Related references
Note: Only part of the references are listed.
Article
Engineering, Electrical & Electronic
Linying Zhao et al.
Summary: This study compared the performance of different deep learning structures in extracting spatio-temporal information. The results showed that 3D CNN and Vision Transformer achieved the best performance, followed by 2D CNN and conventional methods. It was also found that using optical images alone was sufficient for land-cover classification, but SAR images could provide satisfactory results when optical images were unavailable.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Geography, Physical
Xin-Yi Tong et al.
Summary: High-resolution satellite images are valuable for land cover classification, but their application in detailed mapping at large scale is limited. To address this, we present a large-scale land cover dataset called Five-Billion-Pixels, with over 5 billion labeled pixels from 150 high-resolution Gaofen-2 satellite images. We also propose a deep-learning-based unsupervised domain adaptation approach for large-scale land cover mapping. Experimental results show promising performance even with entirely unlabeled images.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Yanqiao Chen et al.
Summary: Remote sensing image scene classification has become popular, but manually labeling large amounts of remote sensing images is difficult and time-consuming. Therefore, few-shot scene classification of remote sensing images is an urgent research task. This paper proposes a deep nearest neighbor neural network based on attention mechanism (DN4AM) to solve this task. By using scene class-related attention maps to reduce interference from irrelevant objects, our method achieves promising results in few-shot scene classification of remote sensing images, outperforming several state-of-the-art methods.
Article
Geography, Physical
Jiaqi Yang et al.
Summary: Due to the rich spectral and spatial information, hyperspectral image (HSI) can be used for accurately classifying land covers. Deep learning techniques such as CNN, FCN, and RNN have been widely applied in HSI classification. However, current methods still face issues with geometric constraints, contribution fuzziness of central pixels, and interaction gap between local and further areas.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Review
Computer Science, Artificial Intelligence
Mohammed Abdulmajeed Moharram et al.
Summary: Recently, there has been increasing focus on land use land cover (LULC) classification due to various challenges like urbanization, environmental pollution, drought, floods, and climate change. Hyperspectral imaging has gained attention because of its informative features, such as spectral-spatial features. This paper provides a comprehensive review of LULC classification using hyperspectral images, covering four significant research investigations.
Article
Environmental Sciences
Huiqing Pei et al.
Summary: In this study, a novel MSG-GCN model was compared with other state-of-the-art methods for the accurate classification of forest types using high-resolution aerial photographs. The MSG-GCN model outperformed other models in terms of classification accuracy and had clear boundaries between different forest types.
Article
Environmental Sciences
Bing Li et al.
Summary: The application of Transformer in computer vision has had a significant impact in the field of deep learning. While convolutional neural networks (CNN) have shown exceptional performance in hyperspectral image (HSI) classification, Transformer has not produced satisfactory results in that area. Recently, a Sequencer structure that replaces the Transformer self-attention layer with a BiLSTM2D layer has achieved satisfactory results in image classification. This paper proposes a unique network called SquconvNet, which combines CNN with Sequencer block to improve hyperspectral classification. Rigorous experiments on three relevant baseline datasets demonstrate that our proposed method outperforms in terms of classification accuracy and stability.
Article
Environmental Sciences
Xiaotong Ma et al.
Summary: A 3D-1D-CNN model was proposed for processing cloud-shadow affected hyperspectral images in complex urban areas. Spectral composition parameters, vegetation index, and texture characteristics were extracted and segmented into small patches for input to a 3D-CNN classifier. The overall accuracy of the proposed 3D-1D-CNN was 96.32%, significantly higher than SVM, RF, 1D-CNN, or 3D-CNN.
Article
Environmental Sciences
Chengjun Xu et al.
Summary: A novel supervised adversarial Lie Group feature learning network is proposed to effectively generate data samples with category information even with limited data samples. This model takes category information and data samples as input and optimizes the constraint of category information in the loss function, producing data samples containing category information. Additionally, an object scale sample generation strategy is introduced to generate data samples of different scales, ensuring the generated samples contain richer feature information.
Article
Plant Sciences
Hengbin Wang et al.
