HyperKD: Lifelong Hyperspectral Image Classification With Cross-Spectral–Spatial Knowledge Distillation

Hyperspectral image (HSI) classification has been extensively applied for analyzing remotely sensed images. HSI data consist of multiple bands that provide abundant spatial information. Convolutional neural networks (CNNs) have emerged as powerful deep learning methods for processing visual data. In recent work, CNNs have shown impressive results in HSI classification. In this paper, we propose a hierarchical neural network architecture called feature extraction with hybrid spectral CNN (FE-HybridSN) to extract superior spectral–spatial features. FE-HybridSN effectively captures more spectral–spatial information while reducing computational complexity. Considering the prevalent issue of class imbalance in experimental datasets (IP, UP, SV) and real-world hyperspectral datasets, we apply the focal loss to mitigate these problems. The focal loss reconstructs the loss function and facilitates effective achievement of the aforementioned goals. We propose a framework (FEHN-FL) that combines FE-HybridSN and the focal loss for HSI classification and then conduct extensive HSI classification experiments using three remote sensing datasets: Indian Pines (IP), University of Pavia (UP), and Salinas Scene (SV). Using cross-entropy loss as a baseline, we assess the hyperspectral classification performance of various backbone networks and examine the influence of different spatial sizes on classification accuracy. After incorporating focal loss as our loss function, we not only compare the classification performance of the FE-HybridSN backbone network under different loss functions but also evaluate their convergence rates during training. The proposed classification framework demonstrates satisfactory performance compared to state-of-the-art end-to-end deep-learning-based methods, such as 2D-CNN, 3D-CNN, etc.

Hyperspectral image (HSI) classification has many applications in different diverse research fields. We propose a method for HSI classification using principal component analysis (PCA), 2D spatial convolution, and support vector machine (SVM). Our method takes advantage of correlation in both spatial and spectral domains in an HSI data cube at the same time. We use PCA to reduce the dimensionality of an HSI data cube. We then perform spatial convolution to the dimension-reduced data cube once and then to the convolved data cube for the second time. As a result, we have generated two convolved PCA output data cubes in a multiresolution way. We feed the two convolved data cubes to SVM to classify each pixel to one of the known classes. Experiments on three widely used hyperspectral data cubes (i.e., Indian Pines, Pavia University, and Salinas) demonstrate that our method can improve the classification accuracy significantly when compared to a few existing methods. Our method is relatively fast in terms of central processing unit computational time as well.

Hyperspectral images (HSI) offer detailed spectral reflectance information about sensed objects through provision of information on hundreds of narrow spectral bands. HSI have a leading role in a broad range of applications, such as in forestry, agriculture, geology, and environmental sciences. The monitoring and management of agricultural lands is of great importance for meeting the nutritional and other needs of a rapidly and continuously increasing world population. In relation to this, classification of HSI is an effective way for creating land use and land cover maps quickly and accurately. In recent years, classification of HSI using convolutional neural networks (CNN), which is a sub-field of deep learning, has become a very popular research topic and several CNN architectures have been developed by researchers. The aim of this study was to investigate the classification performance of CNN model on agricultural HSI scenes. For this purpose, a 3D-2D CNN framework and a well-known support vector machine (SVM) model were compared using the Indian Pines and Salinas Scene datasets that contain crop and mixed vegetation classes. As a result of this study, it was confirmed that use of 3D-2D CNN offers superior performance for classifying agricultural HSI datasets.

In recent studies, superpixel segmentation has been integrated into hyperspectral (HS) image classification methods. However, the existing superpixel-based classification methods usually suffer from two serious problems. First, the accuracy and efficiency of current superpixel segmentation approaches cannot meet the demands of practical applications for HS images; second, conventional superpixel-based classification methods generally consider each generated superpixel as a unit for the image classification, which may help to reduce the computing time but result in a significant decrease of the classification accuracy. To solve the problems, we propose a fast region growing based superpixel segmentation (FRGSS) algorithm and a novel texture-adaptive superpixel integration strategy (TASIS) for the HS image classification. Experimental results on real Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) HS images demonstrate that the proposed FRGSS outperforms the state-of-the-art superpixel algorithm. In addition, the superiority of the TASIS is verified compared to the pixel-wise and the conventional superpixel-based classification methods.

