Article
Remote Sensing
Yuan Yuan, Lei Lin, Qingshan Liu, Renlong Hang, Zeng-Guang Zhou
Summary: This article introduces a pre-trained representation model called SITS-Former for Sentinel-2 time series classification. The model is pre-trained using self-supervised learning on a large amount of unlabeled data and then fine-tuned for a target classification task. Experimental results on two crop classification tasks show that SITS-Former outperforms state-of-the-art approaches and greatly reduces the burden of manual labeling.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Geography, Physical
Wenqiang Xi, Shihong Du, Shouhang Du, Xiuyuan Zhang, Haiyan Gu
Summary: This study proposed a novel approach to accurately classify dense satellite image time series for mapping intra-annual land cover dynamics. The method involves segmentation of dense SITS to generate optimal spatiotemporal cubes and utilizes spectral, textural, spatial, and temporal features for classification. By modeling spatiotemporal context with a conditional random field model, the approach demonstrates significant improvements in classification accuracy over existing methods.
GISCIENCE & REMOTE SENSING
(2021)
Article
Computer Science, Information Systems
Muhammad Aman, Said Jadid Abdulkadir, Izzatdin Abdul Aziz, Hitham Alhussian, Israr Ullah
Summary: This paper proposes a semantic-based unsupervised approach (KP-Rank) for keyphrase extraction, utilizing Latent Semantic Analysis (LSA) and clustering techniques, and introducing a novel frequency-based algorithm considering locality-based sentence, paragraph, and section frequencies. Experimental results demonstrate that KP-Rank achieved significant improvements on benchmark datasets from different domains.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Chandrabali Karmakar, Corneliu Octavian Dumitru, Nick Hughes, Mihai Datcu
Summary: This article presents a novel framework to model and understand image dynamics in Openly available satellite image time series (SITS). The framework utilizes visualizations and domain knowledge to efficiently integrate machine learning pipelines in the absence of ground truth data. The framework is validated through a case study in a Polar region, where limited ground truth data is extended to discover temporal evolution at the image patch level.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Computer Science, Information Systems
Peng Tao, Xiaohu Hao, Jie Cheng, Luonan Chen
Summary: In this study, a spatiotemporal information scheme is utilized to transform high-dimensional/spatial information into temporal information, and a new method called multitask Gaussian process regression machine (MT-GPRM) is developed for accurate predictions from short-term time series.
INFORMATION SCIENCES
(2023)
Article
Environmental Sciences
Ruifu Wang, Dongdong Teng, Wenqing Yu, Xi Zhang, Jinshan Zhu
Summary: This research proposes a generative adversarial network (GAN) model for time series satellite cloud image prediction. The model learns the data feature distribution of satellite cloud images and predicts future time series cloud images by considering the time series information. Through the integration of the Mish activation function and implementation of improvement measures such as using the Wasserstein distance, establishing a multiscale network structure, and combining image gradient difference loss, the model achieves better predictive performance. The experimental results demonstrate that the improved GDL-GAN model maintains good visualization effects while accurately capturing the overall changes and movement trends of the predicted cloud images, thereby enhancing the cooperation ability of satellite cloud images in disastrous weather forecasting and early warning.
Article
Environmental Sciences
Zheng Zhang, Ping Tang, Weixiong Zhang, Liang Tang
Summary: This paper introduces a new time series similarity measure method TAOT for SITS clustering, which effectively alleviates the issues of DTW and improves clustering accuracy according to statistical and visual results on real datasets. TAOT serves as a useful tool to explore the potential of valuable SITS data.
Article
Geochemistry & Geophysics
Xiangtao Zheng, Xiumei Chen, Xiaoqiang Lu, Bangyong Sun
Summary: This article proposes a cross-resolution difference learning method to detect changes from multitemporal images in their original different resolutions without resizing operations. The method involves segmenting input images into homogeneous regions, generating pixelwise difference maps based on mutual information distance and deep feature distance, and then fusing and binarizing the cross-resolution difference maps to produce the final binary change map. Extensive experiments on four datasets demonstrate the effectiveness of the proposed method for detecting changes from different resolution images.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Naik Nitesh Navnath, Kandasamy Chandrasekaran, Andrzej Stateczny, Venkatesan Meenakshi Sundaram, Prabhavathy Panneer
Summary: This study evaluates various deep learning models with SITS data for land cover classification, finding that combining recurrent neural networks with pixel coordinates enhances accuracy and F1 scores. Partioned execution of LCC tasks also outperformed temporal models.
