Pixel In Hyperspectral Image Research Articles

A Spectral–Spatial Method Based on Fractional Fourier Transform and Collaborative Representation for Hyperspectral Anomaly Detection

Anomaly detection (AD) is one of the most important tasks in hyperspectral image (HSI) processing. Most of the traditional AD methods fail to take the advantage of rich spatial information of HSIs and suffer the problem of noise contamination. To solve these problems, we propose a fractional Fourier transform and collaborative representation-based spectral-spatial hyperspectral anomaly detector (SSFrFTCRD). Different from the previous work, fractional Fourier transform (FrFT) is associated with collaborative representation detector (CRD) in the proposed method. FrFT can transfer HSI pixels into a FrFT domain, which can suppress noise and improve the discrimination between background and anomalies. By taking advantage of the CRD, the SSFrFTCRD can adaptively estimate the background through a sliding dual window without assuming its distribution. Furthermore, both spectral and spatial information are utilized to enhance the performance of the proposed detector. Experiments show that the proposed anomaly detector SSFrFTCRD can achieve superior results compared with the other state-of-the-art methods.

Spectral–Spatial Hyperspectral Image Classification Using Dual-Channel Capsule Networks

Deep learning methods have shown their marvel performance on hyperspectral image (HSI) classification tasks. In particular, algorithms based on convolution neural network (CNN) outperformed most of the conventional machine learning-based algorithms and have become the mainstream of the current HSI classification research works. Recently, a newly proposed neural network called capsule network (CapsNet) showed its potential to replace the CNNs in various classification tasks with its amazing performance. In this letter, we proposed a new network architecture based on the CapsNet for HSI classification tasks, called dual-channel capsule network (DCCapsNet). Our DCCapsNet model extracts the features from spectral and spatial domains, respectively, with two separate convolution channels and then concatenates and feeds them into the following capsule layers to classify each of the HSI pixels. The model was trained and validated on four real HSI data sets and achieved high accuracy. We also compared our network with some of the state-of-the-art models and found that our model outperformed these competitor models.

Multidimensional relation learning for hyperspectral image classification

Classification is a fundamental technique in hyperspectral image understanding field, which is to give each pixel a specific semantic label based on its characteristics automatically. Recently, even though convolutional neural network (CNN) based methods have been applied to hyperspectral image classification, their performances are not in accordance with our expectation. The reason is that, most of the existing methods can not exploit the intrinsic characteristics of different pixels in hyperspectral image effectively, they often ignore the inherent relationships among different spatial pixels, different spectral bands and different category dependencies. Specifically, the relationships among different spatial pixels mean that the similarities of attribute and location information among different pixels. The relationships among different spectral bands mean that the different contributions of them to the final classification. The relationships among different categories mainly mean that the severe sample imbalance problem in hyperspectral image classification task. To address the aforementioned problems, a deep convolutional network based multidimensional relation learning method is proposed in this paper. Firstly, we propose a weighted gaussian mask based spatial relation learning module to alleviate the spatial diffusion issue. Besides, we propose a channel attention alike spectral relation learning module to alleviate the interferences from noisy and redundant bands. Additionally, we propose an integrated decision based category relation learning module to alleviate the ill-conditioned classifying issue from imbalanced samples. Finally, we test the proposed method on four public and challenging datasets, and the experimental results validate the effectiveness and robustness of our method.

Spectral–Spatial-Weighted Multiview Collaborative Sparse Unmixing for Hyperspectral Images

Spectral unmixing is an important task in hyperspectral image (HSI) analysis and processing. Sparse representation has become a promising semisupervised method for remotely sensed hyperspectral unmixing and incorporating the spectral or spatial information to improve the spectral unmixing results under a weighted sparse unmixing framework is a recent trend. While most methods focus on analyzing HSI by exploring the spatial information, it is known that hyperspectral data are characterized by its large contiguous set of wavelengths. This information can be naturally used to improve the representation of pixels in HSI. In order to take the advantage of the hyper spectral information as well as the spatial information for hyperspectral unmixing, in this article, we explore and introduce a multiview data processing approach through spectral partitioning to benefit from the abundant spectral information in HSI. Some important findings on the application of multiview data set in sparse unmixing are discussed. Meanwhile, we develop a new spectral–spatial-weighted multiview collaborative sparse unmixing (MCSU) model to tackle such a multiview data set. The MCSU uses a weighted sparse regularizer, which includes both multiview spectral and spatial weighting factors to further impose sparsity on the fractional abundances. The weights are adaptively updated associated with the abundances, and the proposed MCSU can be solved by the alternating direction method of multipliers efficiently. The experimental results on both the simulated and real hyperspectral data sets demonstrate the effectiveness of the proposed MCSU, which can significantly improve the abundance estimation results.

