TensorlinearLaplaciandiscrimination(TLLD)forfeatureextraction

摘要: Discriminant feature extraction plays a central role in pattern recognition and classification. In this paper, we propose the tensor linear Laplacian discrimination (TLLD) algorithm for extracting discriminant features from data. TLLD is an extension of analysis (LDA) (LLD) directions both nonlinear subspace learning representation. Based on contextual distance, weights within-class scatters between-class scatter can be determined to capture principal structure data clusters. This makes free metric sample space, which may not known. Moreover, unlike LLD, parameter tuning very easy. Experimental results face recognition, texture classification handwritten digit show that effective discriminative features.