Hyperspectral Image Super-resolution Research Articles

MTLSC-Diff: Multitask learning with diffusion models for hyperspectral image super-resolution and classification

Hyperspectral image (HSIs) super-resolution (SR) can improve the spatial resolution of images for subsequent application tasks. In recent years, SR methods based on deep learning have gained widespread attention. However, most of the existing SR methods do not take into account the needs of specific application tasks when designing the network structure. These methods may not be able to efficiently generate high-quality images that satisfy the specific application tasks, leading to degradation of the performance of subsequent application tasks. To solve this problem, we propose a multi-task learning architecture based on the diffusion model, namely MTLSC-Diff. MTLSC-Diff combines the SR network and the classification network in a multi-task learning manner on the basis of the diffusion model. MTLSC-Diff achieves mutual guidance of the two tasks by iterating the image super-resolution and classification tasks, thus gradually reconstructing high-quality images and improving classification accuracy. The guided operations for each time step are performed by the specially designed Mutual-Guidance SR-Classification Synergy Module (M-GSCS). M-GSCS refines the multi-scale image obtained at the previous time step and uses the predicted high spatial resolution image for classification. Meanwhile, a class-guided SR dynamic refinement strategy (C-GSR) is proposed in M-GSCS, which uses multi-scale classification results to guide target scale images to learn new knowledge to further reconstruct high-quality images. Experimental results on relevant datasets show that our method significantly improves the super-resolution performance as well as the classification performance.

Deep blind super-resolution for hyperspectral images

Current deep learning methods for single hyperspectral image super-resolution are non-blind ones, which adopt the simplistic bicubic degradation model. These models have poor generalization performance and cannot handle unknown degradation. Additionally, blind super-resolution methods for RGB images neglect rich spectral information in hyperspectral images, which causes spectral distortion in results. To address these issues, we consider the degeneration estimation and propose a single hyperspectral image blind super-resolution algorithm. Specifically, we first use a blur kernel estimation network and a deblurring network to obtain the image without blur. We change the receptive field by exchanging spatial channel information when estimating the kernel, thereby obtaining blurry information at different scales. In addition, to reduce the errors introduced by kernel mismatch and the relevant information among bands ignored by the group convolution, we fuse the spatial information of the blurred image to compensate for the errors and then utilize the spatial and spectral information of other bands to guide the feature extraction of the current band. Finally, we integrate the feature information from different branches in a global fusion network to improve spatial and spectral information fidelity further. Extensive experiments on synthetic and real-world hyperspectral datasets unequivocally demonstrate that our approach outperforms other state-of-the-art approaches in terms of both quantitative and qualitative evaluations. Our code is publicly available at https://github.com/YoungP2001/DBSR.

Multi-dimensional attention-aided transposed ConvBiLSTM network for hyperspectral image super-resolution

Hyperspectral (HS) image always suffers from the deficiency of low spatial resolution, compared with conventional optical image types, which has limited its further applications in remote sensing areas. Therefore, HS image super-resolution (SR) techniques are broadly employed in order to observe finer spatial structures while preserving the spectra of ground covers. In this paper, a novel multi-dimensional attention-aided transposed convolutional long-short term memory (LSTM) network is proposed for single HS image super-resolution task. The proposed network employs the convolutional bi-directional LSTM for the purpose of local and non-local spatial–spectral feature explorations, and transposed convolution for the purpose of image amplification and reconstruction. Moreover, a multi-dimensional attention module is proposed, aiming to capture the salient features on spectral, channel, and spatial dimensions, simultaneously, to further improve the learning abilities of network. Experiments on four commonly-used HS images demonstrate the effectiveness of this approach, compared with several state-of-the-art deep learning-based SR methods.

Unsupervised Test-Time Adaptation Learning for Effective Hyperspectral Image Super-Resolution With Unknown Degeneration.

