Hyperspectral Image Datasets Research Articles

Hyperspectral Image Classification Based on Hybrid Depth-Wise Separable Convolution and Dual-Branch Feature Fusion Network

Recently, advancements in convolutional neural networks (CNNs) have significantly contributed to the advancement of hyperspectral image (HSI) classification. However, the problem of limited training samples is the primary obstacle to obtaining further improvements in HSI classification. The traditional methods relying solely on 2D-CNN for feature extraction underutilize the inter-band correlations of HSI, while the methods based on 3D-CNN alone for feature extraction lead to an increase in training parameters. To solve the above problems, we propose an HSI classification network based on hybrid depth-wise separable convolution and dual-branch feature fusion (HDCDF). The dual-branch structure is designed in HDCDF to extract simultaneously integrated spectral–spatial features and obtain complementary features via feature fusion. The proposed modules of 2D depth-wise separable convolution attention (2D-DCAttention) block and hybrid residual blocks are applied to the dual branch, respectively, further extracting more representative and comprehensive features. Instead of full 3D convolutions, HDCDF uses hybrid 2D–3D depth-wise separable convolutions, offering computational efficiency. Experiments are conducted on three benchmark HSI datasets: Indian Pines, University of Pavia, and Salinas Valley. The experimental results show that the proposed method showcases superior performance when the training samples are extremely limited, outpacing the state-of-the-art method by an average of 2.03% in the overall accuracy of three datasets, which shows that HDCDF has a certain potential in HSI classification.

TBSSF-Net: three-branch spatial-spectral fusion network for hyperspectral image classification.

Hyperspectral images (HSI) have been extensively applied in a multitude of domains, due to their combined spatial and spectral characteristics along with a wealth of spectral bands. The ingenious combination of spatial and spectral information in HSI classification has remained a central research area for an extended period. In the classification process, it is essential to choose an expanded neighborhood window for learning. Nonetheless, employing an extensive window could lead to the problem of a lack of independence between the training dataset and the test dataset. Hence, this paper puts forward a three-branch spatial-spectral fusion network (TBSSF-Net) for HSI classification based on a smaller patch size. The network is composed of a spatial key details aggregation branch, a spatial semantic knowledge refinement branch, and a spectral band signal granularity branch. By employing the spatial branch, the network not only retains the key characteristics of details within the space but also captures the contextual relationships of global semantic information. The introduction of the spectral branch permits the combination of signal granularity at diverse levels, supplementing the performance of the spectral dimension. The TBSSF-Net has been validated for its superiority and effectiveness on four public HSI datasets. Additionally, it demonstrates significant classification performance across diverse amounts of training sets.

A multi-spectral and hyperspectral image dataset for evaluating chemical traits and the water status of avocado, olive and grape through leaf dehydration under laboratory conditions

Assessing the health status of vegetation is of vital importance for all stakeholders. Multi-spectral and hyper-spectral imaging systems are tools for evaluating the health of vegetation in laboratory settings, and also hold the potential of assessing vegetation of large portions of land. However, the literature lacks benchmark datasets to test algorithms for predicting plant health status, with most researchers creating tailored datasets. This work presents a dataset composed of multi-spectral images, hyper-spectral reflectance values, and measurements of weight, chlorophyll, and nitrogen content of leaves at five different drying stages, from avocado, olive, and grape trees, which are common crops in the Valparaíso region of Chile. This dataset is a valuable asset for developing tools in the field of precision agriculture and assessing the general health status of vegetation.

Vis/NIR Spectroscopy and Vis/NIR Hyperspectral Imaging for Non-Destructive Monitoring of Apricot Fruit Internal Quality with Machine Learning.

