Article
Geochemistry & Geophysics
Wei Lu, Si-Bao Chen, Jin Tang, Chris H. Q. Ding, Bin Luo
Summary: Remote-sensing (RS) images present unique challenges for computer vision (CV) due to lower resolution, smaller objects, and fewer features. To address this problem, we propose a new and universal downsampling module named robust feature downsampling (RFD), which creates a more accurate and robust analysis of RS images by fusing multiple feature maps extracted by different downsampling techniques. We develop two versions of the RFD module (SRFD and DRFD) and conduct comparative experiments, showing significant improvements in RS image classification, object detection, and semantic segmentation.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Chemistry, Multidisciplinary
Josept David Revuelta-Acosta, Edna Suhail Guerrero-Luis, Jose Eduardo Terrazas-Rodriguez, Cristian Gomez-Rodriguez, Gerardo Alcala Perea
Summary: This study provides the most recent analysis of land use and land cover change in Coatzacoalcos, Mexico over the past six years. The study utilized remote sensing technology and ground-based surveys to analyze the changes. The findings indicate a slowdown in growth for residential, industry, and commercial areas, as well as significant degradation of swamps. Dunes and areas with high vegetation density transitioned to low vegetation density areas.
APPLIED SCIENCES-BASEL
(2022)
Article
Environmental Sciences
F. Milazzo, P. Fernandez, A. Pena, T. Vanwalleghem
Summary: Land use change is one of the main drivers of soil erosion in the Mediterranean. This study examines the relationship between land use and erosion in Southern Spain from 1956 to 2018 and highlights the mitigation role of permanent grassland. The research findings show that maintaining permanent grassland can help reduce erosion, which should be considered in future agricultural policy.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Computer Science, Interdisciplinary Applications
Wenliang Li
Summary: Detailed urban land-use patterns are crucial for urban management, economic analysis, and policy-making towards sustainable urban development. This study proposed a framework combining social sensing data and remote sensing images to map the land-use patterns of New York City, achieving high classification accuracy that can be applied in urban planning and building energy use modeling.
EARTH SCIENCE INFORMATICS
(2021)
Article
Geochemistry & Geophysics
Ba Hung Ngo, Ju Hyun Kim, So Jeong Park, Sung In Cho
Summary: This article introduces a unified framework called MECKA, which achieves multi-source unsupervised domain adaptation for remote sensing data through the processes of multi-view generation and collaborative learning. Experimental results show that the proposed method achieves the best classification accuracy on RS scene benchmark datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Hardware & Architecture
S. Ansith, A. A. Bini
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
Engineering, Electrical & Electronic
Jiangtao Peng, Yi Huang, Weiwei Sun, Na Chen, Yujie Ning, Qian Du
Summary: Traditional remote sensing image classification methods heavily rely on labeled samples, which may fail when labeled samples are unavailable or have different distributions. Cross-domain or cross-scene remote sensing image classification is developed to address these issues. The distribution inconsistency problem can be caused by differences in acquisition environment conditions, scenes, time, and sensors. To tackle the cross-domain remote sensing image classification problem, various domain adaptation techniques have been developed.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Geography, Physical
Burak Ekim, Elif Sertel
Summary: Utilizing three different Deep Neural Network Ensemble methods can improve performance in remote sensing image classification tasks and increase accuracy. This approach enhances the generalizability of the models, generates more robust and generalizable outcomes, and promotes the widespread use of the method.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2021)
Article
Geochemistry & Geophysics
Jie Geng, Bohan Xue, Wen Jiang
Summary: Few-shot learning is addressed for remote sensing image scene classification by proposing a foreground-background contrastive learning method. It includes a foreground-background separation module for distinguishing object and background features and a channel weight allocator for balancing feature dimensions. Experimental results on three remote sensing datasets demonstrate its superior classification performance compared to other approaches.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Remote Sensing
Jiangfan Feng, Wei Zheng, Zhujun Gu, Dongen Guo, Rui Qin
Summary: This article introduces a specific attention-based network called PaANet for semantic segmentation of remote sensing images. By incorporating position-aware attention and pyramid pooling expectation-maximization modules, this method significantly improves recognition accuracy and the continuity of ground object recognition while preserving structural classification details. The research also proposes a multiresolution data augmentation method that further enhances the model's performance and generalization ability.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2023)
Article
Environmental Sciences
Luiz Fernando Favacho Morais Filho, Kamila Cunha de Meneses, Gustavo Andre de Araujo Santos, Elton da Silva Bicalho, Glauco de Souza Rolim, Newton La Scala Jr
