Unsupervised Hyperspectral Image Classification Research Articles

Domain-Invariant Feature and Generative Adversarial Network Boundary Enhancement for Multi-Source Unsupervised Hyperspectral Image Classification

Hyperspectral image (HIS) classification, a crucial component of remote sensing technology, is currently challenged by edge ambiguity and the complexities of multi-source domain data. An innovative multi-source unsupervised domain adaptive algorithm (MUDA) structure is proposed in this work to overcome these issues. Our approach incorporates a domain-invariant feature unfolding algorithm, which employs the Fourier transform and Maximum Mean Discrepancy (MMD) distance to maximize invariant feature dispersion. Furthermore, the proposed approach efficiently extracts intraclass and interclass invariant features. Additionally, a boundary-constrained adversarial network generates synthetic samples, reinforcing the source domain feature space boundary and enabling accurate target domain classification during the transfer process. Furthermore, comparative experiments on public benchmark datasets demonstrate the superior performance of our proposed methodology over existing techniques, offering an effective strategy for hyperspectral MUDA.

Max–min distance embedding for unsupervised hyperspectral image classification in the satellite Internet of Things system

In modern society, hyperspectral images have widely been used in many applications. In the earth observation system, hyperspectral images are acquired by hyperspectral sensors and transmitted to the cloud computing center for analysis, which are the part of satellite Internet of Things(IoT) to build a large scale land-cover surveillance system. Thus, during recent years, researchers throughout academia and industry have been advancing the classification of hyperspectral images. Luckily for swamped remote sensing researchers, new technology is allowing brands to extract useful knowledge using artificial intelligence to enhance supervised hyperspectral image classification technology. However, conventional methods that label and process hyperspectral images manually are labor and time-consuming. Hence, in this paper, we proposed an unsupervised hyperspectral image classification framework. The paper differs from the existing methods by three aspects: (1) We adopt multi-scale spatial features to enhance the discriminability of data. (2) We propose deep autoencoder with max–min distance embedding to extract features and reduce dimensions of data, which is more suitable for large-scale data clustering. (3) We apply k-means clustering method to get original classification, then optimize the results by the guided filter. Experimental results on two widely used hyperspectral data, Salinas and Pavia University, show the competitive performance obtained by the proposed framework compared with other approaches.

Superpixel Contracted Neighborhood Contrastive Subspace Clustering Network for Hyperspectral Images

Deep subspace clustering has achieved remarkable performances in the unsupervised classification of hyperspectral images. However, previous models based on pixel-level self-expressiveness of data suffer from the exponential growth of computational complexity and access memory requirements with increasing number of samples, thus leading to poor applicability to large hyperspectral images. This paper presents a Neighborhood Contrastive Subspace Clustering network (NCSC), a scalable and robust deep subspace clustering approach, for unsupervised classification of large hyperspectral images. Instead of using a conventional autoencoder, we devise a novel superpixel pooling autoencoder to learn the superpixel-level latent representation and subspace, allowing a contracted self-expressive layer. To encourage a robust subspace representation, we propose a novel neighborhood contrastive regularization to maximize the agreement between positive samples in subspace. We jointly train the resulting model in an end-to-end fashion by optimizing an adaptively weighted multi-task loss. Extensive experiments on three hyperspectral benchmarks demonstrate the effectiveness of the proposed approach and its substantial advancement of state-of-the-art approaches.

Graph Convolutional Sparse Subspace Coclustering With Nonnegative Orthogonal Factorization for Large Hyperspectral Images

Sparse subspace clustering (SSC) is a representative data clustering paradigm that has been broadly applied in the unsupervised classification of hyperspectral images (HSIs). Existing SSC methods usually produce a subspace affinity matrix between representations of hyperspectral pixels first, followed by spectral clustering for the affinity matrix. To this end, the separated framework fails to exploit the dualities contained in both features and pixels or higher order entities at the same time, and thus, it is difficult to compute coclusters simultaneously. In addition, SSC methods often require expensive computational consumption and memory capacity to approximate the spectral decomposition of the affinity matrix, thus hindering the applicability of SSC methods for large HSIs. To overcome these limitations, we propose a novel graph convolutional sparse subspace coclustering (GCSSC) model with nonnegative orthogonal factorization for large HSIs in which affinity matrix learning and spectral coclustering are integrated into a unified optimizing model to obtain the optimal clustering results. Specifically, to form a more compact self-representation, the superpixel-based adaptive dictionary construction strategy is proposed instead of the global dictionary to precisely represent the pixels. To explore the spatial–contextual and spectral neighboring characteristics between dictionary atoms, graph convolution is incorporated into the dictionary atoms to aggregate the local neighborhood information, and the affinity matrix in the proposed coclustering framework is constructed under a joint sparsity constrained representation model. To reduce high computational consumption and memory capacity, a nonnegative orthogonal factorization constraint is proposed to offer an alternative spectral clustering for hyperspectral pixels and dictionary atoms simultaneously. The clustering performance of the proposed method is evaluated for three classical HSIs, and the experimental results illustrate that the proposed method is memory and computationally efficient and outperforms the state-of-the-art HSI clustering methods.