The fruit supply chain requires simple, non-destructive, and fast tools for quality evaluation both in the field and during the post-harvest phase. In this study, a portable visible and near-infrared (Vis/NIR) spectrophotometer and a portable Vis/NIR hyperspectral imaging (HSI) device were tested to highlight genetic differences among apricot cultivars, and to develop multi-cultivar and multi-year models for the most important marketable attributes (total soluble solids, TSS; titratable acidity, TA; dry matter, DM). To do this, the fruits of seventeen cultivars from a single experimental orchard harvested at the commercial maturity stage were considered. Spectral data emphasized genetic similarities and differences among the cultivars, capturing changes in the pigment content and macro components of the apricot samples. In recent years, machine learning techniques, such as artificial neural networks (ANNs), have been successfully applied to more efficiently extract valuable information from spectral data and to accurately predict quality traits. In this study, prediction models were developed based on a multilayer perceptron artificial neural network (ANN-MLP) combined with the Levenberg-Marquardt learning algorithm. Regarding the Vis/NIR spectrophotometer dataset, good predictive performances were achieved for TSS (R2 = 0.855) and DM (R2 = 0.857), while the performance for TA was unsatisfactory (R2 = 0.681). In contrast, the optimal predictive ability was found for models of the HSI dataset (TSS: R2 = 0.904; DM: R2 = 0.918, TA: R2 = 0.811), as confirmed by external validation. Moreover, the ANN allowed us to identify the most predictive input spectral regions for each model. The results showed the potential of Vis/NIR spectroscopy as an alternative to traditional destructive methods to monitor the qualitative traits of apricot fruits, reducing the time and costs of analyses.

Black-Grass Monitoring Using Hyperspectral Image Data Is Limited by Between-Site Variability

Many important ecological processes play out over large geographic ranges, and accurate large-scale monitoring of populations is a requirement for their effective management. Of particular interest are agricultural weeds, which cause widespread economic and ecological damage. However, the scale of weed population data collection is limited by an inevitable trade-off between quantity and quality. Remote sensing offers a promising route to the large-scale collection of population state data. However, a key challenge is to collect high enough resolution data and account for between-site variability in environmental (i.e., radiometric) conditions that may make prediction of population states in new data challenging. Here, we use a multi-site hyperspectral image dataset in conjunction with ensemble learning techniques in an attempt to predict densities of an arable weed (Alopecurus myosuroides, Huds) across an agricultural landscape. We demonstrate reasonable predictive performance (using the geometric mean score-GMS) when classifiers are used to predict new data from the same site (GMS = 0.74-low density, GMS = 0.74-medium density, GMS = 0.7-High density). However, even using flexible ensemble techniques to account for variability in spectral data, we show that out-of-field predictive performance is poor (GMS = 0.06-low density, GMS = 0.13-medium density, GMS = 0.08-High density). This study highlights the difficulties in identifying weeds in situ, even using high quality image data from remote sensing.

Unmixing Autoencoder for Image Reconstruction from Hyperspectral Data.

Due to the complexity of samples and the limitations in spatial resolution, the spectra in hyperspectral imaging (HSI) are generally contributed to by multiple components, making univariate analysis ineffective. Although feature extraction methods have been applied, the chemical meaning of the compressed variables is difficult to interpret, limiting their further applications. An unmixing autoencoder (UAE) was developed in this work for the separation of the mixed spectra in HSI. The proposed model is composed of an encoder and a fully connected (FC) layer. The former is used to compress the input spectrum into several variables, and the latter is employed to reconstruct the spectrum. Combining reconstruction loss and sparse regularization, the weights and the spectral profiles of the components will be encoded in the compressed variables and the connection weights of FC, respectively. A simulated and three experimental HSI data sets were adopted to investigate the performance of the UAE model. The spectral components were successfully obtained, from which the handwriting under papers was revealed from the image of near-infrared (NIR) diffusive reflectance spectroscopy, and the images of lipids, proteins, and nucleic acids were reconstructed from the Raman and stimulated Raman scattering (SRS) images.

