摘要
提出了一种应用于人脸识别的监督线性维数约简算法。首先引入图像距离度量方法以确定人脸数据之间的相似程度,之后将训练样本的类标先验信息融入到邻域保持嵌入(NPE,neighborhood preserving embedding)算法的目标函数中,使得降维后的嵌入空间的投影数据呈多流形分布,不仅最优保持了样本空间的局部几何结构,同时各类样本投影的类内散度最小化,类间散度最大化,增大了各类数据分布之间的间隔,提高了嵌入空间的辨别能力。在Extended Yale B和CMU PIE两个开放人脸数据库上进行了识别实验,结果表明,本文算法取得了很好的识别效果。
In order to extract the facial features from face images effectively, a novel supervised linear method of reducing dimensionality is proposed for face recognition. In this study, the concept of image distance is first introduced to measure the similarity between face samples, which enhances the robust- ness to the translation and deformation of the face image. And then the prior class label information of train samples is incorporated into the criterial equation of neighborhood preserving embedding (NPE) al- gorithm which is a manifold learning method developing from the classical algorithm of locality linear embedding (LLE). After optimizing the criterial equation, the distribution of the reduced subspace is made to be the structure of multi-manifold,which not only optimally preserves the local geometry of the original space,but also minimizes the intra-class scatter while maximizes the between-class scatter of the projected data. Thus the discrimination of the embedding is enhanced, and then the recognition rate of the proposed algorithm is improved obviously. Experiments are conduced on the two open face databases, the" Extended Yale B and CMU PIE face databases, and the results show that the proposed method can effec- tively find the key facial features form face images and can achieve better recognition rate compared with other existing ones.
出处
《光电子.激光》
EI
CAS
CSCD
北大核心
2013年第2期365-371,共7页
Journal of Optoelectronics·Laser
基金
国家自然科学基金(61064003)资助项目
关键词
人脸识别
维数约简
图像距离
流形学习
face recognition
dimensionality reduction
image distance
manifold learning