Research

20 pages, 11271 KiB

Open AccessArticle

Hyperspectral Image Super-Resolution under the Guidance of Deep Gradient Information

by Minghua Zhao, Jiawei Ning, Jing Hu and Tingting Li

Remote Sens. 2021, 13(12), 2382; https://doi.org/10.3390/rs13122382 - 18 Jun 2021

Cited by 12 | Viewed by 2125

Hyperspectral image (HSI) super-resolution has gained great attention in remote sensing, due to its effectiveness in enhancing the spatial information of the HSI while preserving the high spectral discriminative ability, without modifying the imagery hardware. In this paper, we proposed a novel HSI [...] Read more.

Hyperspectral image (HSI) super-resolution has gained great attention in remote sensing, due to its effectiveness in enhancing the spatial information of the HSI while preserving the high spectral discriminative ability, without modifying the imagery hardware. In this paper, we proposed a novel HSI super-resolution method via a gradient-guided residual dense network (G-RDN), in which the spatial gradient is exploited to guide the super-resolution process. Specifically, there are three modules in the super-resolving process. Firstly, the spatial mapping between the low-resolution HSI and the desired high-resolution HSI is learned via a residual dense network. The residual dense network is used to fully exploit the hierarchical features learned from all the convolutional layers. Meanwhile, the gradient detail is extracted via a residual network (ResNet), which is further utilized to guide the super-resolution process. Finally, an empirical weight is set between the fully obtained global hierarchical features and the gradient details. Experimental results and the data analysis on three benchmark datasets with different scaling factors demonstrated that our proposed G-RDN achieved favorable performance. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Graphical abstract

29 pages, 2978 KiB

Open AccessArticle

Hyperspectral Unmixing Based on Constrained Bilinear or Linear-Quadratic Matrix Factorization

by Fatima Zohra Benhalouche, Yannick Deville, Moussa Sofiane Karoui and Abdelaziz Ouamri

Remote Sens. 2021, 13(11), 2132; https://doi.org/10.3390/rs13112132 - 28 May 2021

Cited by 8 | Viewed by 2523

Abstract

Unsupervised hyperspectral unmixing methods aim to extract endmember spectra and infer the proportion of each of these spectra in each observed pixel when considering linear mixtures. However, the interaction between sunlight and the Earth’s surface is often very complex, so that observed spectra [...] Read more.

Unsupervised hyperspectral unmixing methods aim to extract endmember spectra and infer the proportion of each of these spectra in each observed pixel when considering linear mixtures. However, the interaction between sunlight and the Earth’s surface is often very complex, so that observed spectra are then composed of nonlinear mixing terms. This nonlinearity is generally bilinear or linear quadratic. In this work, unsupervised hyperspectral unmixing methods, designed for the bilinear and linear-quadratic mixing models, are proposed. These methods are based on bilinear or linear-quadratic matrix factorization with non-negativity constraints. Two types of algorithms are considered. The first ones only use the projection of the gradient, and are therefore linked to an optimal manual choice of their learning rates, which remains the limitation of these algorithms. The second developed algorithms, which overcome the above drawback, employ multiplicative projective update rules with automatically chosen learning rates. In addition, the endmember proportions estimation, with three alternative approaches, constitutes another contribution of this work. Besides, the reduction of the number of manipulated variables in the optimization processes is also an originality of the proposed methods. Experiments based on realistic synthetic hyperspectral data, generated according to the two considered nonlinear mixing models, and also on two real hyperspectral images, are carried out to evaluate the performance of the proposed approaches. The obtained results show that the best proposed approaches yield a much better performance than various tested literature methods. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Figure 1

30 pages, 10415 KiB

Open AccessArticle

Towards On-Board Hyperspectral Satellite Image Segmentation: Understanding Robustness of Deep Learning through Simulating Acquisition Conditions

by Jakub Nalepa, Michal Myller, Marcin Cwiek, Lukasz Zak, Tomasz Lakota, Lukasz Tulczyjew and Michal Kawulok

Remote Sens. 2021, 13(8), 1532; https://doi.org/10.3390/rs13081532 - 15 Apr 2021

Cited by 32 | Viewed by 8024

Abstract

Although hyperspectral images capture very detailed information about the scanned objects, their efficient analysis, transfer, and storage are still important practical challenges due to their large volume. Classifying and segmenting such imagery are the pivotal steps in virtually all applications, hence developing new [...] Read more.

