Major depressive disorder (MDD) is a common neuropsychiatric disorder, yet its underlying physiological mechanisms remain unclear, limiting diagnostic advances. Functional connectivity (FC) derived from resting-state functional magnetic resonance imaging (rs-fMRI), when combined with deep learning methods, has shown promise as diagnostic biomarker. Currently, most FC-based diagnostic methods rely on graph structures modeled by FC, and are limited to pairwise interactions between brain regions. Hypergraph representations enable the characterization of higher-order interactions across multiple regions. However, existing hypergraph models ignore the directionality of these interactions, thus limiting their ability to capture complex neural dynamics. To address these limitations, this study proposes a dynamic weight aware directed hypergraph learning method - dwDHGL, for MDD identification and subtype analysis. dwDHGL captures asymmetric causal interactions by modeling temporal lag effects and constructs a directed hypergraph network(DHN). It further utilizes a self-attention mechanism to dynamically learn inter node interactions during message passing and adaptively differentiate node importance. A node weight aware directed hypergraph convolution is designed to aggregate features based on hyperedge directions, incorporating dynamic weights to enhance representation learning. The proposed method is evaluated on the large-scale REST-meta-MDD dataset, achieving an MDD identification accuracy of 73.75 %, and outperforming existing advanced methods in subtype identification. Furthermore, dwDHGL identifies discriminative directed hyperedges, with the inferior frontal gyrus triangular part emerging as key biomarkers, providing new insights into the neural mechanisms of MDD.

The hyperspectral anomaly detection (HAD) aims to identify potential anomalies from complex backgrounds. Most reconstruction-based autoencoders equally treat background pixels and anomalies or ignore potential spatial information. In this letter, we propose an HAD method based on multiscale memory autoencoder and spatial filtering, abbreviated as SFM2AE. Specifically, by introducing memory modules into different hidden layers of the autoencoder, multiscale reconstruction of background and anomaly pixels is achieved in the spectral domain. In addition, morphological filtering in the spatial domain is used to extract spatial structural information from anomalies. Joint spatial-spectral anomaly detection is achieved by combining multiscale memory autoencoder and spatial filtering. Experiments demonstrate superior detection performance of the proposed method over the state-of-the-art methods.

Pansharpening of multispectral images aims to merge multispectral (MS) and panchromatic (PAN) images to produce high-resolution multispectral (HRMS) images. Unsupervised pansharpening algorithms, which are widely used in deep learning for pansharpening tasks, commonly employ generative adversarial networks (GANs). However, existing unsupervised methods based on GANs have some limitations: 1) restricted ability for joint preservation of spatial and spectral information, and 2) the receptive field of general convolutions restricts the extraction of long-range dependencies. To address these issues, we propose a frequency-guided generative adversarial networks for unsupervised pansharpening (FG-GAN). Our unsupervised framework uses high-frequency and low-frequency information as prior constraints to guide the training of the FG-GAN’s generator and discriminator networks, thereby enhancing the joint preservation of spatial and spectral details. Furthermore, a graph convolution-based generator network is designed, in which long-range edge dependencies are extracted and propagated by learning the relationships between distant edge feature nodes. Extensive experiments on the Quickbird and Gaofen-2 datasets demonstrate the effectiveness of our design: our method enhances spatial details and reduces the spatial distortion index (Ds) by 44.8%, while achieving the fidelity of spatial-spectral information with a HQNR score of 0.99.

