基于双流网络的步态特征表述算法  

A gait feature representation algorithm based on double flow network

作　　者：杜晓雨 张学东[1] DU Xiaoyu;ZHANG Xuedong(School of Computer and Software Engineering,University of Science and Technology Liaoning,Anshan 114051,China)

机构地区：[1]辽宁科技大学计算机与软件工程学院,辽宁鞍山114051

出　　处：《辽宁科技大学学报》2022年第1期65-70,共6页Journal of University of Science and Technology Liaoning

摘　　要：针对步态特征识别中步态帧图像数据量大、网络处理速度慢、计算效率低等问题,本文提出基于双流网络的步态特征表述方法,先利用边缘检测算法处理步态序列集合,再利用主成分分析进行降维处理得到特征提取结果。采用CAISIA-B步态数据库进行验证,仿真实验表明该方法不仅能够保证识别率,同时还能提高网络的运算效率。Aiming at the problems of large amount of gait frame image data,slow network processing,and low computational efficiency in gait feature recognition,a gait feature representation method based on dual flow network is proposed.First,the edge detection algorithm is used to process the gait sequence set.Then,principal component analysis is used to reduce the dimension and obtain feature extraction results.The CAI-SIAB gait database is used for verification,and the results show that this method can not only ensure the rec-ognition rate,but also improve the computing efficiency of the network.

关 键 词：特征提取 主成分分析法 步态序列 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]