Related references
Note: Only part of the references are listed.
Article
Remote Sensing
Yue Xu et al.
Summary: This paper constructs a large HSSR dataset based on aerial hyperspectral imagery and proposes a new deep learning model, 3D-HRNet, for interpreting HSSR images. The dataset is the largest of its kind to date and can be used for spatial-spectral feature extraction. The 3D-HRNet model shows good interpreting ability for the dataset, with a high FWIoU score of 80.54%.
GEO-SPATIAL INFORMATION SCIENCE
(2023)
Article
Geochemistry & Geophysics
Muhammad Ahmad et al.
Summary: This study proposes a 3-D CNN model that utilizes both spatial-spectral feature maps to improve the performance of HSIC. By processing small overlapping 3-D patches and generating 3-D feature maps, the model demonstrates remarkable performance in terms of accuracy and computational time.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Uzair Aslam Bhatti et al.
Summary: This article introduces a novel HSI classification algorithm, LSPGF, which utilizes LSP and Gabor filtering to extract deeper features from original hyperspectral data, achieving higher classification accuracy compared to other algorithms.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Le Sun et al.
Summary: In this article, the spectral-spatial feature tokenization transformer (SSFTT) method is proposed to capture spectral-spatial and high-level semantic features. Experimental analysis confirms that this method outperforms other deep learning methods in terms of computation time and classification performance.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geography, Physical
Shivam Pande et al.
Summary: This article introduces a more efficient method for hyperspectral image classification, which utilizes self-looping convolutional neural networks for better feature extraction. Compared to traditional CNNs, this method has higher information flow efficiency and is able to extract abstract representations at multiple spatial levels.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Remote Sensing
Yuanlin Wang et al.
Summary: In this study, we propose a probabilistic neighbourhood pooling-based attention network (PNPAN) for hyperspectral image (HSI) classification. By dividing the input HSI cubes into ring-shaped neighbouring regions and probabilistically selecting them as pooling regions, PNPAN is able to better consider the spatial dependencies of the central pixel and extract spectral-spatial features, achieving promising results for HSI classification with limited training samples.
REMOTE SENSING LETTERS
(2022)
Article
Automation & Control Systems
He Sun et al.
Summary: In this article, an adaptive distance-based band hierarchy (ADBH) clustering framework is proposed for unsupervised band selection (UBS) in hyperspectral image (HSI) processing. The framework helps to avoid noisy bands while reflecting the hierarchical data structure of HSI. The experiments on four datasets from two HSI systems have validated the superiority of the proposed framework.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Remote Sensing
Yifan Sun et al.
Summary: We propose a fully octave convolution network with a pyramid attention mechanism (FOctConvPA) for whole Hyperspectral Image (HSI) classification. The method effectively extracts global context and long-distance information to improve classification performance. Experimental results on real benchmark hyperspectral datasets demonstrate that the proposed method achieves better classification performance compared with other state-of-the-art deep learning models.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Article
Computer Science, Information Systems
Heena Patel et al.
Summary: This paper discusses an approach for hyperspectral image classification using an autoencoder and convolutional neural network. The method enhances features in HSI data through preprocessing and optimized weights, showing superiority in classification compared to other deep learning methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Jinxiong Gao et al.
Summary: In this paper, a Bi-LSTM-based multi-scale dense attention framework, MBDA-Net, is proposed for hyperspectral image classification. The framework utilizes multi-scale feature extraction and attention mechanism for feature selection, and employs bi-directional LSTM to capture contextual semantic information. Experimental results demonstrate the effectiveness of the proposed method in identifying hyperspectral images.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Engineering, Multidisciplinary
Zhiyong Li et al.
Summary: This paper proposes an HSI classification method based on octave convolution and multi-scale feature fusion. By introducing attention mechanism and extracting spectral-spatial fusion features, the method achieves high classification accuracy on the WHU-Hi dataset.
PRECISION ENGINEERING-JOURNAL OF THE INTERNATIONAL SOCIETIES FOR PRECISION ENGINEERING AND NANOTECHNOLOGY
(2022)
Article
Geochemistry & Geophysics
Fengfeng Wang et al.