Hyperspectral image (HSI) classification is a key task in the field of remote sensing, aiming to assign category labels to each pixel by leveraging the spectral and spatial information in HSIs. Recently, many deep learning (DL) methods, such as convolutional neural networks (CNNs) and Transformers, have been applied to this task, achieving significant results. However, most existing patch-based DL methods often overlook the potential relationships between the central pixel and its surrounding pixels. Additionally, the unique spectral characteristics of HSIs, such as the high correlation between adjacent spectral bands and low dependence between distant bands, also require special attention. Based on this, we propose a novel dual-branch convolution-Transformer network with spectral-spatial attention (CTSSA), which can effectively aggregate both local and global spectral-spatial features. Specifically, CTSSA comprises two core modules: the Pyramid Spectral Attention Module (PSAM) and the Center Transformer Encoder (CenterTE). The former extracts highly discriminative spectral features through a hierarchical multi-scale attention mechanism, capturing subtle differences between adjacent spectral bands. The latter improves the original Transformer encoder (TE) by introducing a center-attention mechanism to model the global relationship between the central pixel and its surrounding pixels, thereby enhancing classification accuracy while reducing computational complexity. Experimental results on four public datasets (Salinas, Pavia University, Houston, and WHUHi-LongKou) demonstrate that, compared with nine other networks, CTSSA achieves satisfactory performance with fewer parameters and relatively high efficiency.

The joint clustering of multimodal remote sensing (RS) data represents a multiobjective optimization challenge involving conflicting modality-specific objectives and diverse regularization objectives. Current approaches to multiview subspace clustering (MVSC) often oversimplify this task by transforming it into a weighted single-objective optimization problem, neglecting the intricate interactions between objectives and leading to suboptimal subspace representations. The presence of quadratic decision variables in MVSC renders direct application on large-scale RS data impracticable using multiobjective evolutionary algorithms (MOEAs). To overcome this challenge, we propose a novel MVSC method termed multiobjective multiview attributed graph learning (MMAGL). Instead of optimizing every self-representation coefficient individually, our method transforms MVSC into a link prediction task over a sparse attributed graph that fuses different modalities. We incorporate superpixel-based sample reduction and proximity-based population coding, leveraging spatial and structural priors, respectively. This results in a significantly compressed decision space, enabling optimization with MOEAs. To fully exploit node attributes and the graph structure, we redefine self-representation using contrastive learning and introduce an efficient graph filtering (GF) through a generalized spectral graph convolution, enhancing clustering discriminability. The proposed MMAGL constitutes a hybrid and versatile framework, adaptable to any MOEA. Extensive experimental evaluations demonstrate that our MMAGL method surpasses the current state-of-the-art on multimodal RS benchmarks (e.g., with nearly 2% gain on Trento and 3% on Houston) on overall accuracy.

The joint clustering of multimodal remote sensing (RS) data poses a critical and challenging task in Earth observation. Although recent advances in multiview subspace clustering have shown remarkable success, existing methods become computationally prohibitive when dealing with large-scale RS datasets. Moreover, they neglect intrinsic nonlinear and spatial interdependencies among heterogeneous RS data and lack generalization ability for out-of-sample data, thereby restricting their applicability. This article introduces a novel unified framework called anchor-based multiview kernel subspace clustering with spatial regularization (AMKSC). It learns a scalable anchor graph in the kernel space, leveraging contributions from each modality instead of seeking a consensus full graph in the feature space. To ensure spatial consistency, we incorporate a spatial smoothing operation into the formulation. The method is efficiently solved using an alternating optimization strategy, and we provide theoretical evidence of its scalability with linear computational complexity. Furthermore, an out-of-sample extension of AMKSC based on multiview collaborative representation-based classification is introduced, enabling the handling of larger datasets and unseen instances. Extensive experiments on three real heterogeneous RS datasets confirm the superiority of our proposed approach over state-of-the-art methods in terms of clustering performance and time efficiency. The source code is available at https://github.com/AngryCai/AMKSC.

Functional connectivity (FC) derived from fMRI reflects the interactions between brain regions of interest (ROIs). It has become one of the important features of individualized prediction. However, in studies using FC as input, some studies have extracted global features directly from the entire brain FC, lacking focus on local critical information. On the other hand, some studies have extracted local critical features from selected ROIs or connections, but lack access to global contextual information. In this paper, we propose a novel method, namely, Global-Local Brain Network (GLBN), focusing on both global contextual information and critical local information. We validate our proposed method on the largescale public dataset, the Cambridge Centre for Ageing and Neuroscience (Cam-CAN). For the prediction of age and fluid intelligence, GLBN achieves the mean absolute errors of 6.084, and 4.160, with Pearson’s correlations of 0.908, and 0.641, respectively. Our method demonstrates superior prediction accuracy compared to existing studies. Additionally, we visualize the brain ROIs that play crucial roles in the prediction tasks, affirming the biological interpretability of GLBN.