Summary: In this study, a new deep learning approach called Cropformer was proposed for multi-scenario crop classification. Cropformer solves the problem of current crop classification methods extracting only a single feature by extracting both global and local features. Experimental results showed that Cropformer achieved significant accuracy advantage in crop classification and achieved higher accuracy with fewer samples.
FRONTIERS IN PLANT SCIENCE
(2023)
Article
Agriculture, Multidisciplinary
Changhong Xu et al.
Summary: This study proposes a multi-layer pyramid crop classification network (MPNet) to solve the challenges faced in crop classification tasks based on neural networks. By using a pyramid pooling module to improve global information acquisition and an information concatenation module to retain upper features, feature loss during crop extraction is reduced. Experimental results show that the proposed model achieves the highest accuracy compared to five other deep learning models, and it also has a shorter training time and higher efficiency under the same running conditions. Overall, this study improves the efficiency and accuracy of crop classification tasks in unbalanced temporal and spatial distribution, providing a feasible solution for crop classification tasks in complex growing areas.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Forestry
Xiangsuo Fan et al.
Summary: This paper proposes a hierarchical convolutional recurrent neural network (HCRNN) model that combines CNN and RNN modules for pixel-level classification of multispectral remote sensing images. The experimental results show that the HCRNN model achieves an overall accuracy of 97.62% on the Sentinel-2 dataset, improving the performance by 1.78% compared to the RNN model. Furthermore, the study focuses on the changes in forest cover in the study area of Laibin City, Guangxi Zhuang Autonomous Region.
Article
Forestry
Yurong Li et al.
Summary: A measurement method for P. massoniana seedling morphological indicators based on machine vision was proposed in this paper. The method includes image processing, point cloud segmentation, and skeleton extraction. Experimental results showed that the method can accurately extract morphological indicators.
Article
Environmental Sciences
Sultan Daud Khan et al.
Article
Agronomy
Ke Wu et al.
Summary: This study provides high-precision detection models and relevant spectral information for the detection of different states of kiwifruit affected by environmental stress. Through laboratory experiments, spectral reflectance data of kiwifruit leaves in four different states were collected and analyzed. Multiple modeling approaches were used to extract bands and evaluate the performance of the models. The study successfully achieved the detection of kiwifruit under different stages of environmental stress.
Review
Computer Science, Information Systems
Garima Jaiswal et al.
Summary: The combination of hyperspectral imaging (HSI) and autoencoders (AE) has the potential to revolutionize various industries, including healthcare and environmental monitoring. AE can uncover hidden patterns and insights in large-scale datasets, while HSI can benefit from the scalability and flexibility offered by AE, leading to more efficient data processing.
COMPUTER SCIENCE REVIEW
(2023)
Article
Engineering, Electrical & Electronic
Caihao Sun et al.
Summary: In this article, a semisupervised dual-branch spectral-spatial adversarial representation learning (SSARL) method is proposed for HSI classification. SSARL adaptively assigns attention weights to different pixels and adds a spectral constraint to spatial features. Experimental results demonstrate that SSARL outperforms competitive methods on small-sized labeled samples and exhibits superior performance for boundary test pixels.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Engineering, Electrical & Electronic
Alvaro G. Dieste et al.
Summary: This study presents ResBaGAN, a GAN-based method for remote sensing image classification, which overcomes the challenges of limited labeled data and class imbalances through an advanced data augmentation framework. Compared to other machine learning methods, ResBaGAN achieves higher overall classification accuracies, particularly improving the accuracy of minority classes with F1-score enhancements up to 22%.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Yonghao Yi et al.
Summary: In this article, we propose an essential feature mining network (EFM-net) based on deep convolutional neural network (DCNN) to obtain discriminative features for fine-grained classification in remote sensing. The proposed EFM-net includes two modules, Miner and Refiner, which work together to extract essential features of the targets. Experimental results show the superiority of the proposed method over existing alternatives on public datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Anastasios Temenos et al.
Summary: An interpretable deep learning framework for land use and land cover (LULC) classification in remote sensing is proposed, combining a compact CNN model for image classification with a SHAP deep explainer to improve classification results. The framework is applied to Sentinel-2 satellite images with a reduced input data size of 77% using three-band combinations. Experimental results on the EuroSAT dataset show accurate classification with an overall accuracy of 94.72% and improved accuracy compared to standard approaches. The SHAP explainable results enhance classifications in urban and rural areas with different land uses.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Engineering, Multidisciplinary
Weizhuang Gong et al.