Although hyperspectral images capture very detailed information about the scanned objects, their efficient analysis, transfer, and storage are still important practical challenges due to their large volume. Classifying and segmenting such imagery are the pivotal steps in virtually all applications, hence developing new techniques for these tasks is a vital research area. Here, deep learning has established the current state of the art. However, deploying large-capacity deep models on-board an Earth observation satellite poses additional technological challenges concerned with their memory footprints, energy consumption requirements, and robustness against varying-quality image data, with the last problem being under-researched. In this paper, we tackle this issue, and propose a set of simulation scenarios that reflect a range of atmospheric conditions and noise contamination that may ultimately happen on-board an imaging satellite. We verify their impact on the generalization capabilities of spectral and spectral-spatial convolutional neural networks for hyperspectral image segmentation. Our experimental analysis, coupled with various visualizations, sheds more light on the robustness of the deep models and indicate that specific noise distributions can significantly deteriorate their performance. Additionally, we show that simulating atmospheric conditions is key to obtaining the learners that generalize well over image data acquired in different imaging settings. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Figure 1

24 pages, 2922 KiB

Open AccessArticle

Spectral-Spatial Joint Classification of Hyperspectral Image Based on Broad Learning System

by Guixin Zhao, Xuesong Wang, Yi Kong and Yuhu Cheng

Remote Sens. 2021, 13(4), 583; https://doi.org/10.3390/rs13040583 - 06 Feb 2021

Cited by 9 | Viewed by 2480

Abstract

At present many researchers pay attention to a combination of spectral features and spatial features to enhance hyperspectral image (HSI) classification accuracy. However, the spatial features in some methods are utilized insufficiently. In order to further improve the performance of HSI classification, the [...] Read more.

At present many researchers pay attention to a combination of spectral features and spatial features to enhance hyperspectral image (HSI) classification accuracy. However, the spatial features in some methods are utilized insufficiently. In order to further improve the performance of HSI classification, the spectral-spatial joint classification of HSI based on the broad learning system (BLS) (SSBLS) method was proposed in this paper; it consists of three parts. Firstly, the Gaussian filter is adopted to smooth each band of the original spectra based on the spatial information to remove the noise. Secondly, the test sample’s labels can be obtained using the optimal BLS classification model trained with the spectral features smoothed by the Gaussian filter. At last, the guided filter is performed to correct the BLS classification results based on the spatial contextual information for improving the classification accuracy. Experiment results on the three real HSI datasets demonstrate that the mean overall accuracies (OAs) of ten experiments are 99.83% on the Indian Pines dataset, 99.96% on the Salinas dataset, and 99.49% on the Pavia University dataset. Compared with other methods, the proposed method in the paper has the best performance. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Graphical abstract

21 pages, 2493 KiB

Open AccessArticle

CSR-Net: Camera Spectral Response Network for Dimensionality Reduction and Classification in Hyperspectral Imagery

by Yunhao Zou, Ying Fu, Yinqiang Zheng and Wei Li

Remote Sens. 2020, 12(20), 3294; https://doi.org/10.3390/rs12203294 - 10 Oct 2020

Cited by 5 | Viewed by 2448

Abstract

Hyperspectral image (HSI) classification has become one of the most significant tasks in the field of hyperspectral analysis. However, classifying each pixel in HSI accurately is challenging due to the curse of dimensionality and limited training samples. In this paper, we present an [...] Read more.

