【作者】 王晓梅；

【导师】 王养利；

【作者基本信息】 西安电子科技大学 ， 计算机应用技术， 2006， 硕士

【摘要】 步态识别是指通过人体走路的姿势来识别人的身份。近来年,步态识别作为一种生物特征识别技术而备受关注。步态识别的三大优势:远距离识别,非侵犯性和难于隐藏性,使得它可以广泛地应用到安全部门、身份鉴别、数字监控等领域。步态识别主要由三部分构成:步态序列图像预处理、特征提取和分类识别。本文采用傅立叶描述子,对基于轮廓的步态识别技术做了探索性的研究。首先,采用一种自适应背景模型,实现了动态场景中的背景获取;其次,使用背景差分图像法,结合直方图自动阈值分割和数学形态学算法实现了运动人体检测;利用运动人体的宽高比的周期性变化来提取关键帧,并采用傅立叶描述子对关键帧步态轮廓进行分析和优化,构建特征矢量;然后,对特征矢量进行特征空间变换以获得可分类的低维步态特征;最后,在时空相关性分析的基础上,通过使用标准的模式分类器——最近邻法实现身份识别。本文采用中科院自动化所提供的CASIA步态数据库,对本文中的算法进行了实验,结果表明该算法不仅获得了令人鼓舞的识别性能,而且拥有相对较低的计算代价。更多还原

【Abstract】 Human gait recognition is the process of identifying individuals by their walking manners. In the last years, as one of biometrical features, gait recognition has attracted more and more research interest. Its advantages are that it is a long distant recognition technology, noninvasive and difficult to conceal。Therefore, it can be applied to security system, Human ID management, digital surveillance and so on.. Generally, gait recognition consists of three parts: preprocessing of gait sequence, feature extraction and classification. The goal of this thesis is to explore the gait recognition technology based on body silhouettes by Fourier descriptors.Firstly, An adaptive background model is applied to extracting background in dynamic environment. At the same time the algorithms of automatic histogram thresholding segmentation and morphological operators are used to accomplish the moving person segmentation. For each image sequence, cyclic width of gait analysis is performed to extract key frames, and Fourier descriptor is utilized to describe gait contour. Then eigenspace transformation based on the traditional Principal Component Analysis is applied to time-varying distance signals derived from a sequence of silhouette images to reduce the dimensionality of the input feature space. A supervised pattern classification technique called nearest neighbor algorithm is finally performed in the lower-dimensional eigenspace for recognition. Extensive experimental results on CASIA database demonstrate that the proposed algorithm has an encouraging recognition performance with relatively lower computational cost.更多还原