Summary: This study proposes a capacitive flexible tactile sensor that utilizes a dielectric layer with a microcylindrical structure. A decoupling model is presented based on the analysis of sensor coupling. The sensor has been validated in applications such as tangential detection and object grasping, laying the foundation for future self-aware operations.
Article
Geochemistry & Geophysics
Wang Miao et al.
Summary: The paper proposes a semi-supervised representation consistency Siamese network (SS-RCSN) for remote-sensing image scene classification, utilizing GAN to extract discriminative features and reducing the differences between labeled and unlabeled data, experimental results demonstrate superior performance compared to other semi-supervised learning methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Ailong Ma et al.
Summary: In this article, a supervised progressive growing generative adversarial network (SPG-GAN) is proposed for remote sensing image scene classification, which can generate labeled samples and significantly improve the classification accuracy in the case of limited samples.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Chao Xie et al.
Summary: A novel coordinate attention mechanism is explored in this study to improve the performance of image super-resolution and is successfully applied in a deep residual coordinate attention SR network.
IET IMAGE PROCESSING
(2022)
Article
Environmental Sciences
Nan Jiang et al.
Summary: Few-shot remote sensing image scene classification is a challenging problem. This paper proposes a multi-scale graph-based feature fusion model, which effectively represents spatial relations and integrates different scale information, achieving accurate classification with limited labeled data.
Article
Environmental Sciences
Jiahang Liu et al.
Summary: Water body extraction from remote sensing images using deep learning faces challenges in detecting both large and small water bodies, accurately predicting the edge position, and lack of labeled samples. This paper proposes a novel SFnet-DA network based on domain adaptation and multi-scale feature fusion to address these issues, which shows better performance in water body segmentation compared to current methods.
Article
Computer Science, Hardware & Architecture
Muhammad Sohail et al.
Summary: This paper presents a multiscale spectral-spatial feature learning network (MulNet) for hyperspectral image (HSI) classification. By integrating a 3D Residual Network (3DResNet), a Feature Fusion Module (FFM), and a Recurrent Neural Network (RNN), the network effectively handles the spectral heterogeneity of HSIs and improves classification accuracy. Experimental results on real-world datasets demonstrate the efficacy and efficiency of the proposed network.
Article
Computer Science, Hardware & Architecture
S. Ansith et al.
Summary: The development of new deep learning algorithms has significantly changed land use classification. Recent models combine deep neural network structures with machine learning algorithms for feature extraction and classification. The proposed model based on the modified GAN architecture can achieve better results with fewer training samples, making it superior to other deep learning models.
Article
Geography, Physical
Yunwei Tang et al.
Summary: Recent developments in deep learning have introduced new methods for land cover classification. However, most methods neglect the spatial association of land cover classes in remote sensing images. This research proposes a deep relearning method based on recurrent neural networks, which improves classification accuracy by considering the spatial arrangement of land cover classes.
GISCIENCE & REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Damian Ibanez et al.
Summary: This article presents a novel masked auto-encoding spectral-spatial transformer (MAEST) model, which combines two collaborative branches to classify and reconstruct hyperspectral remote sensing images, addressing the noise issue in conventional transformer networks.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Lubin Bai et al.
Summary: This study proposes a model that integrates contrastive learning and adversarial learning to align two remote sensing datasets with different acquisition conditions in both representation space and spatial layout. It achieves adaptive alignment at the pixel level and the predicted results level using a semantic segmentation network and two domain adaptation branches. Additionally, a training strategy called category similarity matching sampling is proposed to improve the model's performance by providing similar category compositions in image pairs.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Pedro J. Soto et al.
Summary: Many deep learning-based domain adaptation methods for remote sensing applications rely on adversarial training strategies to align features extracted from images of different domains. This study proposes a DL-based representation matching approach for domain adaptation in change detection tasks, which effectively mitigates the issue of imbalanced classes and improves the accuracy of cross-domain deforestation detection.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Yansheng Li et al.