In remote sensing image fusion, pansharpening is a type of remote sensing image fusion method that aims to fuse panchromatic (PAN) images and multispectral (MS) images to produce high-resolution multispectral (HRMS) images. Deep learning based pansharpening technology offers a series of advanced unsupervised algorithms. However, there are several challenges: (1) The existing unsupervised pansharpening methods only consider the fusion of single-scale features; (2) for the fusion of MS and PAN image feature branches, the existing pansharpening methods are implemented directly by concatenation and summation, without paying attention to critical features or suppressing redundant features; (3) the semantic gap in the long skip connections of the network architecture will create unexpected results. In this paper, we design a multiscale unsupervised architecture based on generative adversarial networks (GANs) for remote sensing image pansharpening (Mun-GAN), which consists of a generator and two discriminators. The generator includes a multi-scale feature extractor (MFE), a self-adaptation weighted fusion (SWF) module, and a nest feature aggregation (NFA) module. First, the MFE is utilized to extract multiscale feature information from the input images and to then pass this information to the SWF module for adaptive weight fusion. Then, multiscale features are reconstructed by the NFA module to obtain HRMS images. The two discriminators are spectral and spatial discriminators used against the generator. Moreover, we design a hybrid loss function to aggregate the multiscale spectral and spatial feature information. Compared with other state-of-the-art methods using QuickBird, GaoFen-2 and WorldView-3 images, which demonstrate that the Mun-GAN yields better fusion results.

Dimensionality reduction (DR) is important for feature extraction and classification of hyperspectral images (HSIs). Recently proposed superpixel-based DR models have shown promising performance, where superpixel segmentation techniques were applied to segment an HSI and then DR models like principal component analysis (PCA) or linear discriminant analysis (LDA) were employed to extract the local and/or global features. However, superpixelwise PCA based local features are unsatisfactory because PCA aims to extract features with high variance, which could be inefficient in superpixels with mixed objects or strong noise/outliers. In addition, superpixelwise unsupervised LDA based global features may neglect local (spatial-contextual) information. To address these issues, we propose a new spectral-spatial and superpixelwise unsupervised LDA (S3-ULDA) model for unsupervised feature extraction from HSIs. Specifically, the HSI is first segmented into various superpixels with pseudo labels. Then, superpixel based local reconstruction for HSI denoising is conducted. Next, superpixelwise unsupervised LDA (SuperULDA) is performed on both the original HSI and locally reconstructed data to extract global features. Then, superpixelwise unsupervised local Fisher discriminant analysis (SuperULFDA) is developed for local feature extraction, where each superpixel and its adjacent superpixels (along with their pseudo-labels) are fed into local Fisher discriminant analysis (LFDA) to extract local features. The superpixel-level local manifold structures can be effectively modeled by the proposed SuperULFDA. Finally, by fusing the extracted global and local features, novel global-local and spectral-spatial features can be obtained. Our experimental results on several benchmark HSIs demonstrate the superiority of the proposed method over state-of-the-art methods. The code of the proposed model is available at https://github.com/XinweiJiang/S3-ULDA.

This article presents FLGC, a simple yet effective fully linear graph convolutional network for semi-supervised and unsupervised learning. Instead of using gradient descent, we train FLGC based on computing a global optimal closed-form solution with a decoupled procedure, resulting in a generalized linear framework and making it easier to implement, train, and apply. We show that (1) FLGC is powerful to deal with both graph-structured data and regular data, (2) training graph convolutional models with closed-form solutions improve computational efficiency without degrading performance, and (3) FLGC acts as a natural generalization of classic linear models in the non-Euclidean domain (e.g., ridge regression and subspace clustering). Furthermore, we implement a semi-supervised FLGC and an unsupervised FLGC by introducing an initial residual strategy, enabling FLGC to aggregate long-range neighborhoods and alleviate over-smoothing. We compare our semi-supervised and unsupervised FLGCs against many state-of-the-art methods on a variety of classification and clustering benchmarks, demonstrating that the proposed FLGC models consistently outperform previous methods in terms of accuracy, robustness, and learning efficiency. The core code of our FLGC is released at https://github.com/AngryCai/FLGC .

