Related references
Note: Only part of the references are listed.
Article
Remote Sensing
Jinzhou Liu et al.
Summary: In this paper, a Multiscale Joint Deep Neural Network (MJ-DNN) is proposed for large-scale glaciers contour extraction using single-polarimetric SAR intensity images. The method has been improved by using Atrous Separable Convolution, multiscale joint convolution layer, and deeper network with residual connection structure. The network result is optimized with the conditional random field method, showing the proposed method to be the most efficient among the tested methods.
CANADIAN JOURNAL OF REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Maniraman Periyasamy et al.
Summary: This article studies and surveys the segmentation of glaciers (calving front) using a U-Net model, optimizing the model to achieve significant improvements in automatic delineation of SAR images. The optimized U-Net model shows remarkable advancements in glacier segmentation and front delineation.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Yuting Dong et al.
Summary: This study proposes a DEM-based Automatic CF Extraction method (DACE) to efficiently extract the calving fronts (CFs) from DEM data. The algorithm achieves a high extraction accuracy for DEM data of different posting sizes, making it suitable for various applications such as glacier mass balance calculation and temporal analysis of CF changes. DACE can also be used to replace water values and edit DEMs themselves.
REMOTE SENSING OF ENVIRONMENT
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Youngwan Lee et al.
Summary: This article introduces a new Vision Transformer network architecture called Multi-Path Vision Transformer (MPViT), which achieves multi-scale feature representation through multi-scale patch embedding and multi-path structure. Experimental results demonstrate that MPViT consistently outperforms state-of-the-art Vision Transformers in multiple computer vision tasks.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2022)
Article
Geography, Physical
Xinde Chu et al.
Summary: The study proposes an improved model, Attention DeepLab V3+, and a complete solution for automated extraction of glacier boundaries. The results show that the improved model effectively enhances robustness, increases the weight of target pixels, and reduces the influence of non-target elements. Compared with other deep-learning models, the improved model performs better in the test dataset and achieves superior performance in glacier boundary extraction in certain regions.
Article
Geosciences, Multidisciplinary
Nora Gourmelon et al.
Summary: This paper presents a benchmark dataset of SAR images with manually defined labels of calving fronts from multiple regions. The dataset allows for implementation, testing, and comparison of different approaches for calving front detection, and provides two types of labels. The dataset is large enough for training deep learning models and includes glacier data from Antarctica, Greenland, and Alaska, ensuring wide applicability and reproducibility.
EARTH SYSTEM SCIENCE DATA
(2022)
Article
Geography, Physical
Thomas Frank et al.
Summary: The retreat of marine outlet glaciers has a significant impact on the mass balance of inland ice. The geometry of fjords plays a critical role in determining the retreat or stability of glaciers. By analyzing various topographic fjord settings, the study investigates the interactions between ice and topography, finding that different fjord geometries can either stabilize or promote glacier retreat. The study also establishes quantitative relationships between fjord topography and glacier retreat, providing guidance for assessing future topographically induced ice loss.
Article
Agriculture, Multidisciplinary
Omer Wosner et al.
Summary: This paper explores the proper application of modern deep networks in agricultural contexts for object detection tasks and compares their accuracy to human performance. The experiments reveal that handling small objects and large scale variance are important challenges, leading to the development of a multiple-resolution approach that significantly improves detection accuracy. Furthermore, suggested application-specific accuracy metrics are used to estimate the best performing detectors based on agricultural task requirements.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Geochemistry & Geophysics
Julia Liu et al.
Summary: The article introduces a newly developed automated method for mapping glacier calving fronts, which can be applied under various image conditions and is capable of resolving subseasonal to multiyear temporal patterns as well as regional patterns.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Computer Science, Artificial Intelligence
Mart van Rijthoven et al.
Summary: HookNet is a semantic segmentation model for histopathology whole-slide images that combines context and details via multiple branches of encoder-decoder convolutional neural networks. The advantages of using HookNet in two histopathology image segmentation tasks are demonstrated, and the model has been made publicly available by releasing the source code and web-based applications.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Environmental Sciences
Enze Zhang et al.
Summary: Over the past two decades, the data volume of remote sensing imagery in polar regions has significantly increased. Researchers have developed a generalized deep learning method that can automatically identify glaciological features like calving fronts, showing a high generalization ability on various glaciers and data types.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Computer Science, Artificial Intelligence
Andrea Galassi et al.
Summary: This article presents a unified model for attention architectures in natural language processing, with a taxonomy based on four dimensions. It demonstrates how prior information can be exploited in attention models, and discusses ongoing research efforts and challenges in this domain.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Multidisciplinary Sciences
Romain Hugonnet et al.
Summary: The study found that glaciers around the world are melting at an accelerated rate, with contrasting patterns of mass loss in different geographical regions. Overall, glaciers are losing mass more rapidly, with similar or even larger acceleration rates than Greenland or Antarctic ice sheets taken separately. These findings also highlight the importance of understanding drivers that influence glacier change to predict future changes and mitigate the impacts of sea-level rise.
Article
Geography, Physical
Melanie Marochov et al.
Summary: There is a wealth of research on the key controls on mass loss from Greenland and Antarctic ice sheets in response to climate forcing, especially in relation to marine-terminating glacier change. Deep learning has advanced automated delineation of glacier calving fronts, but research is lacking in pixel-level semantic image classification of outlet glacier environments. A novel CNN-Supervised Classification workflow achieved high accuracy in classifying marine-terminating glaciers in Greenland and demonstrated robustness in complex imagery.
Article
Geography, Physical
Celia A. Baumhoer et al.
Summary: The safety band of Antarctica, consisting of floating glacier tongues and ice shelves, plays a crucial role in supporting stability of the Antarctic Ice Sheet and influencing ice discharge. Recent disintegration events of ice shelves and glacier retreat indicate a weakening of this important safety band.
Article
Geography, Physical
Daniel Cheng et al.
Summary: The study introduced an automated method CALFIN for extracting calving fronts of Greenland glaciers from satellite images, improving data availability and accuracy, and providing new constraints for future sea level rise contributions.
Article
Multidisciplinary Sciences
Thomas Frederikse et al.
Proceedings Paper
Computer Science, Artificial Intelligence
Dichao Hu
INTELLIGENT SYSTEMS AND APPLICATIONS, VOL 2
(2020)
Article
Environmental Sciences
Celia A. Baumhoer et al.
Article
Environmental Sciences
Yara Mohajerani et al.
Article
Geography, Physical
Enze Zhang et al.
Article
Multidisciplinary Sciences
Andrew Shepherd et al.
Review
Environmental Sciences
Celia A. Baumhoer et al.
Article
Geochemistry & Geophysics
Bei Zhao et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2016)