Hyperspectral image (HSI) classification has become one of the most significant tasks in the field of hyperspectral analysis. However, classifying each pixel in HSI accurately is challenging due to the curse of dimensionality and limited training samples. In this paper, we present an HSI classification architecture called camera spectral response network (CSR-Net), which can learn the optimal camera spectral response (CSR) function for HSI classification problems and effectively reduce the spectral dimensions of HSI. Specifically, we design a convolutional layer to simulate the capturing process of cameras, which learns the optimal CSR function for HSI classification. Then, spectral and spatial features are further extracted by spectral and spatial attention modules. On one hand, the learned CSR can be implemented physically and directly used to capture scenes, which makes the image acquisition process more convenient. On the other hand, compared with ordinary HSIs, we only need images with far fewer bands, without sacrificing the classification precision and avoiding the curse of dimensionality. The experimental results of four popular public hyperspectral datasets show that our method, with only a few image bands, outperforms state-of-the-art HSI classification methods which utilize the full spectral bands of images. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Graphical abstract

25 pages, 2847 KiB

Open AccessArticle

Blind Unmixing of Hyperspectral Remote Sensing Data: A New Geometrical Method Based on a Two-Source Sparsity Constraint

by Djaouad Benachir, Yannick Deville, Shahram Hosseini and Moussa Sofiane Karoui

Remote Sens. 2020, 12(19), 3198; https://doi.org/10.3390/rs12193198 - 30 Sep 2020

Cited by 1 | Viewed by 2283

Abstract

Blind source separation (or unmixing) methods process a set of mixed signals, which are typically linear memoryless combinations of source signals, so as to estimate these unknown source signals and/or combination coefficients. These methods have been extensively applied to hyperspectral images in the [...] Read more.

Blind source separation (or unmixing) methods process a set of mixed signals, which are typically linear memoryless combinations of source signals, so as to estimate these unknown source signals and/or combination coefficients. These methods have been extensively applied to hyperspectral images in the field of remote sensing, because the reflectance spectrum of each image pixel is often a mixture of elementary contributions, due to the limited spatial resolution of hyperspectral remote sensing sensors. Each spatial source signal then corresponds to a pure material, and its value in each pixel is equal to the “abundance fraction” of the corresponding Earth surface covered by that pure material. The mixing coefficients then form the pure material spectra. Various unmixing methods have been designed for this data model and the majority of them are either geometrical or statistical, or even based on sparse regressions. Various such unmixing techniques mainly consider assumptions that are related to the presence or absence of pure pixels (i.e., pixels which contain only one pure material). The case when, for each pure material, the image includes at least one pixel or zone which only contains that material yielded attractive unmixing methods, but corresponds to a stringent sparsity condition. We here aim at relaxing that condition, by only requesting a few tiny pixel zones to contain two pure materials. The proposed linear and geometrical sparse-based, blind (or unsupervised) unmixing method first automatically detects these zones. Each such zone defines a line in the data representation space. These lines are then estimated and clustered. The pairs of cluster centers, corresponding to lines, which have an intersection, yield the spectra of pure materials, forming the columns of the mixing matrix. Finally, the proposed method derives all abundance fractions, i.e., source signals, by using a least squares method with a non-negativity constraint. This method is applied to realistic synthetic images and is shown to outperform various methods from the literature. Indeed, and for the conducted experiments, when considering the pure material spectra extraction, the obtained improvements, for the considered spectral angle mapper performance criterion, vary between

{0.02}^{\circ}

and

{12.43}^{\circ}

. For the abundance fractions estimation, the proposed technique is able to achieve a normalized mean square error lower than

0.01 %

, while the tested literature methods yield errors greater than

0.1 %

. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Graphical abstract

32 pages, 2544 KiB

Open AccessArticle

Using a Panchromatic Image to Improve Hyperspectral Unmixing

by Simon Rebeyrol, Yannick Deville, Véronique Achard, Xavier Briottet and Stephane May

Remote Sens. 2020, 12(17), 2834; https://doi.org/10.3390/rs12172834 - 01 Sep 2020

Cited by 4 | Viewed by 3165

Abstract

Hyperspectral unmixing is a widely studied field of research aiming at estimating the pure material signatures and their abundance fractions from hyperspectral images. Most spectral unmixing methods are based on prior knowledge and assumptions that induce limitations, such as the existence of at [...] Read more.

