一种基于步幅长度及频域特征的步态识别方法

步态识别作为一种较新的生物认证技术有其独特优势,它旨在根据人行进过程中的行走步态模式来识别其身份.提出一种基于步幅长度及频域特征的步态识别方法:首先采用背景减除算法实现运动人体检测,并根据步幅长度变化特征进行周期分割.然后提取步幅长度极值结合关键帧的频幅均值构成五维特征向量,最后用标准模式分类器在中科院自动化所提供的CASIA步态数据库中进行实验,在降低了算法复杂度的同时,获得了令人鼓舞的实验结果.

A Gait Recognition Method Based on Step Length and Frequency Domain Characteristics

As a new biometrics recognition technology, gait recognition has its unique advantages.It aims to recognize individuals by the people's walking gait pattern.A gait recognition method based on step length and frequency domain characteristics is presented in this way:First,we chose the background subtraction methods to get the motion area,and use step length feature to divide period.Second,we calculate the frequency energy of the key frames,and extract the mean value of them asthe features,together with the step length made a 5 dimensional feathers vector.At last,two supervised pattern classification technique were performed for recognition.Extensive experimental results on CASIA database demonstrate that the proposed algorithm has an encouraging recognition performance with relatively lower computational cost.