Deep Learning for Hyperspectral Image Classification: A Critical Evaluation via Mutation Testing

Recently, there has been a surge in the adoption of deep learning (DL) techniques, especially convolutional neural networks (CNNs), to perform hyperspectral image (HSI) classification. Although deep learners have been shown to achieve impressive performance in HSI classification, they are known to be extremely sensitive to even slight perturbations to their inputs and models. When applied in safety-critical applications, it is crucial to know how robust they really are against perturbations. However, there is still limited tool support for DL testing in terms of their robustness, nor are the existing RGB testing approaches able to address the HSI-specific challenges. In this paper, we propose a mutation analysis framework specialized for DL models trained to classify HSIs, which facilitates a critical evaluation of the robustness of DL-based HSI classifiers. First, we introduce a set of mutation operators to inject faults into the inputs and models to simulate distortions of remote sensing HSI classifiers. By utilizing the mutation testing technique, we implement a novel framework which supports the multidimensional evaluation of individual DL-based classifiers. Finally, a comparative study of the robustness of seven popular CNN-based HSI classifiers (i.e., 3D-CNN, FDSSC, HybridSN, MCNN, FC3DCNN, DWTDENSE, and Tri-CNN) on six HSI datasets is provided. Results show that FDSSC and Tri-CNN achieve higher robustness in the presence of distortions, and FDSSC maintains a relatively stable level of robustness even with few training samples. These empirical findings can be partly explained by the characteristics of the classifiers’ architectures. The results substantiate the efficacy of our evaluation framework in assessing the robustness of HSI classifiers and thus confirm its contribution to the field of remote sensing image classification.

Computer-aided endoscopic diagnostic system modified with hyperspectral imaging for the classification of esophageal neoplasms.

The early detection of esophageal cancer is crucial to enhancing patient survival rates, and endoscopy remains the gold standard for identifying esophageal neoplasms. Despite this fact, accurately diagnosing superficial esophageal neoplasms poses a challenge, even for seasoned endoscopists. Recent advancements in computer-aided diagnostic systems, empowered by artificial intelligence (AI), have shown promising results in elevating the diagnostic precision for early-stage esophageal cancer. In this study, we expanded upon traditional red-green-blue (RGB) imaging by integrating the YOLO neural network algorithm with hyperspectral imaging (HSI) to evaluate the diagnostic efficacy of this innovative AI system for superficial esophageal neoplasms. A total of 1836 endoscopic images were utilized for model training, which included 858 white-light imaging (WLI) and 978 narrow-band imaging (NBI) samples. These images were categorized into three groups, namely, normal esophagus, esophageal squamous dysplasia, and esophageal squamous cell carcinoma (SCC). An additional set comprising 257 WLI and 267 NBI images served as the validation dataset to assess diagnostic accuracy. Within the RGB dataset, the diagnostic accuracies of the WLI and NBI systems for classifying images into normal, dysplasia, and SCC categories were 0.83 and 0.82, respectively. Conversely, the HSI dataset yielded higher diagnostic accuracies for the WLI and NBI systems, with scores of 0.90 and 0.89, respectively. The HSI dataset outperformed the RGB dataset, demonstrating an overall diagnostic accuracy improvement of 8%. Our findings underscored the advantageous impact of incorporating the HSI dataset in model training. Furthermore, the application of HSI in AI-driven image recognition algorithms significantly enhanced the diagnostic accuracy for early esophageal cancer.

SSTF-Unet: Spatial-Spectral Transformer-Based U-Net for High-Resolution Hyperspectral Image Acquisition.

To obtain a high-resolution hyperspectral image (HR-HSI), fusing a low-resolution hyperspectral image (LR-HSI) and a high-resolution multispectral image (HR-MSI) is a prominent approach. Numerous approaches based on convolutional neural networks (CNNs) have been presented for hyperspectral image (HSI) and multispectral image (MSI) fusion. Nevertheless, these CNN-based methods may ignore the global relevant features from the input image due to the geometric limitations of convolutional kernels. To obtain more accurate fusion results, we provide a spatial-spectral transformer-based U-net (SSTF-Unet). Our SSTF-Unet can capture the association between distant features and explore the intrinsic information of images. More specifically, we use the spatial transformer block (SATB) and spectral transformer block (SETB) to calculate the spatial and spectral self-attention, respectively. Then, SATB and SETB are connected in parallel to form the spatial-spectral fusion block (SSFB). Inspired by the U-net architecture, we build up our SSTF-Unet through stacking several SSFBs for multiscale spatial-spectral feature fusion. Experimental results on public HSI datasets demonstrate that the designed SSTF-Unet achieves better performance than other existing HSI and MSI fusion approaches.