Summary: Agriculture and soil management practices have a direct impact on CO2 emissions in crop fields. This study investigated the relationship between NDVI, SIF, and xCO2 in different agroecosystems in southern-central Brazil, finding a negative correlation between SIF and xCO2. The results suggest the potential use of SIF and xCO2 to identify sources and sinks of greenhouse gases in agricultural areas.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2021)
Article
Geochemistry & Geophysics
Jianqi Chen, Keyan Chen, Hao Chen, Wenyuan Li, Zhengxia Zou, Zhenwei Shi
Summary: The article introduces an asynchronous contrastive learning-based method for effective fine-grained ship classification in remote sensing images. The method, called Push-and-Pull Network (P(2)Net), separates images using a dual-branch network and aggregates them into subclasses using an integration module. Experimental results demonstrate the effectiveness of the proposed method.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Claudia Paris, Lorenzo Bruzzone
Summary: The article introduces a novel approach to extract reliable labeled data from existing thematic products, improving the effectiveness of supervised classification algorithms and addressing the issue of insufficient traditional data collection.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Yan-e Hou, Kang Yang, Lanxue Dang, Yang Liu
Summary: Convolutional neural networks have achieved remarkable results in remote sensing scene classification. However, existing methods often ignore object-level information in shallow features. This paper proposes an end-to-end contextual spatial-channel attention network (CSCANet) to fully utilize shallow features and improve classification performance.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Geochemistry & Geophysics
Di Wang, Jing Zhang, Bo Du, Gui-Song Xia, Dacheng Tao
Summary: Deep learning has achieved great success in aerial image understanding for remote sensing research. However, most existing models are pretrained with ImageNet weights which hinder their fine-tuning performance on downstream aerial scene tasks due to domain gaps. This study empirically investigates RS pretraining on aerial images, training different networks from scratch using the MillionAID dataset to obtain pretrained backbones. Results show that RS pretraining enhances performance in scene recognition and RS-related semantics tasks, but task discrepancies still exist, highlighting the need for further research on large-scale pretraining datasets and effective methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Agronomy
Daniel Kukenbrink, Fabian D. Schneider, Bernhard Schmid, Jean-Philippe Gastellu-Etchegorry, Michael E. Schaepman, Felix Morsdorf
Summary: The three-dimensional distribution of light within forest ecosystems plays a key role in species competition, coexistence, ecosystem functioning, productivity, and diversity. Recent advances in technology provide new insights into light distribution within forest canopies. Combining laser scanning and optical measurements, this study analyzes the impact of canopy structure and optical properties on light extinction in temperate and tropical forests. It is found that canopy structure drives light extinction, with tropical forests exhibiting larger 3D heterogeneity. The use of detailed 3D modeling is crucial for understanding light-related mechanisms affecting species in complex forest ecosystems.
AGRICULTURAL AND FOREST METEOROLOGY
(2021)
Article
Environmental Sciences
Jing Xie, Fabia Husler, Rogier de Jong, Barbara Chimani, Sarah Asam, Yeran Sun, Michael E. Schaepman, Mathias Kneubuhler
Summary: Shifts in phenology due to climate change in mountainous ecosystems are evident in the European Alps, with earlier SOS and increased spring temperatures having a predominant influence on phenological changes. Snow cover duration and melting days have a secondary impact on phenology.
JOURNAL OF GEOPHYSICAL RESEARCH-BIOGEOSCIENCES
(2021)
Article
Ecology
Andrew K. Skidmore, Nicholas C. Coops, Elnaz Neinavaz, Abebe Ali, Michael E. Schaepman, Marc Paganini, W. Daniel Kissling, Petteri Vihervaara, Roshanak Darvishzadeh, Hannes Feilhauer, Miguel Fernandez, Nestor Fernandez, Noel Gorelick, Ilse Geizendorffer, Uta Heiden, Marco Heurich, Donald Hobern, Stefanie Holzwarth, Frank E. Muller-Karger, Ruben Van De Kerchove, Angela Lausch, Pedro J. Leitau, Marcelle C. Lock, Caspar A. Mucher, Brian O'Connor, Duccio Rocchini, Woody Turner, Jan Kees Vis, Tiejun Wang, Martin Wegmann, Vladimir Wingate
Summary: Remote sensing of geospatial biodiversity patterns is crucial as a complement to field observations, with high potential to enhance our understanding of global biodiversity. By compiling a prioritized list of remote sensing biodiversity products, it is possible to further improve monitoring and applicability of the EBV framework. Linking remote sensing products to EBVs will accelerate product generation and improve reporting on biodiversity from local to global scales.
NATURE ECOLOGY & EVOLUTION
(2021)
Article
Remote Sensing
Ghasem Ronoud, Parviz Fatehi, Ali A. Darvishsefat, Erkki Tomppo, Jaan Praks, Michael E. Schaepman
Summary: This study investigated the use of various satellite datasets and their combination for estimating aboveground biomass, with the conclusion that the combination of optical (L8 or S2) and SAR (S1) data improves the estimation accuracy of broadleaved Hyrcanian forests.
