基于多分类器投票机的人体姿态识别算法  

Human Posture Recognition Method Based on Multiple Classifiers

作　　者：陈慧杰[1] 谢毅雄 

机构地区：[1]厦门大学信息科学与技术学院,福建厦门361005

出　　处：《计算机与现代化》2014年第4期64-68,共5页Computer and Modernization

摘　　要：为了获得准确的人体姿态识别结果,满足智能视频监控的需求,提出一种融合Hu不变矩特征和傅里叶描述子特征的人体姿态识别算法,并将ReliefF算法引入特征选择过程中,区分特征的重要性,然后使用大数投票法构建多分类器投票机制进行姿态识别,该机制很好地发挥了各个分类器的优势,提高了识别的准确率。实验结果表明,提出的算法对各种姿态取得了很好的分类效果。To get an accurate body posture recognition result and meet the needs of intelligent video surveillance, we present a fu- sion of Hu invariant moments and Fourier descriptors characteristics of human posture recognition algorithm. The ReliefF feature selection algorithm is introduced into the process, for distinguishing the importance of characteristics. Voting mechanism of using multiple classifier is constructed with large numbers to vote, plays good advantages of the various classifiers to improve the recog- nition accuracy. Finally, experimental results show that the proposed algorithm achieves good classification results for various pos- tures.

关 键 词：姿态识别 HU不变矩 傅里叶描述子 多分类器 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]