Deep neural networks have gained increasing interest in hyperspectral image (HSI) processing. However, prior arts often neglect the high-order correlation among data points, failing to capture intraclass variations. In this letter, we present a unified neural network framework, termed as hypergraph-structured autoencoder (HyperAE), to leverage the high-order relationship among data and learn robust deep representation for downstream tasks. Technically, the proposed method adopts a deep autoencoder regularized by hypergraph structure as the backbone network, which is jointly trained with a task-specific branch, resulting in a multitask architecture. We separately combine the subspace clustering model and the softmax classifier into the HyperAE to deal with HSI unsupervised and semisupervised classification problems. Benefiting from the hypergraph, HyperAE endows traditional networks with the capacity of preserving the high-order structured information. We evaluate the proposed methods on three benchmarking HSI data sets, demonstrating that the proposed HyperAE dramatically outperforms many existing methods with significant margins in both unsupervised and semisupervised HSI classification problems.

The bilateral filter (BF) is a nonlinear filtering method, which can remove noise and retain better edge information. It has been widely used in the field of hyperspectral images (HSIs) filtering. In this letter, we propose a novel spectral-spatial information integration method based on the BF with multispatial domain (MBF). The proposed method includes three steps. First, principal component analysis (PCA) is used for the original HSI to obtain multiple components containing almost all information; second, multiple principal components are used as both spatial domain and range domain information for BF; finally, the extreme learning machine (ELM) is used for classification. To verify the effectiveness of the proposed approach, we evaluate performance on three benchmark data sets. Our method will improve the existing filtering methods by constructing multiple spatial domains for filtering, which will make more effective use of spatial features and solve the problem of lack of spatial information in HSIs. This method is compared with other filtering algorithms. Comparative experiments show that our proposed method can improve the classification accuracy. And the MBF information is more effective than the BF with single spatial domain information and other filtering methods.

Hyperspectral images (HSIs) consisting of abundant spectral bands could lead to the curse of dimensionality issue when performing HSIs classification. In this letter, an unsupervised dimensionality reduction (DR) method termed Laplacian regularized collaborative representation projection (LRCRP) is proposed, where Laplacian regularization and local enhancement are introduced into collaborative representation (CR) to construct adjacent graph and then to reduce the spectral dimension in graph embedding framework. As the constructed graph simultaneously preserves the local manifold and global information in HSIs, the proposed LRCRP could be used to extract effective low-dimensional features for accurate HSIs classification. The experimental results on two HSI datasets demonstrate the effectiveness of the proposed model. The source code the proposed model is available athttps://github.com/XinweiJiang/LRCRP.

Hyperspectral image (HSI) clustering has attracted a great deal of attention, owing to lower cost and higher application prospects. Deep subspace clustering has been proved to be an effective method to explore the sample relationship of HSI clustering. However, due to the complex distribution of HSI data, the problem of data cluster overlap occurs frequently. In the actual sample distribution, a sample may belong to multiple subspaces. The complex sample distribution brings challenges to subspace clustering. In this letter, we propose a deep mutual information subspace clustering network (DMISC) to find a more intuitive feature space for non-linear subspace clustering. Technically, we maximize the mutual information between the samples and their generated features to enlarge the inter-class dispersion and intra-class compactness. The deep subspace method can find a more suitable non-linear intrinsic relationship, benefitting from the generated feature distribution. We evaluate DMISC on four HSI data sets and compare the performances with 12 popular clustering methods. The experiment results demonstrate our method outperforms many prior unsupervised methods.

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.

