摘要
基于分类的跟踪算法成为当前目标跟踪的研究热点.首先把跟踪问题看成是一个目标和背景的二分类问题,根据每一帧的正负样本数据训练SVM分类器,通过分类器的分类概率值确定目标位置.然而,采集正负样本边界的那些样本很容易出现异常点,当把它们作为目标的下一帧位置时将会出现严重的跟踪漂移问题.本文在此基础上提出一种基于单类支持向量机(One-class support vector machine)的目标跟踪算法,基于One-class SVM分类能有效地排除其他类的干扰,有效地防止异常样本的出现.并结合加权多示例采样方法,使得每个采样样本会根据不同的权值对于分类器的贡献而不同.实验结果表明本文改进跟踪方法的鲁棒性.
Object tracking algorithm based on binary classification has become the research hot issues.The tracker is firstly as binary classification between obj ect and background,and both positive and negative samples data have been applied to train SVM classifier.The obj ect’s location is determined by the proba-bility of the classifier.However,such binary classification may not fully handle the outliers,which may cause tracking drifting.To improve the robustness of the tracker,a novel obj ect tracking algorithm has been proposed based on one-class SVM.This method based on one-class support vector machine （SVM） can effectively rule out other types of interference,to effectively prevent the emergence of abnormal sam-ples.Furthermore,the tracker integrates weighted multi-instance sampling method,which can consider the sample importance by the different weights.The experimental results show the robustness of the im-proved method.
出处
《西南师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2014年第3期82-89,共8页
Journal of Southwest China Normal University(Natural Science Edition)
基金
四川省教育厅科研资助项目(13ZA0135)
关键词
二分类
目标跟踪
单类支持向量机
加权多示例采样方法
binary classification
obj ect tracking
one-class SVM
weighted multi-instance sampling method