Pixel-Aware Deep Function-Mixture Network for Spectral Super-Resolution

Spectral super-resolution (SSR) aims at generating a hyperspectral image (HSI) from a given RGB image. Recently, a promising direction is to learn a complicated mapping function from the RGB image to the HSI counterpart using a deep convolutional neural network. This essentially involves mapping the RGB context within a size-specific receptive field centered at each pixel to its spectrum in the HSI. The focus thereon is to appropriately determine the receptive field size and establish the mapping function from RGB context to the corresponding spectrum. Due to their differences in category or spatial position, pixels in HSIs often require different-sized receptive fields and distinct mapping functions. However, few efforts have been invested to explicitly exploit this prior.To address this problem, we propose a pixel-aware deep function-mixture network for SSR, which is composed of a new class of modules, termed function-mixture (FM) blocks. Each FM block is equipped with some basis functions, i.e., parallel subnets of different-sized receptive fields. Besides, it incorporates an extra subnet as a mixing function to generate pixel-wise weights, and then linearly mixes the outputs of all basis functions with those generated weights. This enables us to pixel-wisely determine the receptive field size and the mapping function. Moreover, we stack several such FM blocks to further increase the flexibility of the network in learning the pixel-wise mapping. To encourage feature reuse, intermediate features generated by the FM blocks are fused in late stage, which proves to be effective for boosting the SSR performance. Experimental results on three benchmark HSI datasets demonstrate the superiority of the proposed method.

Subspace Structure Regularized Nonnegative Matrix Factorization for Hyperspectral Unmixing

Hyperspectral unmixing is a crucial task for hyperspectral images (HSIs) processing, which estimates the proportions of constituent materials of a mixed pixel. Usually, the mixed pixels can be approximated using a linear mixing model. Since each material only occurs in a few pixels in real HSI, sparse nonnegative matrix factorization (NMF), and its extensions are widely used as solutions. Some recent works assume that materials are distributed in certain structures, which can be added as constraints to sparse NMF model. However, they only consider the spatial distribution within a local neighborhood and define the distribution structure manually, while ignoring the real distribution of materials that is diverse in different images. In this article, we propose a new unmixing method that learns a subspace structure from the original image and incorporate it into the sparse NMF framework to promote unmixing performance. Based on the self-representation property of data points lying in the same subspace, the learned subspace structure can indicate the global similar graph of pixels that represents the real distribution of materials. Then the similar graph is used as a robust global spatial prior which is expected to be maintained in the decomposed abundance matrix. The experiments conducted on both simulated and real-world HSI datasets demonstrate the superior performance of our proposed method.

Hyperspectral Image Classification via Slice Sparse Coding Tensor Based Classifier With Compressive Dimensionality Reduction

Tensor representation is the most natural and effective way to preserve the structural information of hyperspectral image (HSI), and thus is very beneficial to HSI processing. This paper represents the spectral features of each testing pixel and its spatial neighbors as a Spatial Neighbor Tensor (SNT), whose spectral vectors can be simultaneously sparse coded by the spectral vectors of a few common training pixels. The obtained sparse coding coefficients could be regarded as a Slice Sparse Coding Tensor (SSCT), which can be adaptively learnt and utilized to predict the labels of HSI pixels. Furthermore, to improve the efficiency, the Compressive Dimensionality Reduction (CDR) is introduced into tensor slice sparse coding for optimizing SSCT and thus an SSCT based Classifier with CDR (SSCTC-CDR) is proposed for hyperspectral image classification (HIC). The performance of SSCTC-CDR is evaluated on three real HSI data, and the results show that it can obtain high-accurate classification result with relatively low computation lost.

Hyperspectral image classification using graph-based wavelet transform

ABSTRACTGraph-based methods are developed to efficiently extract data information. In particular, these methods are adopted for high-dimensional data classification by exploiting information residing on weighted graphs. In this paper, we propose a new hyperspectral texture classifier based on graph-based wavelet transform. This recent graph transform allows extracting textural features from a constructed weighted graph using sparse representative pixels of hyperspectral image. Different measurements of spectral similarity between representative pixels are tested to decorrelate close pixels and improve the classification precision. To achieve the hyperspectral texture classification, Support Vector Machine is applied on spectral graph wavelet coefficients. Experimental results obtained by applying the proposed approach on Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and Reflective Optics System Imaging Spectrometer (ROSIS) datasets provide good accuracy which could exceed 98.7%. Compared to other famous classification methods as conventional deep learning-based methods, the proposed method achieves better classification performance. Results have shown the effectiveness of the method in terms of robustness and accuracy.