3D‐TabNetHS: A hyperspectral image classification method based on improved interpretable 3D attentive TabNet

AbstractThe classification methods for hyperspectral images (HSI) based on decision trees and convolutional neural networks have shown increasing advantages, but these methods often require a large number of labelled samples for learning, which is difficult for HSI, and the interpretability of the network is not high. Therefore, this paper proposes classification methods based on improved attention interpretable table learning (TabNet) named 3D TabNet HSI (3D‐TabNetHS) and unsupervised 3D TabNet HSI (U3D‐TabNetHS). These methods use sequential attention to select appropriate HSI spatial‐spectral features and add a space spectral information extraction (SSE) module composed of a 3D convolutional neural network (3D‐CNN) and fully connected layers to the Attention Transformer module in the original TabNet network to extract spatial‐spectral soft features. At the same time, unsupervised learning can be used to retrain the 3D‐TabNetHS network, and the classification accuracy of the resulting U3D‐TabNetHS network can be further improved. Compared with other HSI classification methods based on decision trees, the HSI classification accuracy of 3D‐TabNetHS is higher. On three typical HSI datasets, the accuracy metric overall accuracy of 3D‐TabNetHS reached as high as 98.71%, 94.73%, and 97.23%, respectively. Simultaneously, the consistency evaluation metric Kappa also reached 98.56%, 93.98%, and 96.31% respectively. The experimental results indicate the feasibility and reliability of the proposed method in HSI classification.

Rapid and non-destructive classification of salinity levels in brined kimchi cabbage using hyperspectral imaging

In this study, we demonstrate the potential of a non-destructive hyperspectral imaging processing method in the near-infrared (NIR) region (874–1734 nm) for classifying the quality of brined kimchi cabbage. The salinity level of brined kimchi cabbage is an important indicator of consumer preference and the quality of kimchi. Hence, we compared the water content and salinity of brined kimchi cabbage via hyperspectral data. We extracted the optimal wavelengths from the hyperspectral image dataset to classify the salinity level of the predicted brined kimchi cabbage, and thus, established a novel approach for classifying kimchi samples into quality-unacceptable and quality-acceptable groups. Standard normal variate and multiplicative scatter correction (MSC) were used for pathlength correction. The Savitzky–Golay first and second derivatives were used for the deconvolution of the raw spectral data. The experimental results confirmed that the decision tree model combined with MSC pathlength correction and Savitzky–Golay first derivative preprocessing was the best classification model. The proposed hyperspectral image–NIR system can be applied to the detection of salinity during industrial kimchi manufacturing.

Domain-Invariant Few-Shot Contrastive Learning for Hyperspectral Image Classification

In Hyperspectral Image (HSI) classification, acquiring large quantities of high-quality labeled samples is typically costly and impractical. Traditional deep learning methods are limited in such scenarios due to their dependence on sample quantities. To address this challenge, researchers have turned to Few-Shot Learning (FSL). Although existing FSL methods improve classification performance by enhancing domain invariance through domain adaptation, they often overlook the critical issue of high inter-class similarity and large intra-class variability. Moreover, during domain alignment, features of different categories within the same domain may become confused. To address these issues, this paper proposes a novel Domain-Invariant Few-Shot Contrastive Learning (DIFSCL) method, which combines domain adaptation and contrastive learning strategies to not only learn domain-invariant features but also significantly enhance inter-class discriminability. Based on this, we further design a multi-scale adaptive attention mechanism for a hyperspectral feature extraction network to more effectively extract and optimize generalized features within the DIFSCL framework, significantly improving intra-class consistency and inter-class discriminative capability of features. Experimental results on three widely used HSI datasets demonstrate that our method significantly outperforms existing techniques in few-shot classification tasks.

Brain tissue classification in hyperspectral images using multistage diffusion features and transformer.

