Semi-supervised Hyperspectral Image Classification with Graph Clustering Convolutional Networks
This work provides an incremental improvement for hyperspectral image classification, a task relevant to remote sensing and environmental monitoring, by enhancing GCN-based methods to better handle data scarcity and computational efficiency.
The paper addresses the challenge of hyperspectral image classification (HIC) with limited labeled data by proposing a graph convolutional network (GCN) framework. This framework employs two clustering operations: first, grouping pixels into superpixels, and second, partitioning the superpixel graph into sub-graphs by pruning weak edges. This approach aims to better exploit multi-hop node correlations and reduce computational burden.
Hyperspectral image classification (HIC) is an important but challenging task, and a problem that limits the algorithmic development in this field is that the ground truths of hyperspectral images (HSIs) are extremely hard to obtain. Recently a handful of HIC methods are developed based on the graph convolution networks (GCNs), which effectively relieves the scarcity of labeled data for deep learning based HIC methods. To further lift the classification performance, in this work we propose a graph convolution network (GCN) based framework for HSI classification that uses two clustering operations to better exploit multi-hop node correlations and also effectively reduce graph size. In particular, we first cluster the pixels with similar spectral features into a superpixel and build the graph based on the superpixels of the input HSI. Then instead of performing convolution over this superpixel graph, we further partition it into several sub-graphs by pruning the edges with weak weights, so as to strengthen the correlations of nodes with high similarity. This second round of clustering also further reduces the graph size, thus reducing the computation burden of graph convolution. Experimental results on three widely used benchmark datasets well prove the effectiveness of our proposed framework.