Hyperspectral Image (HSI) classification plays a key role remote sensing field. Recently, deep learning has demonstrated its effectiveness in HSI Classification field. This paper presents a spectral-spatial HSI classification technique established on the deep learning based deep belief network (DBN) for deep and abstract feature extraction and adaptive boundary adjustment based segmentation. Proposed approach focuses on integrating the deep learning based spectral features and segmentation based spatial features into a framework for improved performance. Specifically, first the deep DBN model is exploited as a spectral feature extraction based classifier to extract the deep spectral features. Second, spatial contextual features are obtained by utilizing effective adaptive boundary adjustment based segmentation technique. Finally, maximum voting based criteria is operated to integrate the results of extracted spectral and spatial information for improved HSI classification. In general, exploiting spectral features from DBN process and spatial features from segmentation and integration of spectral and spatial information by maximum voting based criteria, has a substantial effect on the performance of HSI classification. Experimental performance on real and widely used hyperspectral data sets with different contexts and resolutions demonstrates the accuracy of the proposed technique and performance is comparable to several recently proposed HSI classification techniques.

Hyperspectral image (HSI) classification aiming at identifying the pixel-level attributes, is a challenging task in remote sensing image interpretation. Informative features play an important role in HSI classification. Existing attention-based convolutional neural networks treat each convolutional layer as a separate process that miss the correlation among different layers. To slove the problem, we introduce the 3D calibrated attention module (3DCAM) to model the holistic dependencies among channels and positions. In particular, The 3DCAM can be used for the correlation among channels to achieve more powerful feature representation and correlation learning. Furthermore, we design a 3D calibrated attention-based network, which integrates 3D-CNN and 2D-CNN layers. Both the spectral and spatial feature are employed for classification. Extensive experiments on the Indian Pines and Pavia University datasets demonstrate that the proposed method performs favorably against three closely related methods.

A spatial-spectral metric learning (SSML) framework for hyperspectral image (HSI) classification is proposed. SSML learns a metric by considering both the spectral characteristics and spatial features represented as the mean of neighboring pixels. It first performs the local pixel neighborhood preserving embedding (LPNPE) to reduce the dimensionality of HSI and meanwhile to preserve the spatial local similarity structure. Then, it learns a spectral and spatial distance metric, separately. Finally, the combination of the spectral and spatial metrics yields a joint spatial-spectral metric. It is followed by a nearest neighbor (NN) classifier for HSI classification. SSML shows good performance over the spectral and spatial NN and SVM on the benchmark hyperspectral data set of Indian Pines.

Most hyperspectral image (HSI) classification methods assume that all classes in the test set are present during training. However, in real-world applications, acquiring labeled training samples is challenging. As a result, it is difficult for the training dataset to cover all possible land cover types, leading to the generalized zero-shot learning (GZSL) problem. Recently, vision-language models (VLMs) have provided rich semantic priors for land cover classes, offering promising potential for GZSL. However, two fundamental gaps hinder their application to HSI classification: the task paradigm gap, arising from the difference between image-level VLMs and the pixel-level HSI classification task; and the knowledge gap, due to the inconsistency between VLM features and HSI spectral–spatial representations. To bridge both gaps, a novel framework leveraging VLM semantic priors for GZSL in HSI classification is proposed, primarily using pseudo-labeling technique to provide knowledge for unseen classes. Specifically, a pseudo-label generation and enhancement module enables a paradigm transition from image-level understanding to pixel-level classification by incorporating HSI’s spatial information. A pseudo-label correction module then refines noisy labels using spectral cues to address the knowledge gap. Finally, a global learning strategy integrates pseudo-label distillation, supervised learning, and feature regularization to classify seen classes while enabling generalization to unseen ones. Experiments on benchmark HSI datasets demonstrate the proposed method’s superiority in generalized zero-shot classification. This work highlights the potential of VLMs in advancing HSI classification in practical applications.