Fusing a low-resolution hyperspectral image (HSI) with a high-resolution (HR) multi-spectral image has provided an effective way for HSI super-resolution (SR). The key lies on inferring the posteriori of the latent (i.e., HR) HSI using an appropriate image prior and the likelihood determined by the degeneration between the latent HSI and the observed images. However, in scenarios with complex imaging environments and various imaging scenes, the prior of HSIs can be prohibitively complicated and the degeneration is often unknown, which causes it difficult to accurately infer the posteriori of each latent HSI. To tackle this problem, we present an unsupervised test-time adaptation learning (UTAL) framework for HSI SR under unknown degeneration. Instead of directly modeling the complicated image prior, it first implicitly learns a content-agnostic prior shared across different images through supervisedly pre-training a mutual-guiding fusion module on extensive synthetic data. Then, it adapts the shared prior to those private characteristics in the latent HSI for posteriori inference through unsupervisedly learning a self-guiding adaptation module and a degeneration estimation network on two observed images in the test phase. Such a two-stage learning scheme models the complicated image prior in a divide-and-conquer manner, which eases the modeling difficulty and improves the prior accuracy. Moreover, the unknown degeneration can be estimated properly. Both of these two advantages empower us to accurately infer the posteriori of the latent HSI, thereby increasing the generalization performance in real applications. Additionally, in order to further mitigate the over-fitting in coping with more challenging cases (e.g., degenerations in both spectral and spatial domains are unknown) and speed up, we propose to meta-train UTAL on extensive synthetic SR tasks and solve it using an alternative optimization strategy such that UTAL learns to produce good generalization performance in real challenging cases with a small number of gradient descent steps. To verify the efficacy of UTAL, we evaluate it on HSI SR tasks with different unknown degenerations as well as some other HSI restoration tasks (e.g., compressive sensing), and report strong results superior to that of existing competitors.

HCT: a hybrid CNN and transformer network for hyperspectral image super-resolution

HCT: a hybrid CNN and transformer network for hyperspectral image super-resolution

SSAformer: Spatial–Spectral Aggregation Transformer for Hyperspectral Image Super-Resolution

The hyperspectral image (HSI) distinguishes itself in material identification through its exceptional spectral resolution. However, its spatial resolution is constrained by hardware limitations, prompting the evolution of HSI super-resolution (SR) techniques. Single HSI SR endeavors to reconstruct high-spatial-resolution HSI from low-spatial-resolution inputs, and recent progress in deep learning-based algorithms has significantly advanced the quality of reconstructed images. However, convolutional methods struggle to extract comprehensive spatial and spectral features. Transformer-based models have yet to harness long-range dependencies across both dimensions fully, thus inadequately integrating spatial and spectral data. To solve the above problem, in this paper, we propose a new HSI SR method, SSAformer, which merges the strengths of CNNs and Transformers. It introduces specially designed attention mechanisms for HSI, including spatial and spectral attention modules, and overcomes the previous challenges in extracting and amalgamating spatial and spectral information. Evaluations on benchmark datasets show that SSAformer surpasses contemporary methods in enhancing spatial details and preserving spectral accuracy, underscoring its potential to expand HSI’s utility in various domains, such as environmental monitoring and remote sensing.

Multispectral and hyperspectral images fusion based on subspace representation and nonlocal low-rank regularization

ABSTRACT Multispectral image (MSI) and hyperspectral image (HSI) fusion is a popular method for HSI super-resolution reconstruction (HSI-SR). MSI-HSI fusion problem is ill-posed and demands several image priors or regularization terms to solve accurately, which is a challenging issue. In this paper, we propose an MSI-HSI fusion model via subspace representation and nonlocal low-rank regularization (SRNLRR). SRNLRR model incorporates the global spectral correlations and spatial nonlocal similarities of HSI to improve the fusion results, where the priors complement each other. First, we use the mode-n tensor-matrix product to project latent high spatial resolution HSI (HR-HSI) into spectral subspace, which can capture spectral low-rankness and reduce computational complexity. Then, based on low-rank representation (LRR) and nonlocal processing strategy, we design a spatial nonlocal LRR regularization (spa-NLRR) and a spectral global LRR regularization (spe-GLRR). These two regularizations analyze the spatial nonlocal similarities and spectral global correlations from intermediate-level vision. Finally, we use the residual regularization program to obtain more image information and input it into the fusion model. We use alternate minimization (AM) methods to optimize the SRNLRR model and employ the alternating direction method (ADM) on spatial/spectral LRR learning. Comparison experiments between the SRNLRR method and six advanced methods on three HSI datasets indicate the superiority of our method.