Convolutional neural networks (CNNs) are widely used in hyperspectral image (HSI) classification. However, the network architecture of CNNs is often designed manually, which requires careful fine-tuning. Recently, many techniques for neural architecture search (NAS) have been proposed to design the network automatically but most of the methods are only concerned with the overall classification accuracy and ignore the balance between the floating point operations per second (FLOPs) and the number of parameters. In this paper, we propose a new multi-objective optimization (MO) method called MO-CNN to automatically design the structure of CNNs for HSI classification. First, a MO method based on continuous particle swarm optimization (CPSO) is constructed, where the overall accuracy, floating point operations (FLOPs) and the number of parameters are considered, to obtain an optimal architecture from the Pareto front. Then, an auxiliary skip connection strategy is added (together with a partial connection strategy) to avoid performance collapse and to reduce memory consumption. Furthermore, an end-to-end band selection network (BS-Net) is used to reduce redundant bands and to maintain spectral-spatial uniformity. To demonstrate the performance of our newly proposed MO-CNN in scenarios with limited training sets, a quantitative and comparative analysis (including ablation studies) is conducted. Our optimization strategy is shown to improve the classification accuracy, reduce memory and obtain an optimal structure for CNNs based on unbiased datasets.

As the most classical unsupervised dimension reduction algorithm, principal component analysis (PCA) has been widely used in hyperspectral images (HSIs) preprocessing and analysis tasks. Recently proposed superpixelwise PCA (SuperPCA) has shown promising accuracy where superpixels segmentation technique was first used to segment an HSI to various homogeneous regions and then PCA was adopted in each superpixel block to extract the local features. However, the local features could be ineffective due to the neglect of global information especially in some small homogeneous regions and/or in some large homogeneous regions with mixed ground truth objects. In this article, a novel spectral–spatial and SuperPCA (S3-PCA) is proposed to learn the effective and low-dimensional features of HSIs. Inspired by SuperPCA we further adopt superpixels-based local reconstruction to filter the HSIs and use the PCA-based global features as the supplement of local features. It turns out that the global–local and spectral–spatial features can be well exploited. Specifically, each pixel of an HSI is reconstructed by the nearest neighbors’ pixels in the same superpixel block, which could eliminate the noise and enhance the spatial information adaptively. After the local reconstruction-based data preprocessing, PCA is performed on each region and the entire HSI to obtain local and global features, respectively. Then we simply concatenate them to get the global–local and spectral–spatial features for HSIs classification. The experimental results on two HSIs data sets demonstrate the superiority of the proposed method over the state-of-the-art methods. The source code of the proposed model is available athttps://github.com/XinweiJiang/S3-PCA.

Recently, deep learning-based methods have made great progress in hyperspectral image (HSI) classification (HSIC). Different from ordinary images, the intrinsic complexity of HSIs data still limits the performance of many common convolutional neural network (CNN) models. Thus, the network architecture becomes more and more complex to extract discriminative spectral-spatial features. For instance, 3-D CNN usually has a large number of trainable parameters, thus increasing the computational complexity of the HSIC. In this letter, we designed a cooperative spectral-spatial attention dense network (CS2ADN) that takes raw 3-D HSI data as input data. Specifically, the attention module consists of spectral and spatial axes, by which the salient spectral-spatial features will be emphasized. Furthermore, we combined these attention modules with the dense connection, which is termed as the lightweight dense block; it has a lower computation cost and achieves better classification performance. At the same time, we introduced the center loss, by jointly using the supervision of the center loss and the softmax loss, where the discriminative features could be clearly observed, particularly for small data sets. Experimental results on the biased and unbiased HSI data show that our method outperforms several state-of-the-art methods in HSIC with small training samples.

