New face recognition algorithm based on nonlinear dimensionality reduction
基于非线性降维的人脸识别新算法

A new supervised kernel subspace method for face recognition was proposed. It makes a better use of samples' label information by introducing supervision in the construction of local neighborhood so as to improve the recognition efficiency; and a positive regularization was added in the gaining of the optimal reconstruction weight matrix to make it insensitive to the noise. In experimental phase, the AT&T and Yale face databases were used to test this algorithm by using the Nearest Neighborhood (NN) algorit...