| 基于多特征融合的头部跟踪方法研究* |
| |
| 引用本文： | 曹洁,李伟. 基于多特征融合的头部跟踪方法研究*[J]. 计算机应用研究, 2011, 28(8): 3155-3158. DOI: 10.3969/j.issn.1001-3695.2011.08.099 |
| |
| 作者姓名： | 曹洁 李伟 |
| |
| 作者单位： | 1. 兰州理工大学计算机与通信学院,兰州,730050
2. 兰州理工大学计算机与通信学院,兰州730050;中国人民解放军91666部队,浙江舟山316000 |
| |
| 基金项目： | 甘肃省自然科学基金资助项目（1010RJZA046）；甘肃省教育厅研究生导师基金资助项目（0914ZTB003）；甘肃省财政厅资助项目（0914ZTB148） |
| |
| 摘 要： | 为有效解决头部跟踪鲁棒性较差的问题，利用头部的多种观测信息通过ＤＳ证据理论进行融合跟踪。在粒子滤波的总体框架下，嵌入ＭｅａｎＳｈｉｆｔ算法产生更加逼近真实后验分布的粒子，同时采用颜色和人体头部椭圆轮廓的最大梯度距离测度（ＤＭＧ）作为观测模型，有效避免了单一颜色特征在光照突变、姿态变化、距离较远以及背景相似情况下的跟踪稳定性较差的问题。实验表明该方法对于复杂条件下的人体头部跟踪问题具有较强的鲁棒性。
|
| 关 键 词： | 头部跟踪；多特征融合；粒子滤波；最大梯度距离测量 |
Head tracking method based on multi-feature fusion |
| |
| CAO Jie,LI Wei. Head tracking method based on multi-feature fusion[J]. Application Research of Computers, 2011, 28(8): 3155-3158. DOI: 10.3969/j.issn.1001-3695.2011.08.099 |
| |
| Authors: | CAO Jie LI Wei |
| |
| Affiliation: | CAO Jie1,LI Wei1,2(1.College of Computer & Communication,Lanzhou University of Technology,Lanzhou 730050,China,2.PLA Troops of 91666,Zhoushan Zhejiang 316000,China) |
| |
| Abstract: | In order to effectively solve the poor performance of head tracking,this paper proposed an new method based fusing measurements of head by using D-S evidence theory. It used Mean-Shift algorithm to produce more effective particles that approache the real posterior distribution in the framework of particle filter. The proposed method used the color and distance to maximum gradient point (DMG) features as the observation model, and efficiently avoided the unsatble problems via using single color feature in the illumination of mutation, posture change, greater distance and similar background. Experiment results indicate the proposed method is more robust to present object and has good performance in complex scene. |
| |
| Keywords: | head tracking multi-feature fusion particle filter distance to maximum gradient point(DMG) |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
|