Hyperspectral image (HSI) clustering is a challenging task due to the high complexity of HSI data. Subspace clustering has been proven to be powerful for exploiting the intrinsic relationship between data points. Despite the impressive performance in the HSI clustering, traditional subspace clustering methods often ignore the inherent structural information among data. In this article, we revisit the subspace clustering with graph convolution and present a novel subspace clustering framework called graph convolutional subspace clustering (GCSC) for robust HSI clustering. Specifically, the framework recasts the self-expressiveness property of the data into the non-Euclidean domain, which results in a more robust graph embedding dictionary. We show that traditional subspace clustering models are the special forms of our framework with the Euclidean data. On the basis of the framework, we further propose two novel subspace clustering models by using the Frobenius norm, namely efficient GCSC (EGCSC) and efficient kernel GCSC (EKGCSC). Each model has a globally optimal closed-form solution, making it easier to implement, train, and apply in practice. Extensive experiments strongly evidence that EGCSC and EKGCSC dramatically outperform current models on three popular HSI data sets consistently.

Deep convolutional neural networks(CNNs) have been widely used in hyperspectral image(HSI) classification. However, these CNNs architectures are all handcrafted, which need professional knowledge and consume very significant time. In order to automatically search for cell-based CNNs architectures for HSI classification, we proposed an cell-based CNNs architecture search method by particle swarm optimization(PSO), which is capable of getting the global optimal architecture compared with the gradient descent method. First, the cell-based search space is structured. Then, a novel directly encoding strategy is devised to encode architectures into particles. Finally, PSO is used to search for the optimal deep architecture from the particle swarm. Furthermore, 1-D PSO-NET and 3-D PSO-NET based on PSO-NET are used as spectral and spectral-spatial HSI classifiers respectively. The experiments on two widely used datasets of hyperspectral image show that the method we proposed achieve good performance.

Extreme Learning Machine (ELM) has gained lots of research interest due to its universal approximation capability and fast learning speed. However, traditional ELMs are devised for regular Euclidean data, such as 2D grid and 1D sequence, and thus don't apply to non-Euclidean data, e.g., graph-structured data. To overcome this shortcoming, this paper presents a Graph Convolutional Extreme Learning Machine (termed as GCELM) for semi-supervised classification. Technically, a random graph convolutional layer is introduced to replace the random projection of original ELM, which endues ELM with the capability of dealing with graph-structured data directly. To generate a robust graph from the raw dataset, a self-representation model is adopted to construct a weighted graph. Extensive experiments on 27 UCI datasets demonstrate that GCELM outperforms many popular semi-supervised methods, and with faster learning speed. To the best of our knowledge, this is the first work that combines graph convolution with ELM.

Classification of hyperspectral images (HSIs) by making full use of the spectral and the spatial information has become a research hotspot in the field of remote sensing technology. Aiming at the problems of information redundancy and low utilization of spatial information, this letter proposes a visual saliency-based extended morphological profile (VS-EMP) scheme. First, the morphological features are extracted by the EMP from the HSIs on several principal components. Second, the local binary pattern (LBP) is performed to extract the texture features from morphological scenes. Third, saliency features are captured according to the texture features in an approach of Boolean mapping saliency (BMS). Finally, spectral-spatial features are constructed by feature fusion and are further used for the classification of the HSIs. A number of experiments are performed, including using different classifiers to verify the performance of the proposed scheme, comparing with related variant algorithms, comparing time with deep learning, and testing learning ability in the absence of labeled samples. Experimental results indicate that the proposed method is significantly superior to the previous methods.

Hyperspectral image (HSI) consists of hundreds of continuous narrowbands with high spectral correlation, which would lead to the so-called Hughes phenomenon and the high computational cost in processing. Band selection (BS) has been proven to be effective in avoiding such problems by removing redundant bands. However, many existing BS methods separately estimate the significance for every single band and cannot fully consider the nonlinear and global interaction between spectral bands. In this article, by assuming that a complete HSI band set can be reconstructed from its few informative bands, we propose a unified BS framework, BS Network (BS-Net). The framework consists of a band attention module (BAM), which aims to explicitly model the nonlinear interdependences between spectral bands, and a reconstruction network (RecNet), which is used to restore the original HSI from the learned informative bands, resulting in a flexible architecture. The resulting framework is end-to-end trainable, making it easier to train from scratch and to combine with many existing networks. We implement two versions of BS-Nets, respectively, using fully connected networks (BS-Net-FC) and convolutional neural networks (BS-Net-Conv), and extensively compare their results with popular existing BS approaches on three real hyperspectral data sets, showing that the proposed BS-Nets can accurately select informative band subset with less redundancy and outperform the competitors in terms of classification accuracy with competitive time cost.