Semi-blind hyperspectral and multispectral image fusion via generalized inverse matrix optimization

Combining a low spatial resolution hyperspectral image (HSI) with a high spatial resolution multispectral image (MSI) is an economical way to obtain a high spatial resolution hyperspectral image (HR-HSI). However, existing fusion methods often rely on a critical variable, the point spread function (PSF), which is difficult to obtain in practice. Additionally, these methods often require a long fusion time and have many parameters that need re-debugging on different datasets to achieve satisfactory results. Many existing methods lack robustness in handling complex scenarios, such as complex PSF, changing spatial downsampling factors, and large-scale datasets. These issues can result in degraded performance or render the methods inapplicable to practical scenarios. Therefore, we propose an efficient and parameterless semi-blind fusion method based on generalized inverse matrix optimization (GIMO), which is robust for multiple complex situations. Our method uses the definition of generalized inverse matrix and the relationship among HSI, MSI, and HR-HSI to approximate the function module of PSF. Then, we calculate the generalized inverse matrix of the spectral response function (SRF). Finally, we optimize the generalized inverse matrix of SRF using the Sylvester equation and multiply it with MSI to obtain HR-HSI. This is the first proposed method to achieve hyperspectral image super-resolution by optimizing the inverse matrix of the spectral response function. Experiments on multiple simulated datasets and one real dataset have validated that our GIMO method outperforms existing state-of-the-art methods, especially regarding fusion time and robustness for various complex scenarios.

Enhancing Hyperspectral Images via Diffusion Model and Group-Autoencoder Super-resolution Network

Existing hyperspectral image (HSI) super-resolution (SR) methods struggle to effectively capture the complex spectral-spatial relationships and low-level details, while diffusion models represent a promising generative model known for their exceptional performance in modeling complex relations and learning high and low-level visual features. The direct application of diffusion models to HSI SR is hampered by challenges such as difficulties in model convergence and protracted inference time. In this work, we introduce a novel Group-Autoencoder (GAE) framework that synergistically combines with the diffusion model to construct a highly effective HSI SR model (DMGASR). Our proposed GAE framework encodes high-dimensional HSI data into low-dimensional latent space where the diffusion model works, thereby alleviating the difficulty of training the diffusion model while maintaining band correlation and considerably reducing inference time. Experimental results on both natural and remote sensing hyperspectral datasets demonstrate that the proposed method is superior to other state-of-the-art methods both visually and metrically.

Learning Multi-Modal Cross-Scale Deformable Transformer Network for Unregistered Hyperspectral Image Super-resolution

Hyperspectral image super-resolution (HSI-SR) is a technology to improve the spatial resolution of HSI. Existing fusion-based SR methods have shown great performance, but still have some problems as follows: 1) existing methods assume that the auxiliary image providing spatial information is strictly registered with the HSI, but images are difficult to be registered finely due to the shooting platforms, shooting viewpoints and the influence of atmospheric turbulence; 2) most of the methods are based on convolutional neural networks (CNNs), which is effective for local features but cannot utilize the global features. To this end, we propose a multi-modal cross-scale deformable transformer network (M2DTN) to achieve unregistered HSI-SR. Specifically, we formulate a spectrum-preserving based spatial-guided registration-SR unified model (SSRU) from the view of the realistic degradation scenarios. According to SSRU, we propose multi-modal registration deformable module (MMRD) to align features between different modalities by deformation field. In order to efficiently utilize the unique information between different modals, we design multi-scale feature transformer (MSFT) to emphasize the spatial-spectral features at different scales. In addition, we propose the cross-scale feature aggregation module (CSFA) to accurately reconstruct the HSI by aggregating feature information at different scales. Experiments show that M2DTN outperforms the-state-of-the-art HSI-SR methods. Code is obtainable at https://github.com/Jiahuiqu/M2DTN.