Summary: Remote sensing image scene classification faces challenges such as annotation difficulty and zero-shot classification in the era of RS big data. This article proposes a novel ZSRSSC method based on locality-preservation deep cross-modal embedding networks, which effectively solves the problem of class structure inconsistency and significantly outperforms existing methods in performance.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Computer Science, Artificial Intelligence
Guangyuan Liu et al.
Summary: This study introduces a new approach using a multiobjective evolutionary algorithm assisted stacked autoencoder for PolSAR image classification, which can adaptively optimize parameters and hyperparameters to achieve competitive results.
SWARM AND EVOLUTIONARY COMPUTATION
(2021)
Article
Geochemistry & Geophysics
Chao Tao et al.
Summary: This study introduces a model considering spatial information, which combines CNN and recurrent neural networks to learn more discriminative features, thus improving the accuracy of remote sensing image scene classification. Experimental results demonstrate that the proposed method outperforms other three state-of-the-art methods on the existing dataset.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Engineering, Electrical & Electronic
Zhuang Zhou et al.
Summary: This article introduces a large-scale dataset NaSC-TG2 built from Tiangong-2 remotely sensed imagery, which aims to expand and enrich annotated data for advancing remote sensing classification algorithms, especially for natural scene classification. The dataset contains 20,000 images equally divided into ten scene classes, with the advantages of large scale, intra-class differences, and inter-class similarity.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Computer Science, Information Systems
Haikel Alhichri et al.
Summary: The paper introduces a deep attention Convolutional Neural Network (CNN) for scene classification in remote sensing, which enhances the focus on important regions through the attention mechanism. The proposed method, EfficientNet-B3-Attn-2, shows strong capabilities in classifying RS scenes across multiple remote sensing datasets.
Article
Environmental Sciences
Xin-Yi Tong et al.
REMOTE SENSING OF ENVIRONMENT
(2020)
Article
Engineering, Electrical & Electronic
Zhenfeng Shao et al.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2020)
Article
Engineering, Electrical & Electronic
Patrick Helber et al.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2019)
Article
Geochemistry & Geophysics
Renlong Hang et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Article
Environmental Sciences
Andong Ma et al.
Proceedings Paper
Geosciences, Multidisciplinary
Gencer Sumbul et al.
2019 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2019)
(2019)
Review
Agriculture, Multidisciplinary
Andreas Kamilaris et al.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2018)
Article
Engineering, Electrical & Electronic
Xiangrong Zhang et al.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2018)
Article
Geochemistry & Geophysics
Juan Mario Haut et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2018)
Article
Geography, Physical
Weixun Zhou et al.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2018)
Article
Environmental Sciences
Zhenfeng Shao et al.
Article
Geochemistry & Geophysics
Grant J. Scott et al.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2017)
Article
Geochemistry & Geophysics
Yanqiao Chen et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2017)
Article
Geochemistry & Geophysics
Aoxue Li et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2017)
Article
Geochemistry & Geophysics
Lichao Mou et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2017)
Article
Geochemistry & Geophysics
Gui-Song Xia et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2017)
Article
Engineering, Electrical & Electronic
Gong Cheng et al.
PROCEEDINGS OF THE IEEE
(2017)
Article
Geochemistry & Geophysics
Xiao Xiang Zhu et al.
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
(2017)
Article
Computer Science, Hardware & Architecture
Alex Krizhevsky et al.
COMMUNICATIONS OF THE ACM
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Gao Huang et al.
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017)
(2017)
Article
Geochemistry & Geophysics
Bei Zhao et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2016)
Article
Geochemistry & Geophysics
Liangpei Zhang et al.
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
(2016)
Review
Multidisciplinary Sciences
Yann LeCun et al.
Article
Environmental Sciences
Quanlong Feng et al.
Article
Environmental Sciences
Quanlong Feng et al.
Article
Geochemistry & Geophysics
Xin Huang et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2014)
Article
Remote Sensing
Elhadi Adam et al.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2014)
Article
Environmental Sciences
M. Joseph Hughes et al.
Article
Computer Science, Artificial Intelligence
O Chapelle et al.