This paper presents a novel clustering method, named Laplacian regularized deep subspace clustering (LRDSC), for unsupervised hyperspectral image (HSI) classification. We introduce the Laplacian regularization into the subspace clustering to consider the manifold structure reflecting geometric information. To enable the subspace clustering, which works in linear space, to deal with the complicated HSI data with nonlinear characteristics, we combine the subspace clustering as a self-expressive layer with deep convolutional auto-encoder. Furthermore, the 3-D convolutions and deconvolutions with skip connections are utilized to make full extraction of the spectral-spatial information and full use of the historical feature maps produced by the network. We compare the results of the proposed method with six existing cluster methods on four real hyperspectral data sets, showing that the proposed method is able to achieve state-of-the-art performance.

Hyperspectral images (HSIs) consist of hundreds of continuous bands with high correlation, making it contain great abundant information. Band selection is an effective idea for removing redundant bands and preserving the physical significance at the same time. Popular sparse representation-based band selection commonly introduces an additional coefficient to combine error term and sparse constraint term, making it difficult to find out the optimal balance coefficient. In this letter, we propose a hybrid clustering-based band-selection approach based on using evolutionary multiobjective optimization to solve a sparse self-representation model constituted with two conflicting objectives. The proposed approach simultaneously minimizes two terms of the sparse representation model, avoiding the balance coefficient and producing a set of optimal solutions that are used to construct a similarity matrix for spectral clustering. Finally, a reduced band subset is determined by the cluster centers. We compare the results of the proposed approach with four existing band-selection methods for three real HSI data sets, showing that the proposed approach is able to effectively select representative bands with better classification accuracy.

Hyperspectral image (HSI) consists of hundreds of continuous narrow bands with high redundancy, resulting in the curse of dimensionality and an increased computation complexity in HSI classification. Many clustering-based band selection approaches have been proposed to deal with such a problem. However, a few of them consider the spectral and spatial relationship simultaneously. In this letter, we proposed a novel clustering-based band selection approach using deep subspace clustering (DSC). The proposed approach combines the subspace clustering task into a convolutional autoencoder by treating it as a self-expressive layer, enabling it to be trained end to end. The resulting network can fully extract the interaction of spectral bands based on using spatial information and nonlinear feature transformation. We compared the results of the proposed method with existing band selection methods for three widely used HSI data sets, showing that the proposed method is able to accurately select an informative band subset with remarkable classification accuracy.

Currently, deep neural networks (DNNs) are an important method for handling hyperspectral image (HSI) classification because of their good performance in image processing. However, DNNs' performance depends on a massive number of training data and hyperparameters that are carefully fine-tuned, which results in structural complexity and a time-consuming process. Deep forest is a novel deep learning method that does not need much training data and has a simple structure. In this paper, we first design a deep forest for spectral-based HSI classification and then propose an improved deep forest algorithm, named deep multigrained cascade forest (dgcForest), for spatial-based HSI classification. On the one hand, the cascade forest in dgcForest is used in multigrained scanning, which enhances the performance; on the other hand, a pooling layer is added after the multigrained scanning to reduce the output dimensions. To demonstrate that our proposed algorithm presents a good performance in HSI classification, we analyze the hyperparameters of deep forest and dgcForest and compare them with other methods on the biased and unbiased data sets, which illustrates that our method is superior to other state-of-the-art deep learning methods.

