改进的保持邻域嵌入人脸识别方法

为进一步提高保持邻域嵌入算法在人脸识别中的识别性能，提出一种改进的保持邻域嵌入人脸识别方法LDNPE。利用先验的类标签信息构造权重矩阵，按照线性鉴别的思想把类间散布矩阵嵌入到目标函数中，增加样本类间散布约束，基于修改后的目标函数得到最优变换矩阵，并用最近距离分类器分类。在CAS-PEAL和FERET人脸数据库上的实验结果表明该算法的有效性。

Improved Neighborhood Preserving Embedding Face Recognition Method

In order to improve the recognition performance of neighborhood preserving embedding, this paper proposes a face recognition algorithm based on modified neighborhood preserving embedding LDNPE. Using the priori classification information, the weight value matrix of original Neighborhood Preserving Embedding(NPE) is modified. Subsequently, the objective function of NPE is also changed by combining the between-class scatter matrix. Based on the modified NPE, the transformation matrix can be obtained. As a result, the features of training samples and testing sample are got by projecting them on the subspace spanned by optimal base vectors. Nearest Neighborhood(NN) algorithm is used to construct classifiers. Several experiments are conducted on CAS-PEAL and FERET face database. Results demonstrate the effectiveness of the proposed method.