基于SVM的步态识别方法综述

步态识别是非接触式生物特征识别领域的前沿课题,通过对人体行走方式的识别以确定个体的身份,在智能视频监控领域有较高的研究价值.步态分类是步态识别过程中的重要任务和关键步骤.首先概述了步态识别过程及分类方法,然后重点对基于支持向量机的步态分类方法进行了综述,分析了基于该方法的最新研究进展,对每个具体研究方法的优缺点进行了对比.最后,指出目前步态识别在实际应用中存在的局限性,并对该领域发展方向进行了展望.

Survey on Gait Recognition Methods Based on SVM

Gait recognition is a leading subject in the field of non-contact biometric identification,it can verify the identity of the individual by means of human walking,which has high research value in the field of intelligent video surveillance.Gait classification is an important task and a key step in the process of gait recognition.Firstly,the process of gait recognition and classification approaches are introduced.Secondly,the gait classification approach based on SVM is reviewed and its latest research developments are analyzed.The advantages and disadvantages of each specific research method are compared.Finally,the limitations existing in the gait recognition in practical application are pointed out and the development direction in this field is prospected.