Deep neural networks have been successfully applied to many data mining problems in recent works. The training of deep neural networks relies heavily upon gradient descent methods, however, which may lead to the failure of training due to the vanishing gradient (or exploding gradient) and local optima problems. In this paper, we present SEvoAE method based on using Evolutionary Multiobjective optimization (EMO) algorithm to train single layer auto-encoder, and sequentially learning deeper representation in a stacking way. SEvoAE is able to achieve accurate feature representation with good sparseness by globally simultaneously optimizing two conflicting objective functions and allows users to flexibly design objective functions and evolutionary optimizers. We compare results of the proposed method with existing architectures for seven classification problems, showing that the proposed method is able to outperform existing methods with a reduced risk of overfitting the training data.

Face hallucination can improve the resolution of observed low-resolution (LR) face image to predict the high-resolution (HR) face image. In order to achieve good result performance, the training samples and local structure prior of face image are utilized by some approaches including Least Square Representation (LSR) and convex optimization to obtain the better representation coefficients. However, they do not pay more attention to the relationship between local-pixel structures of HR training samples and input LR face. Thus, the reconstruction coefficients they get are not optimal. Therefore, Optical Flow based face hallucination via weightedly-constrained representation(OFWCR) has been developed in this paper. Compared with LSR and Sparse Representation (SR), our method uses a warping HR training face image strategy to achieve better details from the input LR face. We also take into account the locality constraint in our effective representation scheme to reach locality and sparsity synchronously. Experiments show that our proposed scheme outperforms state-of-the-art approaches in common database.

Transfer learning has attracted more and more attention, and many scholars proposed some useful strategies. Boosting is the main strategy for transfer learning. In boosting, resampling is preferred over reweighting, and it can be applied to any base learner. In this paper, we propose a weighted-resampling method for transfer learning, called TrResampling. Firstly, resampling is applied to the data with heaven weight in the source domain, and the resampled data is used with the target data as the training data to build a classifier. Then the TrAdaBoost algorithm is used to adjust the weights of source data and target data. We discuss Decision Tree, Naive Bayes, and SVM as the base learner in TrResampling, and choose the suitable for TrResampling. In order to illustrate the performance of the proposed algorithm, we compare TrResampling with the state-of-the-art algorithm TrAdaBoost and the base learner Decision Tree, experimental results on UCI data sets indicate that TrResampling is superior to TrAdaBoost and Decision Tree on many data sets.

Extreme Learning Machine (ELM) is an elegant technique for training Single-hidden Layer Feedforward Networks (SLFNs) with extremely fast speed that attracts significant interest recently. One potential weakness of ELM is the random generation of the input weights and hidden biases, which may deteriorate the classification accuracy. In this paper, we propose a new Memetic Algorithm (MA) based Extreme Learning Machine (M-ELM) for classification problems. M-ELM uses Memetic Algorithm which is a combination of population-based global optimization technique and individual-based local heuristic search method to find optimal network parameters for ELM. The optimized network parameters will enhance the classification accuracy and generalization performance of ELM. Experiments and comparisons on 22 benchmark data sets demonstrate that M-ELM is able to provide highly competitive results compared with other state-of-the-art varieties of ELM algorithms.

The lack of labeled training data is a common issue in many machine learning applications. Semi-supervised learning addresses this issue by self-labeling unlabelled examples. Transfer learning tackles it from a different way: borrow labeled examples from a different but related domain (source domain) by assigning weights to those examples based on their suitability on the new domain (target domain). However, it is quite challenging to figure out the suitability. In this paper, we propose a different way for utilizing the labeled examples from source domain. That is, we use them only for labelling the unlabelled examples in the target domain. In this self-labelling, we use the idea of Tri-training. We call our new algorithm: TriTransfer. In TriTransfer, three initial classifiers are generated from the source data and the originally labeled data in the target domain, and an unlabeled example is labeled and added to the labeled data for a classifier if other two classifiers agree on its label. After an expanded labeled data set is obtained, we re-train the classifier. We repeat this process until no more change can be made. At the end, the final classifier, which is a weighted combination of the three classifiers, is output. We conduct an extensive empirical study on 34 UCI datasets, which shows that TriTransfer performs better than the state-of-art algorithms Transfer Boost, Tritraining, and NaiveBayes.