计算机应用研究2012，Vol.29Issue(11)：4338-4341,4.DOI:10.3969/j.issn.1001-3695.2012.11.087
基于监督Isomap的步态识别方法
Method of gait recognition based on supervised Isomap
摘要
Abstract
Because of the complexity of the gait images, many dimension reduction methods can not effectively apply to gait recognition. Isomap is a promising nonlinear dimensionality reduction algorithm. However, in real-world application, Isomap does not use the label information of the observed data, and its generalization ability is weak. Based on Isomap, this paper proposed a novel supervised Isomap applied to gait recognition. The method could not only inherit the advantages of Isomap, but also reduce new samples to low-dimensionality. The experimental results on the three real-world gait databases show that the proposed method is effective and feasible for gait recognition.关键词
步态识别/维数约简/Isomap算法/监督Isomap算法Key words
gait recognition/ dimensional reduction/ Isomap/ supervised Isomap分类
信息技术与安全科学引用本文复制引用
张善文,张传雷..基于监督Isomap的步态识别方法[J].计算机应用研究,2012,29(11):4338-4341,4.基金项目
国家自然科学基金资助项目(60975005) （60975005）
陕西省科技厅自然科学基金资助项目(2011JM8011) （2011JM8011）