Efficient subspace clustering of hyperspectral images using similarity-constrained sampling

The unsupervised classification of hyperspectral images (HSIs) draws attention in the remote sensing community due to its inherent complexity and the lack of labeled data. Among unsupervised methods, sparse subspace clustering (SSC) achieves high clustering accuracy by constructing a sparse affinity matrix. However, SSC has limitations when clustering HSI images due to the number of spectral pixels. Specifically, the temporal complexity grows at a cubic ratio of the size of the data, making it inefficient for addressing HSI subspace clustering. We propose an efficient SSC-based method that significantly reduces the temporal and spatial computational complexity by splitting the HSI clustering task using similarity-constrained sampling. Our similarity-constrained sampling strategy considers both edge and superpixel information of the HSI to boost the clustering performance. This sampling strategy enables an intelligent selection of spectral signatures, and then, we split the clustering problem into multiples threads. Experimental results on widely used HSI datasets show that the efficiency of the proposed method outperforms baseline methods by up to 30% in overall accuracy and up to six times in computing time.

Interference-Suppressed and Cluster-Optimized Hyperspectral Target Extraction Based on Density Peak Clustering

Target extraction can provide a prior knowledge for spectral unmixing, unsupervised hyperspectral image classification, and unsupervised target detection tasks, which is of great practice. Considering that the traditional endmember extraction algorithms such as automatic target generation process (ATGP) and simplex growing algorithm (SGA) are sensitive to the burst noise caused by hyperspectral sensor imaging and cannot extract the appropriate target set to simultaneously guide the various tasks, a weighted target extraction algorithm based on the density peak clustering (DPC) is proposed in this article. First, DPC is extended to hyperspectral images, and the decision graph is generated according to the density and distance of each pixel to effectively classify the regions with different attributes such as burst noise, cluster center, cluster boundary, and cluster core. On the basis, three DPC-based weighted vectors are designed, and the weighted target extraction algorithms W-ATGP and W-SGA are put forward which allocates a lower weight than the core pixel to the interference so as to reduce its impact on ATGP and SGA while obtaining the high-quality targets for various tasks. The experimental results on three datasets show that the accuracies of the target sets extracted by W-ATGP and W-SGA are higher than that of ATGP and SGA, no matter as the prior knowledge of spectral unmixing, unsupervised hyperspectral image classification, or unsupervised target detection, which verifies the effectiveness of the proposed methods.

Unsupervised Clustering for Hyperspectral Images

Hyperspectral images are becoming a valuable tool much used in agriculture, mineralogy, and so on. The challenge is to successfully classify the materials founded in the field relevant for different applications. Due to a large amount of data corresponding to a big number of spectral bands, the classification programs require a long time to analyze and classify the data. The purpose is to find a better method for reducing the classification time. We exploit various algorithms on real hyperspectral data sets to find out which algorithm is more effective. This paper presents a comparison of unsupervised hyperspectral image classification such as K-means, Hierarchical clustering, and Parafac decomposition, which allows the performance of the model reduction and feature extraction. The results showed that the method useful for big data is the classification of data after Parafac Decomposition.

Sequential spectral clustering of hyperspectral remote sensing image over bipartite graph

Unsupervised classification is a crucial step in remote sensing hyperspectral image analysis where producing labeled data is a laborious task. Spectral Clustering is an appealing graph-partitioning technique with outstanding performance on data with non-linear dependencies. However, Spectral Clustering is restricted to small-scale data and neither has been effectively applied to hyperspectral image analysis. In this paper, the unsupervised classification of hyperspectral images is addressed through a sequential spectral clustering that can be extended to the large-scale hyperspectral image. To this end, this paper utilizes a bipartite graph representation along with a sequential singular value decomposition and mini-batch K-means for unsupervised classification of hyperspectral imagery. We evaluate the proposed algorithm with several benchmark hyperspectral datasets including Botswana, Salinas, Indian Pines, Pavia Center Scene and Pavia University Scene. The experimental results show significant improvements made by the proposed algorithm compared to the state-of-art clustering algorithms.

