基于加权行质量向量的步态识别方法
首发时间:2010-05-17
摘要:为了在降低样本训练时间的同时提高数据分类能力,提出一种基于加权行质量的步态识别方法,首先提取人体轮廓行质量向量作为步态特征,分析特征向量各元素的贡献度从而对特征向量进行加权,最后采用归一化欧氏距离度量相似度,使用最近邻分类器进行分类。在CASIA数据库上的实验结果表明,该步态识别方法既满足了步态识别对实时性的要求又保证了较高的识别率。
For information in English, please click here
Gait recognition based on weighted mass vector
Abstract:In order to reduce the training time while increasing the sample data classification capabilities, a new approach to the problem of gait recognition based on weighted mass vector is proposed in this paper, which uses the mass vector as gait feature, then weights it by analyzing the contribution of each mass vector element. The NED (normalized Euclidean distance) is used for matching and the nearest neighbor classifier is used for classifying. Experimental results on CASIA gait database demonstrate that the proposed algorithm can greatly reduce the training time and achive high recognition rate.
Keywords: Gait Recognition Linear Discriminant Analysis mass vector
基金:
论文图表:
引用
No.3588365343511273****
同行评议
共计0人参与
勘误表
基于加权行质量向量的步态识别方法
评论
全部评论0/1000