Summary: In this article, a novel local-global feature-aware transformer based residual network (FATR) is proposed for denoising hyperspectral images (HSIs). The model utilizes spatial-spectral feature extraction, deep feature extraction with local-global feature-aware transformer blocks, and a reconstruction module to merge hierarchical features and recover clean HSIs.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Yuxi Sun et al.
Summary: A novel Multisource data reconstruction-based deep unsupervised Hashing method called MrHash is proposed in this study, aiming to construct reliable pseudolabels by exploring the characteristics of remote sensing images for enhanced image retrieval performance.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Dabing Yu et al.
Summary: This article proposes a cross-level spectral-spatial joint encoding framework for imbalanced hyperspectral image (HSI) classification. By introducing a multiscale input encoder and a spectral-spatial joint attention mechanism, it can obtain abundant features and improve the learning of discriminative spectral-spatial features. Experimental results demonstrate the superiority of this method in terms of classification accuracy and inference speed.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Lianhui Liang et al.
Summary: In this paper, a novel spectral-spatial network classification framework based on multiscale dense connected convolutional network and bidirectional recurrent neural network with attention mechanism network is proposed. By combining innovative spatial and spectral feature extraction, the proposed method achieves better performance in hyperspectral image classification.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Zhengang Zhao et al.
Summary: This letter proposes a novel method called convolutional transformer network (CTN) for HSI classification. By combining convolution and transformer structures, the method effectively utilizes spectral and spatial information to capture local-global features, resulting in improved classification performance.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Wenhui Guo et al.
Summary: In this work, a CNN-enhanced multi-level Haar wavelet features fusion network is proposed to improve the classification performance of hyperspectral images. By combining the spatial features obtained from 2D-CNN with the Haar wavelet decomposition features, spectral-spatial features are obtained and experiments show the superiority of the proposed model compared to state-of-the-art approaches.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Computer Science, Artificial Intelligence
Zhixiang Xue et al.
Summary: A novel deep hierarchical vision transformer architecture is developed for joint classification of hyperspectral and LiDAR data, utilizing long-range dependency modeling and strong generalization ability of transformer networks based on self-attention mechanism for improved collaborative classification performance of remote sensing data.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Geochemistry & Geophysics
Weilian Zhou et al.
Summary: This study proposes a novel multiscanning strategy with recurrent neural networks (RNN) for hyperspectral image (HSI) classification. By considering different scanning orders and concatenating features as input to the RNN, the network can effectively capture the spatial dependence of HSI patches and achieve competitive classification performance on three datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Jiaqi Yang et al.
Summary: This article introduces a novel Enhanced Multiscale Feature Fusion Network (EMFFN) for hyperspectral image classification. EMFFN is able to extract multiscale features from parallel multipath of three stages, achieving better classification performance.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geography, Physical
Xiaoyan Lu et al.
Summary: Road detection based on convolutional neural networks has achieved remarkable performance in very high resolution remote sensing images, but still faces challenges such as the need for massive annotated samples, limited generalization, and degraded performance due to domain gaps. To address this, a global-local adversarial learning framework is proposed for cross-domain road detection, improving feature sharing and road recognition performance through feature space adversarial learning and local alignment operations.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Remote Sensing
Gangwu Jiang et al.
Summary: A new deep learning method has been proposed for successful classification of hyperspectral images, utilizing a fully convolutional network framework with whole HSI input to achieve faster training and testing speeds. The method also incorporates channel and spatial attention models to enhance feature discriminability and improve classification accuracy over patch-based deep learning methods.
REMOTE SENSING LETTERS
(2021)
Article
Geography, Physical
Xiaoyan Lu et al.
Summary: Research analyzed the characteristics of existing deep learning road detection methods, proposed a globally aware multi-scale residual learning network that significantly improves road recognition performance. Experimental results confirmed the superiority of the proposed approach.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Proceedings Paper
Engineering, Electrical & Electronic
Saloua Chlaily et al.