S2CycleDiff: Spatial-Spectral-Bilateral Cycle-Diffusion Framework for Hyperspectral Image Super-resolution

Hyperspectral image super-resolution (HISR) is a technique that can break through the limitation of imaging mechanism to obtain the hyperspectral image (HSI) with high spatial resolution. Although some progress has been achieved by existing methods, most of them directly learn the spatial-spectral joint mapping between the observed images and the target high-resolution HSI (HrHSI), failing to fully reserve the spectral distribution of low-resolution HSI (LrHSI) and the spatial distribution of high-resolution multispectral imagery (HrMSI). To this end, we propose a spatial-spectral-bilateral cycle-diffusion framework (S2CycleDiff) for HISR, which can step-wise generate the HrHSI with high spatial-spectral fidelity by learning the conditional distribution of spatial and spectral super-resolution processes bilaterally. Specifically, a customized conditional cycle-diffusion framework is designed as the backbone to achieve the spatial-spectral-bilateral super-resolution by repeated refinement, wherein the spatial/spectral guided pyramid denoising (SGPD) module seperately takes HrMSI and LrHSI as the guiding factors to achieve the spatial details injection and spectral correction. The outputs of the conditional cycle-diffusion framework are fed into a complementary fusion block to integrate the spatial and spectral details to generate the desired HrHSI. Experiments have been conducted on three widely used datasets to demonstrate the superiority of the proposed method over state-of-the-art HISR methods. The code is available at https://github.com/Jiahuiqu/S2CycleDiff.

A spectral and spatial transformer for hyperspectral remote sensing image super-resolution

ABSTRACT Due to the generally low spatial resolution of hyperspectral images (HSIs), early multispectral images lacked corresponding panchromatic bands, and as a result, fusion methods could not be used to enhance resolution. Many researchers have proposed various image super-resolution methods to address these limitations. However, these methods still suffered from issues, such as inadequate feature representation, lack of spectral feature representation, and high computational cost and inefficiency. To address these challenges, a spectral and spatial transformer (SST) algorithm for hyperspectral remote sensing image super-resolution is introduced. This algorithm uses a spatial transformer structure to extract the spatial features between the image pixels and a spectral transformer structure to extract the spectral features within the image pixels. The integration of these two components is applied to HSI super-resolution. After comparative experiments with currently advanced methods on three publicly available hyperspectral datasets, the results consistently show that our algorithm has better performance in both spectral fidelity and spatial restoration. Furthermore, our proposed algorithm was applied to real-world super-resolution experiments in the region of China's Ruoergai National Park, and subsequently, pixel-based classification was conducted on the super-resolution images, the results indicate that our algorithm could also be applied to future remote sensing interpretation tasks.

Hyperspectral image super resolution using deep internal and self‐supervised learning

AbstractBy automatically learning the priors embedded in images with powerful modelling capabilities, deep learning‐based algorithms have recently made considerable progress in reconstructing the high‐resolution hyperspectral (HR‐HS) image. With previously collected large‐amount of external data, these methods are intuitively realised under the full supervision of the ground‐truth data. Thus, the database construction in merging the low‐resolution (LR) HS (LR‐HS) and HR multispectral (MS) or RGB image research paradigm, commonly named as HSI SR, requires collecting corresponding training triplets: HR‐MS (RGB), LR‐HS and HR‐HS image simultaneously, and often faces difficulties in reality. The learned models with the training datasets collected simultaneously under controlled conditions may significantly degrade the HSI super‐resolved performance to the real images captured under diverse environments. To handle the above‐mentioned limitations, the authors propose to leverage the deep internal and self‐supervised learning to solve the HSI SR problem. The authors advocate that it is possible to train a specific CNN model at test time, called as deep internal learning (DIL), by on‐line preparing the training triplet samples from the observed LR‐HS/HR‐MS (or RGB) images and the down‐sampled LR‐HS version. However, the number of the training triplets extracted solely from the transformed data of the observation itself is extremely few particularly for the HSI SR tasks with large spatial upscale factors, which would result in limited reconstruction performance. To solve this problem, the authors further exploit deep self‐supervised learning (DSL) by considering the observations as the unlabelled training samples. Specifically, the degradation modules inside the network were elaborated to realise the spatial and spectral down‐sampling procedures for transforming the generated HR‐HS estimation to the high‐resolution RGB/LR‐HS approximation, and then the reconstruction errors of the observations were formulated for measuring the network modelling performance. By consolidating the DIL and DSL into a unified deep framework, the authors construct a more robust HSI SR method without any prior training and have great potential of flexible adaptation to different settings per observation. To verify the effectiveness of the proposed approach, extensive experiments have been conducted on two benchmark HS datasets, including the CAVE and Harvard datasets, and demonstrate the great performance gain of the proposed method over the state‐of‐the‐art methods.