Brain surgery is a widely practised and effective treatment for brain tumours, but accurately identifying and classifying tumour boundaries is crucial to maximise resection and avoid neurological complications. This precision in classification is essential for guiding surgical decisions and subsequent treatment planning. Hyperspectral (HS) imaging (HSI) is an emerging multidimensional optical imaging method that captures detailed spectral information across multiple wavelengths, allowing for the identification of nuanced differences in tissue composition, with the potential to enhance intraoperative tissue classification. However, current frameworks often require retraining models for each HSI to extract meaningful features, resulting in long processing times and high computational costs. Additionally, most methods utilise the deep semantic features at the end of the network for classification, ignoring the spatial details contained in the shallow features. To overcome these challenges, we propose a novel approach called MedDiffHSI, which combines diffusion and transformer techniques. Our method involves training an unsupervised learning framework based on the diffusion model to extract high-level and low-level spectral-spatial features from HSI. This approach eliminates the need for retraining of spectral-spatial feature learning model, thereby reducing time complexity. We then extract intermediate multistage features from different timestamps for classification using a pretrained denoising U-Net. To fully explore and exploit the rich contextual semantics and textual information hidden in the extracted diffusion feature, we utilise a spectral-spatial attention module. This module not only learns multistage information about features at different depths, but also extracts and enhances effective information from them. Finally, we employ a supervised transformer-based classifier with weighted majority voting (WMV) to perform the HSI classification. To validate our approach, we conduct comprehensive experiments on in vivo brain database data sets and also extend the analysis to include additional HSI data sets for breast cancer to evaluate the framework performance across different types of tissue. The results demonstrate that our framework outperforms existing approaches by using minimal training samples (5%) while achieving state-of-the-artperformance.

Joint Sparse Local Linear Discriminant Analysis for Feature Dimensionality Reduction of Hyperspectral Images

Although linear discriminant analysis (LDA)-based subspace learning has been widely applied to hyperspectral image (HSI) classification, the existing LDA-based subspace learning methods exhibit several limitations: (1) They are often sensitive to noise and demonstrate weak robustness; (2) these methods ignore the local information inherent in data; and (3) the number of extracted features is restricted by the number of classes. To address these drawbacks, this paper proposes a novel joint sparse local linear discriminant analysis (JSLLDA) method by integrating embedding regression and locality-preserving regularization into the LDA model for feature dimensionality reduction of HSIs. In JSLLDA, a row-sparse projection matrix can be learned, to uncover the joint sparse structure information of data by imposing a L2,1-norm constraint. The L2,1-norm is also employed to measure the embedding regression reconstruction error, thereby mitigating the effects of noise and occlusions. A locality preservation term is incorporated to fully leverage the local geometric structural information of the data, enhancing the discriminability of the learned projection. Furthermore, an orthogonal matrix is introduced to alleviate the limitation on the number of acquired features. Finally, extensive experiments conducted on three hyperspectral image (HSI) datasets demonstrated that the performance of JSLLDA surpassed that of some related state-of-the-art dimensionality reduction methods.

Discrete Cosine Transform-Based Joint Spectral–Spatial Information Compression and Band-Correlation Calculation for Hyperspectral Feature Extraction

Prediction tasks over pixels in hyperspectral images (HSI) require careful effort to engineer the features used for learning a classifier. However, the generated classification map may suffer from an over-smoothing problem, which is manifested in significant differences from the original image in terms of object boundaries and details. To address this over-smoothing problem, we designed a method for extracting spectral–spatial-band-correlation (SSBC) features. In SSBC features, joint spectral–spatial feature extraction is considered a discrete cosine transform-based information compression, where a flattening operation is used to avoid the high computational cost induced by the requirement of distillation from 3D images for joint spectral–spatial information. However, this process can yield extracted features with lost spectral information. We argue that increasing the spectral information in the extracted features is the key to addressing the over-smoothing problem in the classification map. Consequently, the normalized difference vegetation index and iron oxide are improved for HSI data in extracting band-correlation features as added spectral information because their calculations, involving two spectral bands, are not appropriate for the abundant spectral bands of HSI. Experimental results on four real HSI datasets show that the proposed features can significantly mitigate the over-smoothing problem, and the classification performance is comparable to that of state-of-the-art deep features.

