一种基于粒子滤波的自适应运动目标跟踪方法
doi: 10.3724/SP.J.1146.2005.00492
An Adaptive Particle Filter for Moving Objects Tracking
-
摘要: 该文提出了一种基于粒子滤波的自适应运动目标跟踪方法。均值漂移算法是一种最优梯度下降法，通过迭代来搜索目标，从而实现对运动目标的跟踪。而粒子滤波是一种在非线性和非高斯情形下进行跟踪的强有力方法。该文首先对图像的直方图进行改进，提出了一种基于统计直方图分布的目标模型，然后通过这个模型将这两种方法有效地结合起来。根据跟踪的过程，自适应地调整参数，能够较好地处理图像序列中由于光线变化或遮挡所带来的影响。实验证明，该文所提出的方法与均值漂移方法相比，即使在复杂的情形下，也能够准确地对目标进行跟踪。
-
关键词:
- 运动目标跟踪;粒子滤波;直方图
Abstract: In this paper, an adaptive particle filter for moving objects tracking is proposed. Mean shift is optimization algorithm based on gradient descended, which tracks moving targets through iterations. Particle filter is a robust method of tracking in non-Gauss and non-linear case. Firstly, a target model based on statistical histogram is proposed, which improves the classical histogram. Then Mean Shift algorithm and particle filter are integrated novelly through the statistical histogram target model. The parameters are modified according to the processing of tracking, so the effects caused by changed light or occlusion can be overcome. Experiments show that the method proposed by this paper can track moving target more powerful than Mean Shift tracked. Otherwise, even in complicated case, this method can still efficiently work. -
[1] Paragios N and Deriche R. Geodesic active contours and level sets for the detection and tracking of moving objects. IEEE Trans. on Pattern Anal. Mach. Intell., 2000, 3(22): 262－280. [2] Arulampalam M, Maskell S, Gordon N, and Clapp T. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking[J].IEEE Trans. on Singal Processing.2002, 50(2):174- [3] Doucet A, Gordon N, and Krishnamurthy V. Particle filters for state estimation of jump Markov linear systems[J].IEEE Trans. on Signal Processing.2001, 49(3):613- [4] Comaniciu D, Ramesh V, and Meer P. Real-time tracking of non-rigid objects using mean shift. IEEE Conference on Computer Vision and Pattern Recognition. Hilton Head Island, South Carolina. 2000, II: 142－149. [5] Kailath T. The divergence and Bhattacharyya dstance measures in signal selection. IEEE Trans. on Commun. Technol., 1967, COM-15: 52－60. [6] Perez P, Hue C, Vermaak J, and Gangnet M. Color-based probabilistic tracking. European Conference on Computer Vision. Copenhagen, Denmark. 2002, 1: 661－675. [7] Nummiaro K.[J].Koller-Meier E, and Van Gool L. Object tracking with an adaptive color-based particle filter. First International Workshop on Generative-Model-Based Vision, in conjunction with ECCV02.Copenhagen, Denmark.2002,:- -
计量
- 文章访问数: 4064
- HTML全文浏览量: 104
- PDF下载量: 2325
- 被引次数: 0
下载:
