摘要
提出了一种基于SVM(支持向量机)人形识别的算法,通过对静态图像小波变换提取目标的局部形状突变特征,并结合动态帧的步态特征,然后利用支持向量机对小样本进行学习与识别.通过实验验证,该算法具有实时性好、识别率高、可靠性高、适用范围广等特点,以达到实现监控自动化和智能化的目标.
A human detection algorithm based on SVM（Support Vector Machines） is proposed.The local shape mutation features of the targets are extracted through wavelet transformation of the static image.The features are then combined with the gait feature of the dynamic frame to be studied and recognized by support vector machines.It is proved through experiments that this method has the features of real-time,high accuracy and wide range.
出处
《昆明理工大学学报（理工版）》
北大核心
2010年第5期70-74,共5页
Journal of Kunming University of Science and Technology(Natural Science Edition)
基金
昆明市科技局重大计划项目(项目编号:昆科计字08J100305)
关键词
小波变换
步态特征
支持向量机
人形识别
wavelet transformation
gait feature
support vector machines
human detection