Summary: Multimodal remote sensing data analysis improves the characterization of physical phenomena on Earth's surface, but is limited by nonidealities and estimation imperfections. Ensemble learning can address these issues by combining information from multiple weak classifiers to prevent overfitting and biasing. Enhanced ensemble learning scheme optimizes maximum information extraction for a given system by selecting decisions based on capacity optimization.
2021 IEEE STATISTICAL SIGNAL PROCESSING WORKSHOP (SSP)
(2021)
Article
Computer Science, Artificial Intelligence
Yonghao Xu et al.
Summary: Deep learning models have shown great capability for hyperspectral image classification but are vulnerable to adversarial attacks. The study proposes a novel SACNet to enhance robustness against adversarial examples, as demonstrated through experiments on benchmark HSI datasets.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Weilian Zhou et al.
Summary: In this paper, a new method for hyperspectral image classification using recurrent neural networks and multi-scanning direction strategy was proposed, achieving high accuracy on two benchmark datasets. The method introduces residual connections and dense connections to overcome memory problems in the RNN, demonstrating state-of-the-art results in the field.
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2021)
Article
Engineering, Electrical & Electronic
Yuxian Wang et al.
Summary: This article proposes an advanced neural network architecture combined with convolutional conditional random fields and region growing methods for hyperspectral image classification. Through the depthwise separable fully convolutional residual network, RGW method, and ConvCRF, it successfully addresses key issues such as feature extraction, detail preservation, and efficient learning with limited training samples.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Engineering, Electrical & Electronic
Shuxian Dong et al.
Summary: This article proposes a pixel cluster CNN and spectral-spatial fusion (SSF) algorithm for hyperspectral image classification with small training samples. Experimental results demonstrate that the proposed algorithm outperforms traditional CNN and other studied algorithms in cases of small training sets.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Remote Sensing
Yanfei Zhong et al.
Summary: This paper discusses the recent progress in Chinese spaceborne HRS, including typical satellite systems, data processing, and applications, as well as the future development trends of HRS in China.
GEO-SPATIAL INFORMATION SCIENCE
(2021)
Article
Engineering, Electrical & Electronic
Jianxi Wang et al.
Summary: The proposed cross-domain WOASA algorithm simultaneously considers the separability of different land-cover classes and the consistency of selected features between two scenes, leading to discriminative and domain-invariant characters of selected feature subset. The algorithm also includes an improvement for precisely controlling the dimension of selected features based on fitness function value. Experimental results demonstrate the superiority of CDWOASA in cross-scene feature selection.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Remote Sensing
Md. Palash Uddin et al.
Summary: Hyperspectral images contain valuable information for land cover classification, with techniques like PCA for feature extraction and cumulative variance accumulation for feature selection commonly employed. However, non-linear measures like the nMI-based mRMR can be more effective at selecting intrinsic features from the transformed space of FE methods. Experiment results showed that these non-linear measures outperformed existing methods on real HSI datasets.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2021)
Article
Computer Science, Hardware & Architecture
Mercedes E. Paoletti et al.
JOURNAL OF SUPERCOMPUTING
(2020)
Article
Geochemistry & Geophysics
Sen Jia et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2020)
Article
Environmental Sciences
Yanfei Zhong et al.
REMOTE SENSING OF ENVIRONMENT
(2020)
Article
Geochemistry & Geophysics
Bin Pan et al.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2020)
Review
Remote Sensing
Liangpei Zhang et al.
GEO-SPATIAL INFORMATION SCIENCE
(2020)
Article
Computer Science, Information Systems
Yonghao Xu et al.
IEEE TRANSACTIONS ON BIG DATA
(2020)
Article
Geochemistry & Geophysics
Xiaoyan Lu et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Article
Geochemistry & Geophysics
Chunju Zhang et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Article
Geochemistry & Geophysics
Bertrand Le Saux et al.
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
(2018)
Article
Remote Sensing
Saeid Amini et al.
GEO-SPATIAL INFORMATION SCIENCE
(2018)
Article
Remote Sensing
Bin Zhao et al.
GEO-SPATIAL INFORMATION SCIENCE
(2017)