MS<inline-formula><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula>A-Net: Multiscale Spectral&#x2013;Spatial Association Network for Hyperspectral Image Clustering

Abstract

Remote sensing hyperspectral cameras acquire high spectral-resolution data that reveal valuable composition information on the targets (e.g., for Earth observation and environmental applications). The intrinsic high dimensionality and the lack of sufficient numbers of labeled/training samples prevent efficient processing of hyperspectral images (HSIs). HSI clustering can alleviate these limitations. In this study, we propose a multiscale spectral–spatial association network (MS$^{2}$A-Net) to cluster HSIs. The backbone of MS$^{2}$A-Net is an autoencoder architecture that allows the network to capture the nonlinear relation between data points in an unsupervised manner. The network applies a multistream approach. One stream extracts spectral information by deploying a spectral association unit. The other stream derives multiscale contextual and spatial information by employing dilated (atrous) convolutional kernels. The obtained feature representation generated by MS$^{2}$A-Net is fed into a standard k-means clustering algorithm to produce the final clustering result. Extensive experiments on four HSIs for different types of applications (i.e., geological-, rural-, and urban-mapping) demonstrate the superior performance of MS$^{2}$A-Net over the state-of-the-art shallow/deep learning-based clustering approaches in terms of clustering accuracy.

Keywords

• Convolutional autoencoder
• deep learning
• dilated (atrous) convolutions
• hyperspectral imaging
• multiscale information
• remote sensing