一种基于步态分析的身份识别方法

近来年,步态作为一种生物特征识别技术倍受计算机视觉研究者的关注。提出了一种基于计算机视觉运动分析的步态识别方法,从图像序列中提取特征进行识别。本方法提取的特征包含步态外观和运动信息。区别于基于模型的步态识别方法,提出一种新的提取肢体角度的方法,该方法通过分析侧影宽度变化获取角度信息,无需对人体建模。首先,采用背景减除技术提取人的侧影;分析侧影的高宽变化获取步态周期。然后,采用离散余弦变换分析了步态外形特征和角度特征。最后运用K邻法在几个不同的数据库上进行了算法验证。实验结果表明,本方法行之有效。

Automatic Human Identification Method Based On Gait Analysis

The gait as one of biometrics has recently gained more and more interests from computer vision researchers.An automatic gait recognition approach for analyzing and classifying human gait was proposed by computer vision techniques.The approach incorporates knowledge of the shape and dynamics of human gait into the feature extraction process.Different from the model-based approaches,a novel method was proposed to attain the limb angle information by analyzing the variation of silhouette width without needing the human body model.First,a background subtraction was used to separate objects from background.Gait cycle was obtained by analyzing the variety of the silhouette width and height.Then,Discrete Cosine transform was used to analyze the shape and leg angle characteristic and represent the gait features.KNN was used to distinguish the different gaits of persons.The performance of the approach was tested using different gait databases.And the approach shows a better recognition rate.