Adaptive selection of spectral–spatial features for hyperspectral image classification using a modified-CBAM-based network

Convolutional neural networks (CNNs) have demonstrated strong capabilities in hyperspectral image (HSI) classification. However, it is still a challenge to adaptively adjust the size of the receptive fields (RFs) of CNNs base on the information of different scales in HSI to achieve adaptive selection of spectral–spatial features. In the paper, we modify the convolutional block attention module (CBAM) and propose a modified-CBAM-based network (MCNet) to adaptively select spectral–spatial features for HSI classification. In particular, the modified CBAM not only enables the model to adjust its RF size according to the information of different scales in HSI, but also enables the model to achieve a joint focus on important spectral and spatial features. This is very important to adaptively select more descriptive and discriminative spectral–spatial features. The proposed MCNet is compared with currently popular methods on Indian Pines, Kennedy Space Center, University of Pavia, and Botswana HSI datasets. The results show that MCNet has better classification results than other methods on overall accuracy, average accuracy, and Kappa.

Proto-DS: A Self-Supervised Learning-Based Nondestructive Testing Approach for Food Adulteration with Imbalanced Hyperspectral Data.

Conventional food fraud detection using hyperspectral imaging (HSI) relies on the discriminative power of machine learning. However, these approaches often assume a balanced class distribution in an ideal laboratory environment, which is impractical in real-world scenarios with diverse label distributions. This results in suboptimal performance when less frequent classes are overshadowed by the majority class during training. Thus, the critical research challenge emerges of how to develop an effective classifier on a small-scale imbalanced dataset without significant bias from the dominant class. In this paper, we propose a novel nondestructive detection approach, which we call the Dice Loss Improved Self-Supervised Learning-Based Prototypical Network (Proto-DS), designed to address this imbalanced learning challenge. The proposed amalgamation mitigates the label bias on the most frequent class, further improving robustness. We validate our proposed method on three collected hyperspectral food image datasets with varying degrees of data imbalance: Citri Reticulatae Pericarpium (Chenpi), Chinese herbs, and coffee beans. Comparisons with state-of-the-art imbalanced learning techniques, including the Synthetic Minority Oversampling Technique (SMOTE) and class-importance reweighting, reveal our method's superiority. Notably, our experiments demonstrate that Proto-DS consistently outperforms conventional approaches, achieving the best average balanced accuracy of 88.18% across various training sample sizes, whereas the Logistic Model Tree (LMT), Multi-Layer Perceptron (MLP), and Convolutional Neural Network (CNN) approaches attain only 59.42%, 60.38%, and 66.34%, respectively. Overall, self-supervised learning is key to improving imbalanced learning performance and outperforms related approaches, while both prototypical networks and the Dice loss can further enhance classification performance. Intriguingly, self-supervised learning can provide complementary information to existing imbalanced learning approaches. Combining these approaches may serve as a potential solution for building effective models with limited training data.

Tensor-Based Few-Shot Learning for Cross-Domain Hyperspectral Image Classification

Few-shot learning (FSL) is an effective solution for cross-domain hyperspectral image (HSI) classification, which could address the limited labeled samples of the target domain. Current FSL methods mostly utilize the 3D-CNN to transform the spatial and spectral information into a single feature to model an HSI, which means that spatial and spectral information are treated equally in the feature-modeling process. However, spectral information is considered to be more domain-invariant than spatial information. Treating the spatial and spectral information equally may result in parameter redundancy and undesirable cross-domain classification performance. In this paper, we attempt to use tensor mathematics for modeling the HSI and propose a novel few-shot learning method, called tensor-based few-shot Learning (TFSL) for cross-domain HSI classification, which aims to guide the model to focus on the extraction of domain-invariant spectral dependencies. Specifically, we first propose a spatial–spectral tensor decomposition (SSTD) model to provide a mathematical explanation of the input HSI, representing the local spatial–spectral information as 1D and 2D local tensors to reduce the data redundancy. Additionally, a tensor-based hybrid two-stream (THT) model is proposed for extracting the domain-invariant spatial–spectral tensor feature by using 1D-CNN and 2D-CNN. Furthermore, we adopt a 1D-CNN tensor feature enhancement block to enhance the spectral feature of hybrid two-stream tensors and guide the THT model to concentrate on the modeling of spectral dependencies. Finally, the proposed TFSL is evaluated on four public HSI datasets, and the extensive experimental results demonstrate that the proposed TFSL significantly outperforms other advanced FSL methods.

