基于连续HMM与静态外观信息模型融合的步态识别  

Gait Recognition Based on HMM and Static Appearance Information

作　　者：张前进[1] 孙炎增[1] 徐素莉[1] 

机构地区：[1]河南科技大学电子信息工程学院,河南洛阳471003

出　　处：《微电子学与计算机》2009年第3期45-48,共4页Microelectronics & Computer

摘　　要：利用步态轮廓图像下肢左右边界的距离矢量对步态轮廓图像进行描述,采用步态图像两脚的宽度进行准周期性分析.利用隐马尔可夫模型并融合步态的静态外观信息模型进行步态时变数据匹配识别.算法在CMU数据库上进行实验取得了较高的正确识别率.The paper attempts to describe gait contour by using the distance between the legs and feet on the edge,to make quasi-periodic analysis on width of the foot of gait image.And to solve the problems resulting from image sequence of different gait cycle by using fusion of HMM and static cues of body biometrics at a decision level.By utilizing the algorithm in the paper,the experiments with CMU databases have achieved comparatively high correction identification ratio.

关 键 词：步态识别 特征提取 HMM 信息融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]