Hyperspectral image (HSI) classification is one of the growing research areas in Remote Sensing. The high dimensionality of the data cube and a high correlation among the hyperspectral bands, poses a difficulty in HSI processing, and selection is one of the promising solutions to address the issue. In this research paper, a whale optimization-based band selection is proposed. Initially, the informative bands are retrieved via a search mechanism which resembles the hunting behaviour of the humpback whales. Later, a hybrid filter is proposed to extract the intrinsic and edge preserving spatial features. Finally, the smoothened bands are trained using a non-linear support vector machine for efficient classification. The exploitation and exploration in the local and global search for the selection of whale hyperspectral bands from the search space proved to be useful in achieving the high-quality classification maps. The overall accuracy reported on the three benchmark data sets – Indian Pines, University of Pavia and Salinas are 99.05, 97.88, and 97.56%, respectively. The proposed method accuracy is achieved with 5% fewer bands on Indian Pines, 70% on University of Pavia and 50% on Salinas Datasets over the other competing methods, which show the effectiveness of the whale optimization-based band selection.

Dual selective fusion transformer network for hyperspectral image classification.

Hyperspectral image (HSI) contain a rich spectrum of bands, enabling a nuanced understanding of observed scenes. As one primary goal, HSI classification mainly involves identifying subtle material variations pixel-wise using the spectral information. This paper addresses the critical challenge of HSI classification under spatial fragmentation, a scenario where conventional methods that rely on fixed spatial arrangements often fail. In brief, we first introduce the Patch-Level Spatial Perturbation (PLSP) strategy to fragment the HSI data. Then, we propose WESA, a novel Wavelet-Enhanced Spatial-Agnostic Framework, which is designed to be inherently robust to spatial disorder by integrating two key components: Wavelet Multi-Scale Analysis (WMSA) and Frequency-Spatial Aware Knowledge Distillation (FSA-KD). WMSA enhances discriminative feature learning by decomposing inputs into multi-frequency subbands, capturing stable features that are less dependent on spatial location. FSA-KD bridges the domain gap between spatially fragmented and intact data via knowledge distillation, enabling the model to recover crucial spatial-spectral insights. Evaluated on four diverse datasets, WESA demonstrates exceptional robustness. For instance, on the Indian Pines dataset, WESA provides an average OA improvement of 14.88% over baseline models under extreme <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$1\times 1$</tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$5\times 5$</tex-math></inline-formula> patch fragmentation. Ultimately, this work offers a powerful spatial-agnostic solution for HSI classification in disordered environments, with practical implications for agricultural monitoring, urban mapping, and dynamic land cover analysis following sudden natural disasters.

Hyper spectral image (HSI) Classification has become important research areas of remote sensing which can be used in many practical applications, including precision agriculture, Land cover mapping, environmental monitoring etc. HSI Classification includes various steps like Noise removal, dimensionality reduction, and classification. In this work, we adopted structure-preserving recursive filter (SPRF) to noise removal and Probabilistic based principal component analysis (PPCA) is applied to reduce dimensionality. Finally classification is performed using multi class large marginal distribution machine (LDM). The proposed (HSI) Classification method is carried out and results are validated across the three widely used standard datasets like Indian Pines, University of Pavia and Salinas. The obtained results show that the proposed method provides results on par with similar type of methods from literature.

Hyperspectral image classification via nonlocal joint kernel sparse representation based on local covariance

Hyperspectral image (HSI) classification is an important technology for hyperspectral data analysis, which has the challenge of high dimensionality and limited training samples. Moreover, due to the spatial continuity of material distribution, it is reasonable to improve the performance of HSI classification from the local similarity viewpoint, i.e., pixels in the same local region have a high probability to share similar features. Therefore, in this paper, a novel method based on local similarity and kernel low-rank representation is proposed to improve the classification accuracy for HSI. First, taking into consideration that pixels in the same homogeneous region share similar spectral features (termed local similarity), the HSI is segmented by the superpixel segmentation algorithm to obtain the homogeneous regions. Then, a model of kernel low-rank representation based on local similarity (KLRRLS) is proposed for HSI classification. In the model of KLRRLS, the nonlinear structure characteristics of HSI are extracted by employing the kernel trick, and the low-rank constraint is used as a prior to restrict the recovery coefficient matrix in each segmentation region, thus efficiently solving the linear non-separable problem and exploiting the contextual information of HSI. Finally, the classification results are generated by fusing the result of each segmented map via the maximum voting mechanism. Experimental results on three real HSI datasets demonstrate that the proposed method can effectively improve the classification accuracy.

Class incremental learning with analytic learning for hyperspectral image classification