Hyperspectral target detection (HTD) is an important issue in earth observation, with applications in both military and civilian domains. However, conventional representation-based detectors are hindered by the reliance on the unknown background dictionary, the limited ability to capture nonlinear representations using the linear mixing model (LMM), and the insufficient background-target recognition based on handcrafted priors. To address these problems, this paper proposes an interpretable representation network that intuitively realizes LMM for HTD, making nonlinear feature expression and physical interpretability compatible. Specifically, a subspace representation network is designed to separate the background and target components, where the background subspace can be adaptively learned. In addition, to further enhance the nonlinear representation and more accurately learn the coefficients, a lightweight multi-scale Transformer is proposed by modeling long-distance feature dependencies between channels. Furthermore, to supplement the depiction for target-background discrimination, a constrained energy minimization (CEM) loss is tailored by minimizing the output background energy and maximizing the target response. The effectiveness of the proposed method is demonstrated on four benchmark datasets, showing its superiority over state-of-the-art methods. The code for this work is available at https://github.com/shendb2022/HTD-IRN for reproducibility purposes.

Hyperspectral image (HSI) fusion refers to the reconstruction of a high-resolution HSI by fusing a low-resolution HSI (LR-HSI) and a high-resolution multispectral image (HR-MSI) over the same scene. Recently, researchers have proposed many approaches to handle this issue. However, most of them assume that both the spatial and spectral degradation functions are known, which are often limited or unavailable in reality. This article presents a novel model-driven deep network based on matrix decomposition, which considers spectral correlations and reasonably embeds the well-known observation models. Specifically, the proposed method decomposes the desired HSI into spectral basis and coefficients. The spectral basis can be estimated from the LR-HSI via singular value decomposition. To learn the coefficients, a learning model is constructed by merging the observation models, matrix decomposition, and sparsity into a concise single formulation. For solving the proposed model, a deep framework is built by unrolling the alternating direction method of multipliers (ADMM), dubbed as ADMM-HFNet, where the involved parameters can be learned adaptively. It is worth noting that the spectral basis cannot fully represent the desired HSI. Therefore, another model is constructed here to supplement the approximation error, which can also be embedded in the deep network. After checking on three datasets, it is found that the proposed method stands out from advanced competing techniques in both quality measures and visual effects.

Hyperspectral target detection (HTD) identifies objects of interest from complex backgrounds at the pixel level, playing a vital role in Earth observation. However, the limited target priors constrain the ability to obtain sufficient features or patterns for background-target discrimination, and spectral variation further exacerbates the difficulty of achieving reliable and robust performance. To address these challenges, this paper proposes an efficient self-supervised HTD method with a pyramid state space model (SSM), named HTD-Mamba, which employs spectrally contrastive learning to distinguish between target and background based on the similarity measurement of intrinsic features. Specifically, to obtain sufficient training samples and leverage spatial contextual information, we propose a spatial-encoded spectral augmentation technique that encodes all surrounding pixels within a patch into a transformed view of the center pixel. Additionally, to explore global band correlations, we divide pixels into continuous group-wise spectral embeddings and introduce Mamba to HTD for the first time to model long-range dependencies of the spectral sequence with linear complexity. Furthermore, to alleviate spectral variation and enhance robust representation, we propose a pyramid SSM as a backbone to capture and fuse multiresolution spectral-wise intrinsic features. Extensive experiments conducted on four public datasets demonstrate that the proposed method outperforms state-of-the-art methods in both quantitative and qualitative evaluations. Code is available at https://github.com/shendb2022/HTD-Mamba.

Hyperspectral and multispectral image fusion aims to fuse a low-spatial-resolution hyperspectral image (HSI) and a high-spatial-resolution multispectral image to form a high-spatial-resolution HSI. Motivated by the success of model- and deep learning-based approaches, we propose a novel patch-aware deep fusion approach for HSI by unfolding a subspace-based optimization model, where moderate-sized patches are used in both training and test phases. The goal of this approach is to make full use of the information of patch under subspace representation, restrict the scale and enhance the interpretability of the deep network, thereby improving the fusion. First, a subspace-based fusion model was built with two regularization terms to localize pixels and extract texture. Then, the subspace-based fusion model was solved by the alternating direction method of multipliers algorithm, and the model was divided into one fidelity-based problem and two regularization-based problems. Finally, a structured deep fusion network was proposed by unfolding all steps of the algorithm as network layers. Specifically, the fidelity-based problem was solved by a gradient descent algorithm and implemented by a network. The two regularization-based problems were described by proximal operators and learnt by two u-shaped architectures. Moreover, an aggregation fusion technique was proposed to improve the performance by averaging the fused images in all iterations and aggregating the overlapping patches in the test phase. Experimental results, conducted on both synthetic and real datasets, demonstrated the effectiveness of the proposed approach.

Fusing a low-resolution hyperspectral (LRHS) image and a high-resolution multispectral (HRMS) image to generate a high-resolution hyperspectral (HRHS) image has grown a significant and attractive application in remote sensing fields. Recently, the popularization of deep learning has injected more possibilities into the fusion work. However, there still exists a difficulty that is how to make the best of the acquired LRHS and HRMS images. In this paper, we present a twice optimizing net with matrix decomposition to fulfill the fusion task, which can be roughly divided into three stages: pre-optimization, deep prior learning, post-optimization. Specifically, we first transform this fusion problem into a spectral optimization problem and a spatial optimization problem with the help of matrix decomposition. These two optimization problems can be handled sequentially by solving a linear equation respectively and then we can obtain the initial HRHS image by multiplying the two solutions. Next, we establish the mapping between the initial image and the reference image through an end-to-end deep residual network based on local and non-local connectivity. In order to get better performance, we have customized a loss function specifically for the fusion task as well. Finally, we return the predicted result again to the optimization procedure to get the final fusion image. After the evaluation on three simulated datasets and one real dataset, it illustrates that the proposed method outperforms many state-of-the-art ones.

In hyperspectral anomaly detection, tensor low-rankness is essential for effectively separating background and anomaly. However, most of the current low-rank-based methods do not use the spatial-spectral low-rankness and the nonlocal self-similarity simultaneously. To address this issue, we propose a tensor double nuclear norm based tensor approximation (TDNN-TA) model with all the priors in a unified convex framework, which can be efficiently handled through a well-organized alternating direction method of multipliers. Especially, to thoroughly model the background by tensor approximation, we propose a tensor double nuclear norm (TDNN), which achieves a more precise and flexible exploration of low-rankness and nonlocal self-similarity by applying different low-rank constraints to the global tensor and the group tensor. Moreover, to explore the intrinsic characteristics of different priors by various tensor ranks, we employ the Fourier transform-based three-directional tensor nuclear norm to approximate the nonlocal group tensor rank, and the framelet-based three-modal tensor nuclear norm to approximate the global tensor rank. Experimental results validated on several real hyperspectral datasets demonstrate that TDNN-TA is effective in detecting different sizes of anomalous targets and achieves competitive results for various scenes.