CANADIAN JOURNAL OF REMOTE SENSING
(2021)
Article
Ecology
Carla Guillen-Escriba, Fabian D. Schneider, Bernhard Schmid, Andrew Tedder, Felix Morsdorf, Reinhard Furrer, Andreas Hueni, Pascal A. Niklaus, Michael E. Schaepman
Summary: Trait-based ecology has been challenging to combine field-based assessment methods with remote sensing approaches, but this study successfully integrated these two methods in a temperate forest setting. The results showed that taxonomic and environmental variation had significant impacts on plant traits, while within-species variation was influenced by small-scale spatial and residual variation.
ECOLOGY AND EVOLUTION
(2021)
Correction
Ecology
Andrew K. Skidmore, Nicholas C. Coops, Elnaz Neinavaz, Abebe Ali, Michael E. Schaepman, Marc Paganini, W. Daniel Kissling, Petteri Vihervaara, Roshanak Darvishzadeh, Hannes Feilhauer, Miguel Fernandez, Nestor Fernandez, Noel Gorelick, Ilse Geijzendorffer, Uta Heiden, Marco Heurich, Donald Hobern, Stefanie Holzwarth, Frank E. Muller-Karger, Ruben van de Kerchove, Angela Lausch, Pedro J. Leitao, Marcelle C. Lock, Caspar A. Mucher, Brian O'Connor, Duccio Rocchini, Woody Turner, Jan Kees Vis, Tiejun Wang, Martin Wegmann, Vladimir Wingate
NATURE ECOLOGY & EVOLUTION
(2021)
Correction
Ecology
Andrew K. Skidmore, Nicholas C. Coops, Elnaz Neinavaz, Abebe Ali, Michael E. Schaepman, Marc Paganini, W. Daniel Kissling, Petteri Vihervaara, Roshanak Darvishzadeh, Hannes Feilhauer, Miguel Fernandez, Nestor Fernandez, Noel Gorelick, Ilse Geijzendorffer, Uta Heiden, Marco Heurich, Donald Hobern, Stefanie Holzwarth, Frank E. Muller-Karger, Ruben Van De Kerchove, Angela Lausch, Pedro J. Leitao, Marcelle C. Lock, Caspar A. Mucher, Brian O'Connor, Duccio Rocchini, Woody Turner, Jan Kees Vis, Tiejun Wang, Martin Wegmann, Vladimir Wingate
NATURE ECOLOGY & EVOLUTION
(2021)
Article
Geochemistry & Geophysics
David Small, Christoph Rohner, Nuno Miranda, Marius Ruetschi, Michael E. Schaepman
Summary: This article presents a methodology for producing wide-area backscatter images. By combining backscatter measurements of a single region seen from multiple satellite tracks, the method provides wide-area coverage and corrects for slope effects. The approach is suitable for various applications, such as wet snow monitoring, land cover classification, or short-term change detection.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Meinrad Abegg, Ruedi Boesch, Michael E. Schaepman, Felix Morsdorf
Summary: The study focuses on the impact of laser beam diameter and signal triggering on the quality of point clouds in forested environments. It shows that small objects are poorly represented in point clouds and are further affected by large laser beam diameters, dense forest stands, and long distances from the scanning device. Simulations provided in the study offer insights for decision-making regarding the selection of suitable TLS devices and survey configurations for forest inventories.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Environmental Sciences
Rose Pritchard, Thomas Alexandridis, Mary Amponsah, Nabil Ben Khatra, Dan Brockington, Tomas Chiconela, Jesus Ortuno Castillo, Issa Garba, Marta Gomez-Gimenez, Menghestab Haile, Clarisse Kagoyire, Mahlatse Kganyago, Dorothea Kleine, Tesfaye Korme, Alemu A. Manni, Nosiseko Mashiyi, Jadwiga Massninga, Foster Mensah, Maurice Mugabowindekwe, Vivianne Meta, Mark Noort, Patricia Perez Ramirez, Juan Suarez Beltran, Evence Zoungrana
Summary: This article provides a critical perspective on the capacity development required to support the creation of more impactful satellite Earth Observation (EO) data services. Based on a needs assessment conducted as part of the AfriCultuReS project, proximate factors inhibiting the success of EO data services are identified, and these challenges are linked to deeper issues.
