基于特征融合的人体行为识别
DOI:
CSTR:
作者:
作者单位:

作者简介:

邵延华(1982-),男,河南周口人,博士研究 生,主要从事计算机视觉和模式识别方面的研究.

通讯作者:

中图分类号:

基金项目:

教育部重点科研项目(108174)和教育部博士点基金(20130191110021)资助项目 (重庆大学 光电技术及系统教育部重点实验室,重庆 400044)


Human action recognition using multi-feature fusion
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为克服单个行为表达方法有效性上的不足,提出 了一种基于多特征融合和支持向量机(SVM)的人体行为识 别(HAR)方法。首先,利用背景差分提取运动显著区域;然后提取运动显著区域的剪影直方 图和光流直方图,并 采取一定的融合策略,构建融合特征结合SVM识别人体行为。实验以广泛使用的公开 数据集 Weizmann为研究对象,正确识别率达到99.8%以上。结果表明,提出 的特征融合及识别 方法能有效地对人体行为进行识别;而且,由于规避了比较耗时的序列匹配操作,减少了计 算量。

    Abstract:

    Human action recognition (HAR) has become one of the most active topi cs in computer vision and pattern recognition due to a wide range of promising applications.In order to o vercome the deficiency of single representation method,a new recognition a lgorithm of human action based on multi-feature fusion and support vector machine (SVM) is presented in this p aper.The proposed algorithm consists of three essential cascade modules.First,the human silhouette is obta ined by separating the salient regions and the background based on background subtraction.Then,the multi-feature fusio n is constructed by using two types of available features,the histogram of the silhouette and the optic flow. The human activity recognition can commonly be viewed as a multiclass classification problem.Final ly,the multiple features are sent to the SVM for recognizing the human activity.The experimental results sho w that the proposed method can achieve the correct recognition rate above 99.8% for the Weizmann benchmark data set.Inter-related analyses conclude that the proposed algorithm is effective and promising.The recogni tion performance of the SVM classifiers and some other mainstream classification techniques is also com pared,which further verifies the effectiveness of the proposed algorithm.

    参考文献
    相似文献
    引证文献
引用本文

邵延华,郭永彩,高潮.基于特征融合的人体行为识别[J].光电子激光,2014,(9):1818~1823

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2014-04-21
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期:
  • 出版日期:
文章二维码