【作者】 刘翠响； 张艳； 于明； 赵卫萍；

【Author】 LIU Cui-xiang, ZHANG Yan, YU Ming, ZHAO Wei-ping (Department of Information Engineering; School of Computer Science and Engineering, Hebei University of Technology. Tianjin 300130, China.

【机构】 河北工业大学信息工程学院； 河北工业大学计算机软件学院；

【摘要】 针对人脸图像识别,讨论了常规的线性降维算法和新提出的非线性降维算法,并通过大量的实验对不同算法的降维特性进行了分析．实验结果表明,基于Isomap的非线性降维算法的性能优于基于PCA的线性降维算法,采用非线性降维算法可有效地对人脸图像数据进行分析．更多还原

【Abstract】 The algorithms of normal linear dimensionality reduction and the new nonlinear dimensionality reduction are discussed. The algorithms are applied on face image. The effect of dimensionality reduction is analyzed by a large amount of experiments. The Isomap algorithm is better in dimensionality reduction than the PCA. It defectively analyzes to face image data by the new nonlinear dimensionality reduction algorithm.更多还原