Hyperspectral Anomaly Detection via Tensor- Based Endmember Extraction and Low-Rank Decomposition

Due to the limited resolution of hyperspectral sensors, anomalous targets expressed with subpixels are often mixed with nonhomogeneous backgrounds. This fact makes anomalies difficult to distinguish from the surrounding background. From this perspective, a novel hyperspectral anomaly detection (AD) algorithm based on endmember extraction and low-rank representation (LRR) is proposed. For the characteristics of pixels in hyperspectral images (HSIs), the proposed algorithm employs an endmember extraction technology to yield an abundance matrix for AD, thereby gathering more feature information compared with the direct use of a raw image. In addition, a dictionary construction strategy based on Tucker decomposition, and the ${k}$ -means++ clustering method is proposed to make the dictionary more stable and discriminative. An LRR method based on the dictionary is applied to obtain a sparse residual matrix. Finally, anomalies can be determined by the response of the residual matrix. Experiments on three hyperspectral data sets validate the performance of the proposed algorithm.

Hyperspectral image classification using nearest regularized subspace with Manhattan distance

Nearest regularized subspace (NRS) has been recently proposed for hyperspectral image (HSI) classification. The NRS outperforms both collaborative representation classification and sparse representation-based techniques because the NRS makes use of the distance-weighted Tikhonov regularization to ensure appropriate representation from similar samples within-class. However, typical NRS only considers Euclidean distance, which may be suboptimal to resolve the problem of sensitivity in the absolute magnitude of a spectrum. An NRS-Manhattan distance (MD) strategy is proposed for HSI classification. The proposed distance metric controls over magnitude change and emphasizes the shape of the spectrum. Furthermore, the MD metric uses the entire information of the spectral bands in full dimensionality of the HSI pixels, which makes NRS-MD a more efficient pixelwise classifier. Validations are done with several hyperspectral data, i.e., Indian Pines, Botswana, Salinas, and Houston. Results demonstrate that the proposed NRS-MD is superior to other state-of-the-art methods.

Spectral Adversarial Feature Learning for Anomaly Detection in Hyperspectral Imagery

Theoretically, hyperspectral images (HSIs) are capable of providing subtle spectral differences between different materials, but in fact, it is difficult to distinguish between background and anomalies because the samples of anomalous pixels in HSIs are limited and susceptible to background and noise. To explore the discriminant features, a spectral adversarial feature learning (SAFL) architecture is specially designed for hyperspectral anomaly detection in this article. In addition to reconstruction loss, SAFL also introduces spectral constraint loss and adversarial loss in the network with batch normalization to extract the intrinsic spectral features in deep latent space. To further reduce the false alarm rate, we present an iterative optimization approach by a weighted suppression function that depends on the contribution rate of each feature to the detection. In particular, the structure tensor matrix is adopted to adaptively calculate the contribution rate of each feature. Benefiting from these improvements, the proposed method is superior to the typical and state-of-the-art methods either in detection probability or false alarm rate.

Bilateral Kernel Extraction from PCA for Classification of Hyperspectral Images

The improved spatial and spectral resolution in the advanced Hyperspectral (HS) sensors results in images with rich information per pixel. Hence, the development of efficient spatial–spectral feature extraction (FE) techniques is very crucial for a proper characterization of the objects on ground. In this paper, an attempt has been made to develop a simple, yet effective spatial–spectral FE algorithm. In the proposed approach, the following steps are performed. First, Principal Component Analysis (PCA) was applied on the original hyperspectral image (HSI) and the most significant principal component was extracted. Then, the Bilateral Filter (BF), which acts as an edge-preserving filter, was applied on the selected principal component to extract kernel for each pixel in HSI. The extracted kernel bank is then applied on the original HSI. As in general, the principal component image is edge informative, and the BF is an edge-preserving filter; therefore, the extracted kernel bank can be applied on the original HSI to extract spatial–spectral features. Finally, with the help of these features, the performance of Support Vector Machine (SVM) classifier is evaluated. The proposed approach is validated on three popular hyperspectral data sets, namely, Indian Pines, Pavia University, and Botswana. The experimental results reveal that learning the edge information from a reference image (in the present context PCA) is quite essential, rather than applying the edge-preserving filters directly on the HSI. Theoretically, this holds true, as a unique edge (ground) information exists for an HSI, while in reality, the edges have variations due to variation in reflectance over bands.

