A Mean Shift Tracking Algorithm Based on Multi-Cue Fusion

您当前的位置：

首页 >

文章列表页 >

A Mean Shift Tracking Algorithm Based on Multi-Cue Fusion

更新时间：2025-07-09

- A Mean Shift Tracking Algorithm Based on Multi-Cue Fusion
- Vol. 43, Issue 10, Pages: 42-46(2009)
- 作者机构：
  
  西安交通大学电子与信息工程学院,西安,710049
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391
- Online First：10 October 2009，
  
  Published：2009
- 稿件说明：
移动端阅览
A Mean Shift Tracking Algorithm Based on Multi-Cue Fusion[J]. 2009, 43(10): 42-46.
DOI：

A Mean Shift Tracking Algorithm Based on Multi-Cue Fusion[J]. 2009, 43(10): 42-46. DOI：

摘要

针对传统基于的均值偏移(Mean Shift

MS)跟踪算法不能对运动目标准确跟踪的缺点

提出了一种融合多视觉线索的MS跟踪算法.首先根据目标在前一帧的估计位置

将目标搜索区域划分为目标区域和背景区域

其次在区域划分的基础上定义了一种新的颜色、运动线索直方图模型

能有效地抑制目标相邻背景的混乱干扰

最后基于MS理论框架提出了一种融合目标颜色、运动线索的跟踪算法

其颜色、运动线索可在跟踪过程中互补.实验表明

在目标快速运动、姿态发生较大变化或被遮挡的情况下

算法能够获得更为准确、鲁棒的跟踪结果.

Abstract

The traditional color-based mean shift algorithm is unable to accurately track the object in some challenging scenarios. A mean shift tracker that fuses multiple visual cues is proposed to address the problem. The search region of the object is partitioned into one target region and some background regions according to the estimated location of the object in the previous frame. Then based on the region partition

a novel histogram model is presented for characterizing the color and motion cues. The model can effectively suppress the clutter of adjacent backgrounds. Finally the visual tracker is proposed within the mean shift framework to fuse the color and motion cues so that both. The visual cues can be complemented each other and the accuracy and robustness of tracking is improved. Experiments show that the proposed tracker can obtain more accurate and robust tracking results in some challenging scenarios such as rapid motion

large pose variation or occlusion of objects.

关键词

Keywords

references

COLLINS R T. Mean-shift blob tracking through scale space [C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2003: 234-240.

PENG Ningsong, YANG Jie, LIU Zhi. Mean shift blob tracking with kernel histogram filtering and hypothesis testing[J]. Pattern Recognition Letters, 2005, 26(5): 605-614.

YANG C, DURAISWAMI R, DAVIS L S. Efficient mean-shift tracking via a new similarity measure [C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2005: 176-183.

POLAT E, OZDEN M. A nonparametric adaptive tracking algorithm based on multiple feature distributions [J]. IEEE Trans on Multimedia, 2006,8(6):1156-1163.

WANG Junqiu, YAGI Y. Integrating color and shape-texture features for adaptive real-time object tracking[J]. IEEE Trans on Image Processing, 2008, 17(2): 235-240.

JEYAKAR J, BABU V R. Robust object tracking with background-weighted local kernels [J]. Computer Vision and Image Understanding, 2008,112(3): 296-309.

COMANICIU D, MEER P. Kernel-based object tracking [J]. IEEE Trans on PAMI, 2003, 25(5): 564-577.

PÉREZ P, VERMAAK J, BLAKE A. Data fusion for visual tracking with particles [J]. Proceedings of the IEEE, 2004, 92(3): 495-513.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

An Improved Spatial-Temporal Regulari-ation Term Tracking Algorithm Combining Adaptive Spatial Weights

Knowledge Graph Representation Learning Driven by Semantic Discrimination and Subgraph Integration

Numerical Simulation of Heat-Moisture Barrier Performance of a Horizontal Air Curtain in Cold Storage

Electrothermal Coupling Research on the Evolutionary Law of Anti-Corona Performance for High-Altitude Air-Cooled Generators

Windage Loss and Flow Characteristics in Impeller Back Gaps of sCO2 Radial Inflow Turbines

Related Author

王鹏 1

孙梦宇 1

王海燕 3

李晓艳 1

吕志刚 1

Related Institution

School of Electronics and Information Engineering, Xi’an Technological University

School of Marine Science and Technology, Northwestern Polytechnical University

⁰