基于遗传模糊分类的步态识别  

Gait Recognition Based on Genetic Fuzzy Classification

作　　者：申少昊[1] 夏利民[1] 孙正军[1] 

机构地区：[1]中南大学信息科学与工程学院,长沙410075

出　　处：《计算机科学》2009年第1期190-193,共4页Computer Science

基　　金：国家自然科学基金重大项目(79816101);湖南省自然科学基金资助项目(05JJ30121)资助

摘　　要：提出基于遗传模糊分类的步态识别方法。采用新的特征提取方法,该方法将目标按人体结构特点划分为多个子区域,利用各个子区域的质心与头部质心形成的距离和夹角对步态特征进行描述。运用模糊聚类算法构建不同距离函数的分类器,并用遗传算法对分类器进行集成,组成的集成融合分类器对步态序列进行识别。实验结果表明该方法具有较高的识别性能。A gait recognition algorithm based on genetic-fuzzy classification was proposed. A novel method of gait feature extraction was proposed, which divides human body area into several sub-areas according to body unique feature and describes gait features utilizing the distance and angle of the sub-areas centroid and head centroid. Utilizing fuzzy clustering and genetic algorithms constructed and integrated three classifiers of different distance function, then using integrated fuzzy classifier recognized gait sequence. The experiment result shows that the approach is effective and has encouraging recognition performance.

关 键 词：步态识别 特征提取 模糊聚类 遗传算法 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] O242.23[自动化与计算机技术—计算机科学与技术]