Hyperspectral Image Super-Resolution Based on Feature Diversity Extraction

Deep learning is an important research topic in the field of image super-resolution. Problematically, the performance of existing hyperspectral image super-resolution networks is limited by feature learning for hyperspectral images. Nevertheless, the current algorithms exhibit some limitations in extracting diverse features. In this paper, we address limitations to existing hyperspectral image super-resolution networks, focusing on feature learning challenges. We introduce the Channel-Attention-Based Spatial–Spectral Feature Extraction network (CSSFENet) to enhance hyperspectral image feature diversity and optimize network loss functions. Our contributions include: (a) a convolutional neural network super-resolution algorithm incorporating diverse feature extraction to enhance the network’s diversity feature learning by elevating the matrix rank, (b) a three-dimensional (3D) feature extraction convolution module, the Channel-Attention-Based Spatial–Spectral Feature Extraction Module (CSSFEM), to boost the network’s performance in both the spatial and spectral domains, (c) a feature diversity loss function designed based on the image matrix’s singular value to maximize element independence, and (d) a spatial–spectral gradient loss function introduced based on space and spectrum gradient values to enhance the reconstructed image’s spatial–spectral smoothness. In contrast to existing hyperspectral super-resolution algorithms, we used four evaluation indexes, PSNR, mPSNR, SSIM, and SAM, and our method showed superiority during testing with three common hyperspectral datasets.

Hyperspectral Image Super Resolution With Real Unaligned RGB Guidance.

Fusion-based hyperspectral image (HSI) super-resolution has become increasingly prevalent for its capability to integrate high-frequency spatial information from the paired high-resolution (HR) RGB reference (Ref-RGB) image. However, most of the existing methods either heavily rely on the accurate alignment between low-resolution (LR) HSIs and RGB images or can only deal with simulated unaligned RGB images generated by rigid geometric transformations, which weakens their effectiveness for real scenes. In this article, we explore the fusion-based HSI super-resolution with real Ref-RGB images that have both rigid and nonrigid misalignments. To properly address the limitations of existing methods for unaligned reference images, we propose an HSI fusion network (HSIFN) with heterogeneous feature extractions, multistage feature alignments, and attentive feature fusion. Specifically, our network first transforms the input HSI and RGB images into two sets of multiscale features with an HSI encoder and an RGB encoder, respectively. The features of Ref-RGB images are then processed by a multistage alignment module to explicitly align the features of Ref-RGB with the LR HSI. Finally, the aligned features of Ref-RGB are further adjusted by an adaptive attention module to focus more on discriminative regions before sending them to the fusion decoder to generate the reconstructed HR HSI. Additionally, we collect a real-world HSI fusion dataset, consisting of paired HSI and unaligned Ref-RGB, to support the evaluation of the proposed model for real scenes. Extensive experiments are conducted on both simulated and our real-world datasets, and it shows that our method obtains a clear improvement over existing single-image and fusion-based super-resolution methods on quantitative assessment as well as visual comparison. The code and dataset are publicly available at https://zeqiang-lai.github.io/HSI-RefSR/.

Test-time Training for Hyperspectral Image Super-resolution.

The progress on Hyperspectral image (HSI) super-resolution (SR) is still lagging behind the research of RGB image SR. HSIs usually have a high number of spectral bands, so accurately modeling spectral band interaction for HSI SR is hard. Also, training data for HSI SR is hard to obtain so the dataset is usually rather small. In this work, we propose a new test-time training method to tackle this problem. Specifically, a novel self-training framework is developed, where more accurate pseudo-labels and more accurate LR-HR relationships are generated so that the model can be further trained with them to improve performance. In order to better support our test-time training method, we also propose a new network architecture to learn HSI SR without modeling spectral band interaction and propose a new data augmentation method Spectral Mixup to increase the diversity of the training data at test time. We also collect a new HSI dataset with a diverse set of images of interesting objects ranging from food to vegetation, to materials, and to general scenes. Extensive experiments on multiple datasets show that our method can improve the performance of pre-trained models significantly after test-time training and outperform competing methods significantly for HSI SR.