Article
Geography, Physical
Yuan Yuan, Lei Lin, Zeng-Guang Zhou, Houjun Jiang, Qingshan Liu
Summary: Precise crop mapping is crucial for agricultural production and food security. This article proposes a novel method for crop classification, which integrates optical and SAR features through cross-modal contrastive learning. Experimental results demonstrate that our method consistently outperforms traditional supervised learning approaches, regardless of the adequacy of training samples.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Physics, Multidisciplinary
Yujie You, Le Zhang, Peng Tao, Suran Liu, Luonan Chen
Summary: This study proposes a novel spatiotemporal transformer neural network (STNN) for efficient prediction of high-dimensional short-term time-series, which utilizes a continuous attention mechanism and various attention mechanisms to integrate and predict effective information. Experimental results demonstrate that STNN significantly outperforms existing methods in multi-step forecasting.
Article
Engineering, Electrical & Electronic
Xiaowei Xiang, Yang Liu, Gaoyun Fang, Jing Liu, Mengyang Zhao
Summary: Unsupervised Domain Adaptation (UDA) aims to free models from labeled information in the target domain and minimize the distribution discrepancies between different domains. Most existing methods focus only on domain-invariant feature learning through either domain discrimination or matching lower-order moments, resulting in limited robustness for non-Gaussian distributions and inadequate domain matching. To address these issues, we propose a novel Two-Stage Alignments Framework (TSAF) for UDA, which characterizes non-Gaussian distributions through arbitrary-order moment matching and aligns probabilistic outputs of classifiers using domain-specific decision boundaries. Additionally, a reconstruction-based task is introduced to enhance the representation of specific distribution characteristics. Experimental results on three real-world time series datasets demonstrate the superiority of our model in cross-domain classification tasks and the efficient learning of domain-invariant features by TSAF.
IEEE SIGNAL PROCESSING LETTERS
(2023)
Article
Computer Science, Information Systems
Kiyotaka Matsue, Mahito Sugiyama
Summary: This study proposes a tensor-based feature extraction method (UFEKT) for clustering and outlier detection of multivariate time series. The method constructs feature vectors for subsequences by considering both time and variable associations, and can be used as a preprocessing technique for clustering and outlier detection algorithms.
Article
Chemistry, Multidisciplinary
Margherita Berardi, Luigi Santamaria Amato, Francesca Cigna, Deodato Tapete, Mario Siciliani de Cumis
Summary: Volcanic monitoring reports contain valuable geochemical and geophysical data. This study presents a natural language processing system that can extract relevant gas parameters from such reports, as demonstrated by its successful application to monitoring bulletins from Stromboli volcano published between 2015 and 2021.
APPLIED SCIENCES-BASEL
(2022)
Article
Geochemistry & Geophysics
Sudipan Saha, Francesca Bovolo, Lorenzo Bruzzone
Summary: This article proposes an unsupervised deep learning-based method to detect changes in image time series. The method does not require prior knowledge of the change date, and treats change detection as an anomaly detection problem. By using a multilayer LSTM network to learn the representation of the time series, and rearranging the input sequence using an encoder-decoder LSTM model, the method can effectively identify changed pixels.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Simon Daout, Benedetta Dini, Wilfried Haeberli, Marie-Pierre Doin, Barry Parsons
EARTH AND PLANETARY SCIENCE LETTERS
(2020)
Article
Energy & Fuels
M. Prussi, A. Julea, L. Lonza, C. Thiel
Summary: Upgrading biogas to biomethane is a feasible option for greening the European energy sector, particularly in transport. However, challenges such as infrastructure fragmentation and high costs may hinder the widespread diffusion of biomethane as an alternative fuel. Efforts are needed to carefully assess the impact of biomethane on greenhouse gas emissions and to ensure a synchronized deployment of vehicles, infrastructure, and production facilities to prevent barriers to its development.
