Robust Unsupervised Hyperspectral Band Selection via Global Affinity Matrix Reconstruction

Unsupervised band selection is fundamental to alleviate the curse of dimensionality for hyperspectral imagery. Although many researches have been developed, it is still a chanlenging problem to improve the poor classification performance with a small set of selected bands due to the lack of discriminative information that simutaneously distinguishes multiple classes. Moreover, traditional methods utilize the pre-defined graph representation for different HSI datasets that may result in degraded generalization performance. To address this issue, this article proposes an unsupervised band selection method via global affinity matrix reconstruction for hyperspectral image classification, referred to as GAMR briefly. To explore the discriminative information from HSIs, GAMR formulate a convex optimization problem to iteratively refine the randomly generated pseudo-labels. Meanwhile, the similarity between band pairs is estimated by adaptively reconstructing a global affinity matrix from several local graph representations. To solve the proposed formulation, an alternating optimization algorithm is designed to search for the optimal solution. Experimental results on three HSI datasets demonstrate the effectiveness of the proposed method and the superiority over seven state-of-the-art methods. The source code is available athttps://github.com/misteru/BandSelectionGAMR.