CS2DIPs: Unsupervised HSI Super-Resolution Using Coupled Spatial and Spectral DIPs.

In recent years, fusing high spatial resolution multispectral images (HR-MSIs) and low spatial resolution hyperspectral images (LR-HSIs) has become a widely used approach for hyperspectral image super-resolution (HSI-SR). Various unsupervised HSI-SR methods based on deep image prior (DIP) have gained wide popularity thanks to no pre-training requirement. However, DIP-based methods often demonstrate mediocre performance in extracting latent information from the data. To resolve this performance deficiency, we propose a coupled spatial and spectral deep image priors (CS2DIPs) method for the fusion of an HR-MSI and an LR-HSI into an HR-HSI. Specifically, we integrate the nonnegative matrix-vector tensor factorization (NMVTF) into the DIP framework to jointly learn the abundance tensor and spectral feature matrix. The two coupled DIPs are designed to capture essential spatial and spectral features in parallel from the observed HR-MSI and LR-HSI, respectively, which are then used to guide the generation of the abundance tensor and spectral signature matrix for the fusion of the HSI-SR by mode-3 tensor product, meanwhile taking some inherent physical constraints into account. Free from any training data, the proposed CS2DIPs can effectively capture rich spatial and spectral information. As a result, it exhibits much superior performance and convergence speed over most existing DIP-based methods. Extensive experiments are provided to demonstrate its state-of-the-art overall performance including comparison with benchmark peer methods.

Hyperspectral Image Super-Resolution With ConvLSTM Skip-Connections

Hyperspectral Image Super-Resolution With ConvLSTM Skip-Connections

Cross-Scope Spatial-Spectral Information Aggregation for Hyperspectral Image Super-Resolution.

Hyperspectral image super-resolution has attained widespread prominence to enhance the spatial resolution of hyperspectral images. However, convolution-based methods have encountered challenges in harnessing the global spatial-spectral information. The prevailing transformer-based methods have not adequately captured the long-range dependencies in both spectral and spatial dimensions. To alleviate this issue, we propose a novel cross-scope spatial-spectral Transformer (CST) to efficiently investigate long-range spatial and spectral similarities for single hyperspectral image super-resolution. Specifically, we devise cross-attention mechanisms in spatial and spectral dimensions to comprehensively model the long-range spatial-spectral characteristics. By integrating global information into the rectangle-window self-attention, we first design a cross-scope spatial self-attention to facilitate long-range spatial interactions. Then, by leveraging appropriately characteristic spatial-spectral features, we construct a cross-scope spectral self-attention to effectively capture the intrinsic correlations among global spectral bands. Finally, we elaborate a concise feed-forward neural network to enhance the feature representation capacity in the Transformer structure. Extensive experiments over three hyperspectral datasets demonstrate that the proposed CST is superior to other state-of-the-art methods both quantitatively and visually. The code is available at https://github.com/Tomchenshi/CST.git.

Single hyperspectral image super-resolution using a progressive upsampling deep prior network

<p>Hyperspectral image super-resolution (SR) aims to enhance the spectral and spatial resolution of remote sensing images, enabling more accurate and detailed analysis of ground objects. However, hyperspectral images have high dimensional characteristics and complex spectral patterns. As a result, it is critical to effectively leverage the spatial non-local self-similarity and spectral correlation within hyperspectral images. To address this, we have proposed a novel single hyperspectral image SR method based on a progressive upsampling deep prior network. Specifically, we introduced the spatial-spectral attention fusion unit (S<sup>2</sup>AF) based on residual connections, in order to extract spatial and spectral information from hyperspectral images. Then we developed the group convolutional upsampling (GCU) to efficiently utilize the spatial and spectral prior information inherent in hyperspectral images. To address the challenges posed by the high dimensionality of hyperspectral images and limited training dataset, we implemented a parameter-sharing grouped convolutional upsampling framework within the GCU to ensure model stability and enhance performance. The experimental results on three benchmark datasets demonstrated that the proposed single hyperspectral image SR using a progressive upsampling deep prior network (PUDPN) method effectively improves the reconstruction quality of hyperspectral images and achieves promising performance.</p>