基于直方图插值的均值移动小尺寸目标跟踪算法

陈建军; 安国成; 张索非; 吴镇扬

doi:10.3724/SP.J.1146.2009.01245

基于直方图插值的均值移动小尺寸目标跟踪算法

doi: 10.3724/SP.J.1146.2009.01245

陈建军^{① 安国成},
安国成,
张索非,
吴镇扬

基金项目:

国家自然科学基金(60672094)资助课题

计量
- 文章访问数: 3604
- HTML全文浏览量: 99
- PDF下载量: 1088
- 被引次数: 0
出版历程
- 收稿日期: 2009-09-22
- 修回日期: 2010-04-02
- 刊出日期: 2010-09-19

Small Target Tracking Based on Histogram Interpolation Mean Shift

摘要

摘要: 小尺寸目标跟踪是视觉跟踪中的难题。该文首先指出了均值移动小尺寸目标跟踪算法中的两个主要问题：算法跟踪中断和丢失跟踪目标。然后，论文给出了相应的解决方法。对传统Parzen窗密度估计法加以改进，并用于对候选目标区域的直方图进行插值处理，较好地解决了算法跟踪中断问题。论文采用Kullback-Leibler距离作为目标模型和候选目标之间的新型相似性度量函数，并推导了其相应的权值和新位置计算公式，提高了算法的跟踪精度。多段视频序列的跟踪实验表明，该文提出的算法可以有效地跟踪小尺寸目标，能够成功跟踪只有612个像素的小目标，跟踪精度也有一定提高。
- 小尺寸目标跟踪 /
- 均值移动 /
- 直方图插值 /
- Parzen窗 /
- 相似性度量
Abstract: Small scale target tracking is one of the primary difficulties in visual tracking. Two major problems in mean shift small target tracking algorithm are presented in this paper, namely tracking interrupt and target losing. To tackle these problems, the Parzen windows density estimation method is modified to interpolate the histogram of the target candidate. The Kullback-Leibler distance is employed as a new similarity measure function of the target model and the target candidate. And its corresponding weight computation and new location expressions are derived. On the basis of these works, a new mean shift algorithm is proposed for small target tracking. Several tracking experiments for real world video sequences show that the proposed algorithm can track the target successively and accurately. It can successfully track very small targets with only 612 pixels.
- Small target tracking /
- Mean shift /
- Histogram interpolation /
- Parzen window /
- Similarity measure

HTML全文

参考文献(1)

[1] Kaewtrakulpong P and Bowden R. A real time adaptive visual surveillance system for tracking low-resolution colour targets in dynamically changing scenes [J].Image and Vision Computing.2003, 21(10):913-929 [2] Han B and Davis L S. Probabilistic fusion-based parameter estimation for visual tracking [J].Computer Vision and Image Understanding.2009, 113(4):435-445 [3] Formont S, Laude V, and Refregier P. Small target tracking on image sequence using nonlinear optimal filtering [C]. Signal and Data Processing of Small Targets, 1995. San Diego, CA, USA, SPIE, 1995, 2561: 299-307. [4] Wang L, Hu S, and Zhang X. Detecting and tracking of small moving target under the background of sea level [C]. The 9th International Conference on Signal Processing, ICSP'2008, Beijing, China, 2008: 989-992. [5] Fukunaga K and Hostetler L. The estimation of the gradient of a density function, with applications in pattern recognition [J].IEEE Transactions on Information Theory.1975, 21(1):32-40 [6] Yizong C. Mean shift, mode seeking, and clustering [J].IEEE Transactions on Pattern Analysis and Machine Intelligence.1995, 17(8):790-799 [7] Comaniciu D, Ramesh V, and Meer P. Kernel-based object tracking [J].IEEE Transactions on Pattern Analysis and Machine Intelligence.2003, 25(5):564-577 [8] Bajramovic F, Gr C, and Denzler J. Efficient combination of histograms for real-time tracking using mean-shift and trust-region optimization [C]. The 27th Symposium on German Association for Pattern Recognition, DAGM2005, Vienna, Austria, Springer Verlag, 2005, 3663: 254-261. [9] Li P. An Adaptive Binning Color Model for Mean Shift Tracking [J].IEEE Transactions on Circuits and Systems for Video Technology.2008, 18(9):1293-1299 [10] Hongxia C and Wang K. Target tracking based on mean shift and improved kalman filtering algorithm [C]. IEEE International Conference on Automation and Logistics, ICAL '09. Shenyang, China, 2009: 808-812. [11] Li S X, Chang H X, and Zhu C F. Adaptive pyramid mean shift for global real-time visual tracking [J].Image and Vision Computing.2010, 28(3):424-437 [12] Yang C, Duraiswami R, and Davis L. Efficient mean-shift tracking via a new similarity measure [C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR '2005. San Diego, CA, United states, 2005: 176-183. [13] Parzen E. On estimation of a probability density function and mode [J].Annals of Mathematical Statistics.1962, 33(3):1065-1076 [14] Shen H and Yan X. Probability density estimation over evolving data streams using tilted Parzen window [C]. IEEE Symposium on Computers and Communications, ISCC'2008, Marrakech, 2008: 585-589. [15] CAVIAR: EU funded project, IST 2001 37540, URL:http: //homepages.inf.ed.ac.uk/rbf/CAVIAR/(2004).

施引文献

资源附件(0)

访问统计