Semi-supervised enhanced discriminative local constraint preserving projection for dimensionality reduction of medical hyperspectral images

Hongmin Gao, Mengran Yang, Xueying Cao, Qin Liu, Peipei Xu

Abstract: Highlights•A tensor theory and semi-supervised idea inspired EDLCPP DR method is proposed.•An attention mechanism is suggested to enhance the efficiency of the DR.•Select and fully leverage the sample with high discriminability.•EDLCPP overcomes the underutilization of discriminant information in existing methods.