Segmentation-based truncated-SVD for effective feature extraction in hyperspectral image classification

ABSTRACT Remote sensing hyperspectral images (HSIs) are rich sources of information about land cover captured across hundreds of narrow, contiguous spectral wavelength bands. However, using the entire original HSI for practical applications can lead to suboptimal classification accuracy. To address this, band reduction techniques, categorized as feature extraction and feature selection methods, are employed to enhance classification results. One commonly used feature extraction approach for HSIs is Principal Component Analysis (PCA). However, PCA may fall short of capturing the local and specific characteristics present in the HSI data. In this paper, we introduce two novel feature extraction methods: Segmented Truncated Singular Value Decomposition (STSVD) and Spectrally Segmented Truncated Singular Value Decomposition (SSTSVD) to improve classification performance. Segmentation is carried out based on highly correlated bands’ segments and spectral bands’ segments within the HSI data. Our study evaluates and compares these newly proposed methods against classical feature extraction methods, including PCA, Incremental PCA, Sparse-PCA, Kernel PCA, Segmented-PCA (SPCA), and Truncated Singular Value Decomposition (TSVD). We perform this analysis on three distinct HSI datasets, namely the Indian Pines HSI, the Pavia University HSI, and the Kennedy Space Center HSI, using per-pixel Support Vector Machine (SVM) and Random Forest (RF) classification. The experimental results demonstrate the superiority of our proposed methods for all three datasets. The best-performing feature extraction methods when classification is performed using an SVM classifier are STSVD3 (89.03%), SSTSVD2 (95.55%), and STSVD3 (97.74%) for the Indian Pines, Pavia University, and Kennedy Space Center datasets, respectively. Similarly, for the RF classifier, the best-performing feature extraction methods are SSTSVD4 (88.98%), SSTSVD3 (96.04%), and SSTSVD4 (96.09%) for Indian Pines, Pavia University, and Kennedy Space Center datasets, respectively.

Euler Kernel Mapping for Hyperspectral Image Clustering via Self-Paced Learning

Clustering, as a classical unsupervised artificial intelligence technology, is commonly employed for hyperspectral image clustering tasks. However, most existing clustering methods designed for remote sensing tasks aim to solve a non-convex objective function, which can be optimized iteratively, beginning with random initializations. Consequently, during the learning phase of the clustering model, it may easily fall into bad local optimal solutions and finally hurt the clustering performance. Additionally, prevailing approaches often exhibit limitations in capturing the intricate structures inherent in hyperspectral images and are very sensitive to noise and outliers that widely exist in remote sensing data. To address these issues, we proposed a novel Euler kernel mapping for hyperspectral image clustering via self-paced learning (EKM-SPL). EKM-SPL first employs self-paced learning to learn the clustering model in a meaningful order by progressing samples from easy to complex, which can help to remove bad local optimal solutions. Secondly, a probabilistic soft weighting scheme is employed to measure complexity across the data sample, which makes the optimization process more reasonable. Thirdly, in order to more accurately characterize the intricate structure of hyperspectral images, Euler kernel mapping is used to convert the original data into a reproduced kernel Hilbert space, where the nonlinearly inseparable clusters may become linearly separable. Moreover, we innovatively integrate the coordinate descent technique into the optimization algorithm to circumvent the computational inefficiencies and information loss typically associated with conventional kernel methods. Extensive experiments conducted on classic benchmark hyperspectral image datasets illustrate the effectiveness and superiority of our proposed model.