Structure Preserving Transfer Learning for Unsupervised Hyperspectral Image Classification

Recent advances on remote sensing techniques allow easier access to imaging spectrometer data. Manually labeling and processing of such collected hyperspectral images (HSIs) with a vast quantities of samples and a large number of bands is labor and time consuming. To relieve these manual processes, machine learning based HSI processing methods have attracted increasing research attention. A major assumption in many machine learning problems is that the training and testing data are in the same feature space and follow the same distribution. However, this assumption doesn’t always hold true in many real world problems, especially in certain HSI processing problems with extremely insufficient or even without training samples. In this letter, we present a transfer learning framework to address this unsupervised challenge (i.e., without training samples in the target domain), by making the following three main contributions: 1) to the best of our knowledge, this is the first time for transfer learning framework to be used for the classification of totally unknown target HSI data with no training samples; 2) the characteristics of HSI are learned on dual spaces to exploit its structure knowledge to better label HSI samples; and 3) two specific new scenarios suitable for transfer learning are investigated. Experimental results on several real world HSIs support the superiority of the proposed work.

Unsupervised Classification in Hyperspectral Imagery With Nonlocal Total Variation and Primal-Dual Hybrid Gradient Algorithm

In this paper, a graph-based nonlocal total variation method (NLTV) is proposed for unsupervised classification of hyperspectral images (HSI). The variational problem is solved by the primal-dual hybrid gradient (PDHG) algorithm. By squaring the labeling function and using a stable simplex clustering routine, an unsupervised clustering method with random initialization can be implemented. The effectiveness of this proposed algorithm is illustrated on both synthetic and real-world HSI, and numerical results show that the proposed algorithm outperforms other standard unsupervised clustering methods such as spherical K-means, nonnegative matrix factorization (NMF), and the graph-based Merriman-Bence-Osher (MBO) scheme.

Hyperspectral Image Classification Using Unsupervised Algorithms

Hyperspectral Imaging (HSI) is a process that results in collected and processed information of the electromagnetic spectrum by a specific sensor device. It’s data provide a wealth of information. This data can be used to address a variety of problems in a number of applications. Hyperspectral Imaging classification assorts all pixels in a digital image into groups. In this paper, unsupervised hyperspectral image classification algorithms used to obtain a classified hyperspectral image. Iterative Self-Organizing Data Analysis Technique Algorithm (ISODATA) algorithm and K-Means algorithm are used. Applying two algorithms on Washington DC hyperspectral image, USA, using ENVI tool. In this paper, the performance was evaluated on the base of the accuracy assessment of the process after applying Principle Component Analysis (PCA) and K-Means or ISODATA algorithm. It is found that, ISODATA algorithm is more accurate than K-Means algorithm. Since The overall accuracy of classification process using K-Means algorithm is 78.3398% and The overall accuracy of classification process using ISODATA algorithm is 81.7696%. Also the processing time increased when the number of iterations increased to get the classified image.

Unsupervised methods for the classification of hyperspectral images with low spatial resolution

The problem of structure detection and unsupervised classification of hyperspectral images with low spatial resolution is addressed in this paper. Hyperspectral imaging is a continuously growing area in remote sensing applications. The wide spectral range, providing a very high spectral resolution, allows the detection and classification surfaces and chemical elements of the observed image. The main problem of hyperspectral images is that the spatial resolution can vary from a few to tens of meters. Many factors, such as imperfect imaging optics, atmospheric scattering, secondary illumination effects and sensor noise cause a degradation of the acquired image quality, making the spatial resolution one of the most expensive and hardest to improve in imaging systems. Due to such a constraint, mixed pixels, e.g., pixels containing mixture of different materials, are quite common in hyperspectral images. In this work, we exploit the rich spectral information of hyperspectral images to deal with the problem. Two methods, based on the concept of spectral unmixing and unsupervised classification, are proposed to obtain thematic maps at a finer spatial scale in a totally unsupervised way. Experiments are carried out on one simulated and two real hyperspectral data sets and clearly show the comparative effectiveness of the proposed method with respect to traditional unsupervised methods, both for classification and detection of spatial structures.