Spectral-Spatial Hyperspectral Image Classification with Superpixel Pattern and Extreme Learning Machine

Spectral-spatial classification of hyperspectral images (HSIs) has recently attracted great attention in the research domain of remote sensing. It is well-known that, in remote sensing applications, spectral features are the fundamental information and spatial patterns provide the complementary information. With both spectral features and spatial patterns, hyperspectral image (HSI) applications can be fully explored and the classification performance can be greatly improved. In reality, spatial patterns can be extracted to represent a line, a clustering of points or image texture, which denote the local or global spatial characteristic of HSIs. In this paper, we propose a spectral-spatial HSI classification model based on superpixel pattern (SP) and kernel based extreme learning machine (KELM), called SP-KELM, to identify the land covers of pixels in HSIs. In the proposed SP-KELM model, superpixel pattern features are extracted by an advanced principal component analysis (PCA), which is based on superpixel segmentation in HSIs and used to denote spatial information. The KELM method is then employed to be a classifier in the proposed spectral-spatial model with both the original spectral features and the extracted spatial pattern features. Experimental results on three publicly available HSI datasets verify the effectiveness of the proposed SP-KELM model, with the performance improvement of 10% over the spectral approaches.

Region‐division‐based joint sparse representation classification for hyperspectral images

In this article, a region-division-based joint sparse representation classification (RDJSRC) method is proposed to solve the heterogeneous region problem in the joint sparse representation classification (JSRC) method used in hyperspectral image (HSI) classification. The RDJSRC method incorporates regional information, obtained by the hidden Markov random field (HMRF), into the JSRC to reduce the interference of heterogeneous pixels in the neighbourhood of the test pixel and finally improve the classification performance. The framework of this method is as follows. The first several principal components (PCs) are initially selected to be the new HSI by transforming the original HSI with the PC analysis algorithm. Then, the regional information containing the spatial structure of the HSI is obtained by applying the HMRF algorithm to the first PC. Through incorporating this regional information into the JSRC procedure, the initial label of the test pixel can be jointly determined by the new HSI pixels within the homogeneity in the search window. Ultimately, the final label of the test pixel is determined by a voting strategy based on multiple classification results. Compared with several classification methods, experimental results, indicate that this method achieves improvement from 2 to 3% in HSI classification.

Robust Hyperspectral Image Domain Adaptation With Noisy Labels

In hyperspectral image (HSI) classification, domain adaptation (DA) methods have been proved effective to address unsatisfactory classification results caused by the distribution difference between training (i.e., source domain) and testing (i.e., target domain) pixels. However, these methods rely on accurate labels in source domain, and seldom consider the performance drop resulted by noisy label, which often happens since labeling pixel in HSI is a challenging task. To improve the robustness of DA method to label noise, we propose a new unsupervised HSI DA method, which is constructed from both feature-level and classifier-level. First, a linear transformation function is learned in feature-level to align the source (domain) subspace with the target (domain) subspace. Then, a robust low-rank representation based classifier is developed to well cope with the features obtained from the aligned subspace. Since both subspace alignment and the classifier are immune to noisy labels, the proposed method obtains good classification results when confronting with noisy labels in source domain. Experimental results on two DA benchmarks demonstrate the effectiveness of the proposed method.

PSASL: Pixel-Level and Superpixel-Level Aware Subspace Learning for Hyperspectral Image Classification

The performance of hyperspectral image (HSI) classification relies on the pixel information obtained from hundreds of contiguous and narrow spectral bands. Existing approaches, however, are limited to exploit an appropriate latent subspace for data representation within the pixel-level or superpixel-level. To utilize spectral information and spatial correlation among pixels in HSI and avoid the “salt-and-pepper” problem generated in the pixel-based HSI classification, a novel pixel-level and superpixel-level aware subspace learning method called PSASL is developed. The PSASL constructs the subspace learning framework based on the reconstruction independent component analysis algorithm. The spectral–spatial graph regularization and label space regularization are developed as the pixel-level constraints. To avoid the “salt-and-pepper” problem generated in the pixel-based classification methods, superpixel-level constraints are introduced for integrating the data representations defined in the subspace and class probabilities of the pixels in the same superpixel. The subspace learning and the pixel-level regularization are combined with the superpixel-level regularization to form a unified objective function. The solution to the objective function is efficiently achieved by employing a customized iterative algorithm, and it converges very fast. A discriminative data representation and a universal multiclass classifier are learned simultaneously. We test the PSASL on three widely used HSI data sets. Experimental results demonstrate the superior performance of our method over many recently proposed methods in HSI classification.