ENERGY STRATEGY REVIEWS
(2021)
Article
Geochemistry & Geophysics
Tao Li, Jianbao Sun, Yuxin Bao, Yan Zhan, Zheng-Kang Shen, Xiwei Xu, Cecile Lasserre
Summary: This study reports a damaging Mw 5.8 earthquake in Changning, China in June 2019, which is likely the largest induced earthquake by industrial exploitation ever recorded. The research suggests that water injections may have triggered this event. The importance of reassessing seismic hazard over similar tectonic environments with intensive industrial exploitation is highlighted.
Editorial Material
Geosciences, Multidisciplinary
Yosuke Aoki, Masato Furuya, Francesco De Zan, Marie-Pierre Doin, Michael Eineder, Masato Ohki, Tim J. Wright
EARTH PLANETS AND SPACE
(2021)
Article
Geochemistry & Geophysics
C. Pagani, T. Bodin, M. Metois, C. Lasserre
Summary: A transdimensional Bayesian method is proposed to estimate surface strain rates and compared to a standard interpolation scheme. The method shows higher resilience to data errors and uneven data distribution, while providing uncertainties associated with recovered velocities and strain rates.
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
(2021)
Article
Geosciences, Multidisciplinary
P. Pitard, A. Replumaz, M. L. Chevalier, P. H. Leloup, M. Bai, M. P. Doin, C. Thieulot, X. Ou, M. Balvay, H. Li
Summary: Through thermochronology data and thermo-kinematic modeling, the geological features and movement velocities of thrust faults in the crustal thickening history have been determined. The study shows that the structure in the Muli thrust belt has significant changes in thrusting geometry from the surface to depth, and the exhumation of crustal rocks is influenced by deeper crustal processes.
GEOPHYSICAL RESEARCH LETTERS
(2021)
Article
Environmental Sciences
Franck Thollard, Dominique Clesse, Marie-Pierre Doin, Joelle Donadieu, Philippe Durand, Raphael Grandin, Cecile Lasserre, Christophe Laurent, Emilie Deschamps-Ostanciaux, Erwan Pathier, Elisabeth Pointal, Catherine Proy, Bernard Specht
Summary: The FLATSIM service aims to process Sentinel-1 data over large areas using multi-temporal InSAR techniques, providing the ForM@ter scientific community with automatically processed products and quality indicators for research in seismology, tectonics, volcano-tectonics, and hydrological cycle.
Article
Geochemistry & Geophysics
L. Marconato, P. H. Leloup, C. Lasserre, R. Jolivet, S. Caritg, R. Grandin, M. Metois, O. Cavalie, L. Audin
Summary: This study investigates the 2019 Le Teil earthquake in France using various methods and analyzes the geological model and post-seismic displacement. The study found that the La Rouviere Fault near the epicenter of the earthquake was reactivated, and the distribution and depth of fault slip were determined. This study is of great significance for reevaluating the seismic hazard of many faults, including the La Rouviere Fault system.
GEOPHYSICAL JOURNAL INTERNATIONAL
(2022)
Article
Geosciences, Multidisciplinary
M. Mathey, M-P Doin, P. Andre, A. Walpersdorf, S. Baize, C. Sue
Summary: Based on leveling and GNSS data analysis in the Western European Alps, we propose a new approach to map the uplift pattern using InSAR technology, overcoming the challenges of noise and low signal in mountainous areas. The results are consistent with other geodetic measurements and reveal small-scale spatial variations in the uplift pattern.
GEOPHYSICAL RESEARCH LETTERS
(2022)
Article
Engineering, Geological
Aya Cheaib, Pascal Lacroix, Swann Zerathe, Denis Jongmans, Najmeh Ajorlou, Marie-Pierre Doin, James Hollingsworth, Chadi Abdallah
Summary: This study investigates landslides triggered by earthquakes in a semi-arid region, finding that there are more limited-sized rockfalls near the epicenter and larger deep-seated landslides farther away, which is explained as an interaction between earthquake source properties and local geological conditions. The study also examines the kinematics of earthquakes-accelerated slow-moving ancient landslides.
