Superpixel Contracted Graph-Based Learning for Hyperspectral Image Classification

Abstract

A central problem in hyperspectral image classification is obtaining high classification accuracy when using a limited amount of labelled data. In this paper we present a novel graph-based framework, which aims to tackle this problem in the presence of large scale data input. Our approach utilises a novel superpixel method, specifically designed for hyperspectral data, to define meaningful local regions in an image, which with high probability share the same classification label. We then extract spectral and spatial features from these regions and use these to produce a contracted weighted graph-representation, where each node represents a region rather than a pixel. Our graph is then fed into a graph-based semi-supervised classifier which gives the final classification. We show that using superpixels in a graph representation is an effective tool for speeding up graphical classifiers applied to hyperspectral images. We demonstrate through exhaustive quantitative and qualitative results that our proposed method produces accurate classifications when an incredibly small amount of labelled data is used. We show that our approach mitigates the major drawbacks of existing approaches, resulting in our approach outperforming several comparative state-of-theart techniques.

Reconstruction enhanced probabilistic model for semisupervised tongue image segmentation

“…With the emerging technique development of deep learning 17 such as convolution neural networks (CNN), there are significant performance improvements for different tasks in the domain of computer vision, for example, image classification, 18,19 object detection, 20,21 and image segmentation. [22][23][24] Different from traditional machine learning methods [25][26][27][28] applied in different domains, [29][30][31]…”

A Novel Spectral–Spatial Classification Method for Hyperspectral Image at Superpixel Level

“…In [40], a set-to-set distance was defined based on an affine hull model and the singular value decomposition. Sellars et al used a combination of Gaussian kernel technique, Log-Euclidean distance of a covariance matrix and Euclidean spectral distance to construct a weight between two connected superpixels [41]. By using a domain transform recursive filtering and k nearest neighbor rule (KNN), Tu et al [47] gave a representation of the distance between superpixels.…”

“…Compared with the similarity between a pair of superpixels introduced in [40] (an affine hull model and the singular value decomposition), the similarity suggested above is easy to understand since only the sorting rule is used. Contrasted to similarity defined in [41,47], our method is simple to calculate, and has no use of parameters. These advantages of the proposal make it easier to be applied in the field of remote sensing.…”

“…Based on an affine hull model and set-to-set distance, Lu et al developed a spectral-spatial HSI classification method at superpixel level [40]. Sellars et al proposed a graph-based semi-supervised classification technique for HSI by defining the weight between two adjacent superpixels [41]. The classification at superpixel level is, to some degree, a reduction of pixels, which may greatly reduce the number of objects in the classification process.…”

Hyperspectral Image Classification With Spectral and Spatial Graph Using Inductive Representation Learning Network

“…It combines the intrinsic information of labeled and unlabeled samples and uses spectral and spatial information to achieve better results. Recently, Sellars et al [40] also propose a graph-based learning method for hyperspectral image classification using superpixels. However, although the GCN can obtain the embedding of the vertices in the graph, it is a direct push learning method, which requires all nodes to participate in training to obtain node embedding.…”