摘要
现有的视觉跟踪方法在复杂场景中依然难以有效跟踪运动物体,而融合多种特征能有效地提升算法在复杂场景中的跟踪效果。该文研究了在粒子滤波框架下基于多种特征融合的视觉跟踪算法,通过在线训练支持向量机分类器构造判别式目标表观模型。在复杂场景视频中进行测试,实验结果表明该文跟踪算法具有较好的鲁棒性和准确性。
It is still difficult for existing visual tracking algorithms to track effectively moving objects in complex scenes. However, it can promote tracking performances largely by integrating multiple features in complex scenes. This paper proposes a multiple features fusion based visual tracking algorithm in particle filter framework. And the online trained support vector machine classifier is used to construct the discriminative appearance model. The method is tested in videos with complex scenes, and experimental results shows that the proposed tracking algorithm is robust and accurate.
作者
丁建伟
刘伟
DING Jian-wei, LIU Wei （1.People＇s Public Security University of China, Beijing, 102623, China; 2.School of Electro-mechanical Engineering, Nanyang Normal University, Nanyang 473061, China; 3.Oil Equipment Intelligent Control Engineering Laboratory of Henan Province, Nanyang Normal University, Nanyang 473061, China）
出处
《电脑知识与技术》
2016年第1期196-197,共2页
Computer Knowledge and Technology
基金
国家自然科学基金（61503388,61503202）、中国人民公安大学2015年度中央高校基本科研业务费项目《智能视频分析关键技术研究》、河南省重点科技攻关项目（152102210336）
河南省高等学校重点科研项目（15A413005）
南阳市科技攻关项目（KJGG28）
南阳师范学院专项项目（ZX2015007）资助
关键词
视觉跟踪
复杂场景
融合
支持向量机
粒子滤波
visual tracking
complex scenes
fusion
support vector machine
particle filter