ENVIRONMENTAL DEVELOPMENT
(2022)
Article
Ecology
Julia S. Joswig, Christian Wirth, Meredith C. Schuman, Jens Kattge, Bjorn Reu, Ian J. Wright, Sebastian D. Sippel, Nadja Rueger, Ronny Richter, Michael E. Schaepman, Peter M. van Bodegom, J. H. C. Cornelissen, Sandra Diaz, Wesley N. Hattingh, Koen Kramer, Frederic Lens, Ulo Niinemets, Peter B. Reich, Markus Reichstein, Christine Roemermann, Franziska Schrodt, Madhur Anand, Michael Bahn, Chaeho Byun, Giandiego Campetella, Bruno E. L. Cerabolini, Joseph M. Craine, Andres Gonzalez-Melo, Alvaro G. Gutierrez, Tianhua He, Pedro Higuchi, Herve Jactel, Nathan J. B. Kraft, Vanessa Minden, Vladimir Onipchenko, Josep Penuelas, Valerio D. Pillar, Enio Sosinski, Nadejda A. Soudzilovskaia, Evan Weiher, Miguel D. Mahecha
Summary: The study reveals that variation in plant size is mainly influenced by latitudinal gradients in water or energy limitation, while variation in leaf economics traits is primarily influenced by climate, soil fertility, and their interactions. These findings help improve predictions and understanding of biodiversity patterns and the impacts of climate change on biogeochemical cycles.
NATURE ECOLOGY & EVOLUTION
(2022)
Correction
Ecology
Andrew K. Skidmore, Nicholas C. Coops, Elnaz Neinavaz, Abebe Ali, Michael E. Schaepman, Marc Paganini, W. Daniel Kissling, Petteri Vihervaara, Roshanak Darvishzadeh, Hannes Feilhauer, Miguel Fernandez, Nestor Fernandez, Noel Gorelick, Ilse Geijzendorffer, Uta Heiden, Marco Heurich, Donald Hobern, Stefanie Holzwarth, Frank E. Muller-Karger, Ruben Van de Kerchove, Angela Lausch, Pedro J. Leitao, Marcelle C. Lock, Caspar A. Mucher, Brian O'Connor, Duccio Rocchini, Claudia Roeoesli, Woody Turner, Jan Kees Vis, Tiejun Wang, Martin Wegmann, Vladimir Wingate
NATURE ECOLOGY & EVOLUTION
(2021)
Review
Environmental Sciences
Angela Lausch, Michael E. Schaepman, Andrew K. Skidmore, Eusebiu Catana, Lutz Bannehr, Olaf Bastian, Erik Borg, Jan Bumberger, Peter Dietrich, Cornelia Glaesser, Jorg M. Hacker, Rene Hoefer, Thomas Jagdhuber, Sven Jany, Andras Jung, Arnon Karnieli, Reinhard Klenke, Toralf Kirsten, Uta Koedel, Wolfgang Kresse, Ulf Mallast, Carsten Montzka, Markus Moeller, Hannes Mollenhauer, Marion Pause, Minhaz Rahman, Franziska Schrodt, Christiane Schmullius, Claudia Schuetze, Peter Selsam, Ralf-Uwe Syrbe, Sina Truckenbrodt, Michael Vohland, Martin Volk, Thilo Wellmann, Steffen Zacharias, Roland Baatz
Summary: This paper provides a comprehensive overview of using remote sensing techniques for monitoring geomorphology and introduces a new perspective for defining and recording the characteristics of geomorphodiversity using remote sensing data. The five characteristics discussed in this paper are geomorphic genesis diversity, geomorphic trait diversity, geomorphic structural diversity, geomorphic taxonomic diversity, and geomorphic functional diversity. The paper also discusses the challenges and limitations of monitoring geomorphodiversity using remote sensing and presents new approaches and methods for monitoring geomorphodiversity. The importance of the digitization process and data science in geomorphology research is emphasized.
Article
Engineering, Electrical & Electronic
Marius Voegtli, Daniel Schlapfer, Rudolf Richter, Andreas Hueni, Michael E. Schaepman, Mathias Kneubuehler
Summary: This article evaluates the transfer of methods from spaceborne to airborne acquisitions and introduces a new Lambertian/statistical-empirical correction method. The new method showed good performance in airborne data and suggests that effects of higher spatial resolution may compromise previously successful methods.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Geography
T. K. Alexandridis, G. Ovakoglou, I. Cherif, M. Gomez Gimenez, G. Laneve, D. Kasampalis, D. Moshou, S. Kartsios, M. C. Karypidou, E. Katragkou, S. Herrera Garcia, M. Kganyago, N. Mashiyi, K. Pattnayak, A. Challinor, R. Pritchard, D. Brockington, C. Kagoyire, J. Suarez Beltran
Summary: Earth observation data are crucial for monitoring agriculture and providing early warnings in Africa, contributing to optimizing crop production and water resource management to enhance food security.
TRANSACTIONS IN GIS
(2021)