Abundance Estimation Using Discontinuity Preserving and Sparsity-Induced Priors

Abundance estimation is used to infer the proportions of endmembers with the given endmember signatures and reflectance value at each location. In this paper, we propose a two-phase iterative approach to estimate the abundances (fractions) of materials (endmembers) from the pixels of hyperspectral images (HSIs) by using the energy minimization framework. A linear mixture model is used to define the data term. We observe that abundance maps have homogeneous regions with limited discontinuity, and they exhibit spatial redundancy. Hence, we use inhomogeneous Gaussian Markov random field (IGMRF) and sparsity-induced priors as the regularization terms. While the IGMRF prior captures the smoothness and preserves discontinuities among abundance values, the sparsity-induced prior accounts for redundancy. We calculate the IGMRF parameters at every pixel location and learn a dictionary and the sparse representation for abundances using the initial estimate in phase 1, while the final abundance maps are estimated in phase 2. In order to learn the sparsity, we use the approach based on K-singular value decomposition. Both the IGMRF and sparseness parameters are initialized using an initial estimate of abundances and refined using the two-phase iterative approach. The experiments are conducted on synthetic hyperspectral HSIs with different noise levels, as well as on two real HSIs. The results are qualitatively and quantitatively compared with state-of-the-art approaches. Experimental results demonstrate the effectiveness of the proposed approach.

Hyperspectral Image Features Classification Using Deep Learning Recurrent Neural Networks.

The implementation of Deep learning (DL) techniques, Object detection and classification has achieved remarkable results in remote sensing application. Deep learning with Recurrent Neural Network (RNN) technique on hyper-spectral data has been presented here. The only model which can analyze the hyper-spectral pixels as the sequence of information and also to identify the additional information categories through network reasoning is RNN model. This is first time that the framework of RNN has been introduced for the classification of hyper spectral Image. An activation function is proposed by the DL-RNN and also the parameter rectified functions for analyzing the sequence of data in the hyper-spectral images. Throughout the training procedure, the higher learning rates are fairly used by the activation function which has been proposed by avoiding the risk of divergence. In the proposed system the pixels of hyper-spectral images through the sequential perspective has been processed for capturing the sequence based data. The experimental result also shows that the proposed RNN has produced the improved F- score than the traditional deep learning methods.

An Effective Classification Scheme for Hyperspectral Image Based on Superpixel and Discontinuity Preserving Relaxation

Hyperspectral image (HSI) classification is one of the most active topics in remote sensing. However, it is still a nontrivial task to classify the hyperspectral data accurately, since HSI always suffers from a large number of noise pixels, the complexity of the spatial structure of objects and the spectral similarity between different objects. In this study, an effective classification scheme for hyperspectral image based on superpixel and discontinuity preserving relaxation (DPR) is proposed to discriminate land covers of interest. A novel technique for measuring the similarity of a pair of pixels in HSI is suggested to improve the simple linear iterative clustering (SLIC) algorithm. Unlike the existing application of SLIC technique to HSI, the improved SLIC algorithm can be directly used to segment HSI into superpixels without using principal component analysis in advance, and is free of parameters. Furthermore, the proposed three-step classification scheme explores how to effectively use the global spectral information and local spatial structure of hyperspectral data for HSI classification. Compared with the existing two-step classification framework, the use of DPR technology in preprocessing significantly improves the classification accuracy. The effectiveness of the proposed method is verified on three public real hyperspectral datasets. The comparison results of several competitive methods show the superiority of this scheme.

Sparsity-Constrained Distributed Unmixing of Hyperspectral Data

Spectral unmixing (SU) is a technique to characterize mixed pixels in hyperspectral images measured by remote sensors. Most of the spectral unmixing algorithms are developed using the linear mixing models. To estimate endmembers and fractional abundance matrices in a blind problem, nonnegative matrix factorization (NMF) and its developments are widely used in the SU problem. One of the constraints which was added to NMF is sparsity, that was regularized by L q norm. In this paper, a new algorithm based on distributed optimization is suggested for spectral unmixing. In the proposed algorithm, a network including single-node clusters is employed. Each pixel in the hyperspectral images is considered as a node in this network. The sparsity-constrained distributed unmixing is optimized with diffusion least mean p-power (LMP) strategy, and then the update equations for fractional abundance and signature matrices are obtained. Afterward, the proposed algorithm is analyzed for different values of LMP power and L q norms. Simulation results based on defined performance metrics illustrate the advantage of the proposed algorithm in spectral unmixing of hyperspectral data compared with other methods.