustics, Speech, and Signal Processing, Hong Kong, 2003: 177-180.[13]Jung Uk Cho, Seung Hun Jin, Xuan Dai Pham, and Jae Wook Jeon. Multiple objects tracking circuit using particle filters with multiple features. IEEE International Conference on Robotics and Automation, Roma, Italy, 2007: 4639-4644.[14]BoliM, Djuri P M, and Hong S. Resampling algorithms and architectures for distributed particle filters[J].IEEE Transactions on Signal Processing.2005, 53(7):2442-2450

施引文献

资源附件(0)

访问统计