Related references
Note: Only part of the references are listed.
Article
Geochemistry & Geophysics
Sudipan Saha et al.
Summary: This study proposes a multi-sensor change detection method using only unlabeled target bitemporal images for self-supervised network training, employing deep clustering and contrastive learning. Evaluation on four multimodal scenes demonstrates the advantages of our self-supervised approach.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Yuxing Chen et al.
Summary: The study introduces a self-supervised framework for SAR-optical data fusion and land-cover mapping tasks, achieving comparable accuracy with image-level contrastive learning by fusing SAR and optical images through multi-view contrastive loss, and combining pretrained features and spectral information to assign land-cover classes to each pixel.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Haifeng Li et al.
Summary: This study proposes a global style and local matching contrastive learning network (GLCNet) for remote sensing image (RSI) semantic segmentation. By using global style contrastive learning and local feature matching contrastive learning modules, the method achieves superior results compared to state-of-the-art methods and supervised learning methods on various RSI semantic segmentation datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Bianka Trencanova et al.
Summary: This study proposes a method for segmenting shrub cover in high-resolution UAV images by exploring the best practices to train a convolutional neural network. Data augmentation, patch cropping, and rescaling were identified as the most effective practices. Despite being trained on a relatively small dataset, the developed classification model achieved a high average F1 score on separate test datasets.
Article
Environmental Sciences
Hanh Nguyen Pham et al.
Summary: This study proposes a novel deep learning method for monitoring estuarine ecosystems with high accuracy. By integrating spatial and context paths into a new Bilateral Segmentation Network, the processing speed and accuracy of common neural networks are significantly improved. The method utilizes multi-sensor and multi-temporal satellite images for real-time monitoring and assessment of estuarine ecological conditions.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Kaiming He et al.
Summary: This paper presents a self-supervised learning method for computer vision based on masked autoencoders. By masking a portion of the input image and reconstructing the missing pixels, large models can be trained efficiently and effectively. The approach achieves high generalization performance and outperforms supervised pretraining in transfer learning tasks.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Article
Geochemistry & Geophysics
Antonio Montanaro et al.
Summary: This letter presents a framework for self-supervised pretraining of multichannel models, specifically for fusion of multispectral and synthetic aperture radar (SAR) images, and shows its effectiveness in learning features for land cover classification.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Dilxat Muhtar et al.
Summary: Learning effective visual representations without human supervision is crucial for semantic segmentation of remote sensing images (RSIs). Current self-supervised learning (SSL) methods, trained on ImageNet, do not consider spatial position information between objects in RSIs, which is important for multiobject segmentation. In this study, we propose a novel self-supervised dense representation learning method (IndexNet) that takes into account object position changes and combines image-level and pixel-level contrast, outperforming state-of-the-art SSL methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Sikang Hou et al.
Summary: This article proposes a hyperspectral imagery classification algorithm based on contrast learning, which utilizes self-supervised learning and feature fine-tuning to improve the classification of hyperspectral images using unlabeled samples.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Gulcan Can et al.
Summary: The dataset introduced is an outdoor urban 3D pointcloud dataset covering three Swiss cities with different characteristics and a total area of 2.7km(2). This dataset, manually annotated for semantic segmentation, is suitable for various applications such as autonomous driving, gaming, and smart city planning. The quantitative results of PointNet++, a point-based deep 3D semantic segmentation model, are reported as a benchmark, with additional study on the impact of using different cities for model generalization.
PATTERN RECOGNITION LETTERS
(2021)
Article
Geosciences, Multidisciplinary
L. Bragagnolo et al.
Summary: This study evaluated the performance of U-Net, a deep convolutional neural network architecture, for mapping landslide scars from satellite imagery, demonstrating its potential and improved accuracy in identifying landslide scars. The results offer a new method for landslide scar identification in dynamic mapping systems.
Article
Engineering, Electrical & Electronic
Zixian Zheng et al.
Summary: Researchers proposed an end-to-end DCNN network named GAMNet to address issues in semantic segmentation of high-resolution remote sensing images, such as the loss of spatial information, class imbalance, and high diversity of geographic objects. The integration of attention and gate module enables multiscale feature extraction and boundary recovery, leading to state-of-the-art performance on the ISPRS 2-D semantic labeling datasets.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Inigo Alonso et al.
Summary: This work presents a novel approach for semi-supervised semantic segmentation by utilizing a contrastive learning module to enforce similar pixel-level feature representations for same-class samples. The method outperforms the current state-of-the-art for semi-supervised semantic segmentation, particularly in scenarios with limited labeled data.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Article
Computer Science, Artificial Intelligence
Jesper E. Van Engelen et al.
Article
Environmental Sciences
Xin-Yi Tong et al.
REMOTE SENSING OF ENVIRONMENT
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Carl Doersch et al.
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2015)