Article
Environmental Sciences
Floriane Provost, David Michea, Jean-Philippe Malet, Enguerran Boissier, Elisabeth Pointal, Andre Stumpf, Fabrizio Pacini, Marie-Pierre Doin, Pascal Lacroix, Catherine Proy, Philippe Bally
Summary: This article introduces a toolbox called MPIC-OPT for processing optical images, which is aimed at measuring terrain deformation over time. The toolbox provides an end-to-end solution and includes options such as correction and filtering to enhance the accuracy and precision of the measurements. The performance of MPIC-OPT is tested on various use cases and is shown to produce results consistent with in-situ data. The study also highlights the importance of correlation threshold and temporal matching range parameters in the estimation of terrain deformation.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Geosciences, Multidisciplinary
M. Henriquet, B. Kordic, M. Metois, C. Lasserre, S. Baize, L. Benedetti, M. Spelic, M. Vukovski
Summary: This study utilizes the Petrinja earthquake in Croatia as an example to reveal the slip pattern and deformation of the earthquake by re-measuring civilian benchmark networks. The study demonstrates that rapid measurement of existing networks can provide valuable seismic constraints.
GEOPHYSICAL RESEARCH LETTERS
(2022)
Article
Geosciences, Multidisciplinary
Cecile Cornou, Jean-Paul Ampuero, Coralie Aubert, Laurence Audin, Stephane Baize, Jeremy Billant, Florent Brenguier, Mathieu Causse, Mohamed Chlieh, Andy Combey, Marcello de Michele, Bertrand Delouis, Anne Deschamps, Matthieu Ferry, Michael Foumelis, Berenice Froment, Celine Gelis, Raphael Grandin, Jean-robert Grasso, Estelle Hannouz, Sebastien Hok, Axel Jung, Romain Jolivet, Mickael Langlais, Philippe Langlaude, Christophe Larroque, Philippe Herve Leloup, Kevin Manchuel, Leo Marconato, Christophe Maron, Emmanuel Mathot, Emeline Maufroy, Diego Mercerat, Marianne Metois, Emmanuelle Neyman, Ildut Pondaven, Ludmila Provost, Julie Regnier, Jean-Francois Ritz, Diane Rivet, Antoine Schlupp, Anthony Sladen, Christophe Voisin, Andrea Walpersdorf, David Wolynieck, Pascal Allemand, Elise Beck, Etienne Bertrand, Veronique Bertrand, Pierre Briole, Didier Brunel, Olivier Cavalie, Jerome Cheze, Francoise Courboulex, Isabelle Douste-Bacque, Remi Dretzen, Tiziano Giampietro, Maxime Godano, Philippe Grandjean, Marc Grunberg, Gauthier Guerin, Stephane Guillot, Elias el Haber, Alain Hernandez, Herve Jomard, Cecile Lasserre, Chao Liang, Itzhak Lior, Xavier Martin, Daniel Mata, Marine Menager, Antoine Mercier, Aurelien Mordret, Elif Oral, Anne Paul, Fabrice Peix, Catherine Pequegnat, Michel Pernoud, Claudio Satriano, Rihab Sassi, Marc Schaming, Valerie Sellier, Christophe Sira, Anne Socquet, Christian Sue, Aurelie Trilla, Martin Vallee, Martijn van den Ende, Philippe Vernant, Benjamin Vial, Huihui Weng
Summary: In November 11, 2019, a M-w 4.9 earthquake occurred near Montelimar, France, with a very shallow focal depth and causing moderate to large damages in several villages. The lack of monitoring stations prompted a swift response from the French scientific community to deploy instruments, conduct field surveys, and study the earthquake's intensity. This comprehensive dataset aims to unravel the earthquake's causes, rupture mechanism, and contribute to seismic hazard assessment in the region.
COMPTES RENDUS GEOSCIENCE
(2021)
Article
Geochemistry & Geophysics
Lea Pousse-Beltran, Anne Socquet, Lucilla Benedetti, Marie-Pierre Doin, Magali Rizza, Nicola D'Agostino
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
(2020)