J4 ›› 2015, Vol. 42 ›› Issue (3): 67-72.doi: 10.3969/j.issn.1001-2400.2015.03.012

• 研究论文 • 上一篇    下一篇



  1. (西安电子科技大学 计算机学院,陕西 西安  710071)
  • 收稿日期:2014-05-27 出版日期:2015-06-20 发布日期:2015-07-27
  • 通讯作者: 屈鉴铭
  • 作者简介:屈鉴铭(1985-),男,西安电子科技大学博士研究生,E-mail: jmqu@mail.xidian.edu.cn.
  • 基金资助:


Directed AdaBoost weighting classifier based on the scene motion pattern for object detection

QU Jianming;LIU Zhijing;HE Wenhua   

  1. (School of Computer Science and Technology, Xidian Univ., Xi'an  710071, China)
  • Received:2014-05-27 Online:2015-06-20 Published:2015-07-27
  • Contact: QU Jianming



关键词: 目标检测, 场景运动模式, 有向AdaBoost, 加权分类, 光流


The regular behavior reflected by the interaction of spatial layout and interior elements of the scene with the moving object is named the scene motion pattern. In this paper, a novel approach of the object detection combining scene motion pattern with the directed AdaBoost weighting method is proposed to remedy the limitation of the existing detection algorithms that give little consideration of the object motion regularity and the priori information about the scene. For this reason, a model of the scene motion pattern described by a matrix of speed-weighting directed histogram is created, and the information about the scene motion pattern is acquired by the voting of the sparse optical flows on that basis. Meanwhile, a directed AdaBoost weighting detection algorithm is developed correspondingly. A set of directed AdaBoost classifiers which are then weighted according to the motion pattern of the region are established in the algorithm. According to the specially designed cross-validation classification experiments and video tests, the precision rates of the algorithm are about 10% higher than that of standard AdaBoost detector under the same condition of the false positive rate, which proves the effectiveness and the advancements of the proposed approach.

Key words: object detection, scene motion pattern, directed AdaBoost, weighted classification, optical flows


  • P315.69