Hyperspectral unmixing is a widely studied field of research aiming at estimating the pure material signatures and their abundance fractions from hyperspectral images. Most spectral unmixing methods are based on prior knowledge and assumptions that induce limitations, such as the existence of at least one pure pixel for each material. This work presents a new approach aiming to overcome some of these limitations by introducing a co-registered panchromatic image in the unmixing process. Our method, called Heterogeneity-Based Endmember Extraction coupled with Local Constrained Non-negative Matrix Factorization (HBEE-LCNMF), has several steps: a first set of endmembers is estimated based on a heterogeneity criterion applied on the panchromatic image followed by a spectral clustering. Then, in order to complete this first endmember set, a local approach using a constrained non-negative matrix factorization strategy, is proposed. The performance of our method, in regards of several criteria, is compared to those of state-of-the-art methods obtained on synthetic and satellite data describing urban and periurban scenes, and considering the French HYPXIM/HYPEX2 mission characteristics. The synthetic images are built with real spectral reflectances and do not contain a pure pixel for each endmember. The satellite images are simulated from airborne acquisition with the spatial and spectral features of the mission. Our method demonstrates the benefit of a panchromatic image to reduce some well-known limitations in unmixing hyperspectral data. On synthetic data, our method reduces the spectral angle between the endmembers and the real material spectra by 46% compared to the Vertex Component Analysis (VCA) and N-finder (N-FINDR) methods. On real data, HBEE-LCNMF and other methods yield equivalent performance, but, the proposed method shows more robustness over the data sets compared to the tested state-of-the-art methods. Moreover, HBEE-LCNMF does not require one to know the number of endmembers. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Graphical abstract

28 pages, 41587 KiB

Open AccessArticle

A Developed Siamese CNN with 3D Adaptive Spatial-Spectral Pyramid Pooling for Hyperspectral Image Classification

by Mengbin Rao, Ping Tang and Zheng Zhang

Remote Sens. 2020, 12(12), 1964; https://doi.org/10.3390/rs12121964 - 18 Jun 2020

Cited by 25 | Viewed by 3652

Abstract

Since hyperspectral images (HSI) captured by different sensors often contain different number of bands, but most of the convolutional neural networks (CNN) require a fixed-size input, the generalization capability of deep CNNs to use heterogeneous input to achieve better classification performance has become [...] Read more.

Since hyperspectral images (HSI) captured by different sensors often contain different number of bands, but most of the convolutional neural networks (CNN) require a fixed-size input, the generalization capability of deep CNNs to use heterogeneous input to achieve better classification performance has become a research focus. For classification tasks with limited labeled samples, the training strategy of feeding CNNs with sample-pairs instead of single sample has proven to be an efficient approach. Following this strategy, we propose a Siamese CNN with three-dimensional (3D) adaptive spatial-spectral pyramid pooling (ASSP) layer, called ASSP-SCNN, that takes as input 3D sample-pair with varying size and can easily be transferred to another HSI dataset regardless of the number of spectral bands. The 3D ASSP layer can also extract different levels of 3D information to improve the classification performance of the equipped CNN. To evaluate the classification and generalization performance of ASSP-SCNN, our experiments consist of two parts: the experiments of ASSP-SCNN without pre-training and the experiments of ASSP-SCNN-based transfer learning framework. Experimental results on three HSI datasets demonstrate that both ASSP-SCNN without pre-training and transfer learning based on ASSP-SCNN achieve higher classification accuracies than several state-of-the-art CNN-based methods. Moreover, we also compare the performance of ASSP-SCNN on different transfer learning tasks, which further verifies that ASSP-SCNN has a strong generalization capability. Full article

(This article belongs to the Special Issue Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing)

► Show Figures

Graphical abstract

Journal Menu

Journal Browser

Machine Learning and Pattern Analysis in Hyperspectral Remote Sensing

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (8 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI