Article
Computer Science, Artificial Intelligence
Dong Wang, Chanyue Wu, Yunpeng Bai, Ying Li, Changjing Shang, Qiang Shen
Summary: This paper proposes a multitask network (MTNet) to achieve joint multispectral (MS) pansharpening for images acquired by different satellites. The MTNet shares generic knowledge between datasets via a task-agnostic subnetwork (TASNet) and adapts this knowledge to specific satellites using task-specific subnetworks (TSSNets). It also introduces band-aware dynamic convolutions (BDConvs) to accommodate various ground scenes and bands. Experimental results demonstrate that the proposed approach outperforms existing state-of-the-art (SOTA) techniques across different datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Review
Computer Science, Artificial Intelligence
Sin Liang Lim, Jaya Sreevalsan-Nair, B. S. Daya Sagar
Summary: This article provides a brief overview of various aspects of data mining of multi spectral image data, with a focus on remote sensing satellite images acquired using multispectral imaging. It reviews different data mining processes, state-of-the-art methods, and applications. The article also emphasizes the importance of understanding data acquisition and preprocessing, and concludes with applications demonstrating knowledge discovery, challenges, and future directions for MSI data mining research.
WILEY INTERDISCIPLINARY REVIEWS-DATA MINING AND KNOWLEDGE DISCOVERY
(2023)
Article
Geochemistry & Geophysics
Shaojia Ge, Hong Gu, Weimin Su, Anne Lonnqvist, Oleg Antropov
Summary: Accurate forest mapping is crucial for forest management and carbon stocks monitoring. This study introduces a novel semi-supervised regression framework for wall-to-wall mapping of continuous forest variables, using contrastive regression and cross-pseudo regression. It achieves higher prediction accuracies for forest tree height compared to traditional models, with a relative root mean square error (rRMSE) of 15.1% on stand level.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Environmental Sciences
Michiel Vlaminck, Laurens Diels, Wilfried Philips, Wouter Maes, Rene Heim, Bart De Wit, Hiep Luong
Summary: In the past two decades, UAVs have become essential in the remote sensing field. Lightweight sensors like LiDAR scanners and hyperspectral cameras have been introduced to the market, but there are few drone systems that effectively combine different sensing modalities. In this paper, a multimodal drone payload and sensor fusion pipeline is presented, which generates subcentimeter accurate multispectral point clouds. The quality of the reconstructed point cloud is significantly improved by combining high-frequency navigation outputs, photogrammetric bundle adjustment, and a dedicated point cloud registration algorithm.
Article
Computer Science, Information Systems
Guojun Fan, Zhibin Pan, Quan Zhou, Jing Dong, Xiaoran Zhang
Summary: This paper proposes a novel reversible data hiding method for multispectral images, which improves the embedding effectiveness through creative matching and complexity computation, and processes different types of pixels using predictors, achieving lower distortion levels.
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2022)
Article
Meteorology & Atmospheric Sciences
Elisabeth Weisz, W. Paul Menzel
Summary: Multisensor satellite data fusion integrates measurements or products from imaging and sounding instruments with different characteristics to improve the detection and monitoring of atmospheric variables, such as trace gas emissions from volcanoes. By combining data from low Earth and geostationary orbits, spatial-temporal fusion can enhance the spatial detail and temporal resolution of trace gas information, aiding in the monitoring of dispersion patterns. This study utilized fusion techniques to track volcanic sulfur dioxide and ash plumes from the Cumbre Vieja volcano eruptions, providing valuable insights for air quality monitoring and aircraft safety systems.
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2023)
Article
Environmental Sciences
Joshua Hrisko, Prathap Ramamurthy, Jorge E. Gonzalez
Summary: A satellite-derived hysteresis model is established to estimate heat storage in urban areas, by relating multispectral satellite radiances and geophysical properties to ground-truth residual heat storage computed with flux instruments. Gradient-boosted regression trees serve as the method to maximize the relationship between satellite data and flux measurements, with the model performing well under varying weather conditions. The model shows lower RMSE and MAE values compared to some ground-to-ground studies, making it one of the few satellite-derived methods that compute direct comparison over different land cover types.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Engineering, Electrical & Electronic
Fernando Perez-Fontan, Vicente Pastoriza-Santos, Fernando Machado, Francisco Poza, Norbert Witternigg, Roman Lesjak
Summary: This article presents the implementation details of a generative satellite maritime propagation channel model at L-band. The parameterization of the channel model for the narrowband case was previously presented, and wider band measurements were performed to estimate power delay profiles. However, due to limitations in measurement range and delay resolution, conclusive information about time spreading caused by the channel could not be gathered. The article discusses the extension of the model to the wideband case and explores the use of a rough surface scattering model to characterize time dispersion.
IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION
(2022)
Article
Engineering, Electrical & Electronic
Aubin Allies, Antoine Roumiguie, Remy Fieuzal, Jean-Francois Dejoux, Anne Jacquin, Amanda Veloso, Luc Champolivier, Frederic Baup
Summary: This article investigates the potential of assimilating synthetic aperture radar (SAR)-derived dry mass (DM) and optically derived green area index (GAI) in an agro-meteorological model for better rapeseed crops modeling. The results show that the assimilation of both SAR and optical data allows for better control of the model, offering promise as a decision support tool for farmers and decision makers.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Fabio Oriani, Matthew F. McCabe, Gregoire Mariethoz
Summary: This article introduces a statistical approach to downscale and bias-correct multispectral satellite data, utilizing a limited training set of very high-resolution images. The proposed technique aims at extending the coverage of high-resolution images with realistic results. By sampling data from a training set, the approach does not require fine-resolution data in the target zone and preserves important properties such as intensity histogram and spatial correlation.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Environmental Sciences
Francesco Romeo, Luigi Mereu, Simona Scollo, Mario Papa, Stefano Corradini, Luca Merucci, Frank Silvio Marzano
Summary: This study combines satellite microwave and millimeter-wave passive sensors with thermal-infrared radiometric data from a Low Earth Orbit spectroradiometer to characterize volcanic plumes. New physical-statistical methods, incorporating machine learning techniques, were developed to detect and retrieve volcanic clouds from the 2014 Kelud and 2015 Calbuco eruptions. The results demonstrate the effectiveness of machine learning for volcanic cloud detection and the accuracy of using combined TIR and MW-MMW observations for estimating volcanic cloud masses.
Article
Environmental Sciences
Svetlana Illarionova, Sergey Nesteruk, Dmitrii Shadrin, Vladimir Ignatiev, Maria Pukalchik, Ivan Oseledets
Summary: The paper introduces a novel image augmentation approach named MixChannel, which improves the accuracy and performance of solving segmentation and classification tasks with multispectral satellite images by mixing auxiliary data from different locations.
Article
Instruments & Instrumentation
Alkha Mohan, M. Venkatesan, P. Prabhavathy, A. Jayakrishnan
Summary: This paper presents a model for predicting rice crop yield using satellite images and climatic data. By collecting vegetation indices and climatic parameters from Moderate Resolution Imaging Spectroradiometer (MODIS) data, the model utilizes a Temporal Convolutional Network (TCN) with a specially designed dilated convolution module to improve prediction accuracy.
INFRARED PHYSICS & TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Gunjan Joshi, Ryo Natsuaki, Akira Hirose
Summary: In the last decade, the increase in earth observation satellites has led to enhanced interest in data fusion techniques. This article proposes a neural network that combines and analyzes SAR and optical sensor data to provide high-resolution classification maps. It introduces a novel activation function called inverse mapping for feature analysis, which helps understand the prominent contributors for classification outputs. The fusion-based results show improved accuracy compared to independent sensors, and inverse mapping provides reasonable explanations.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Engineering, Electrical & Electronic
Dmitry Rashkovetsky, Florian Mauracher, Martin Langer, Michael Schmitt
Summary: It is critical to accurately determine the extent of areas affected by wildfires for fire management and population protection. Traditional algorithms require complex relationships between remote sensing data parameters, while deep learning can automatically learn patterns in complex data. Detecting fire-affected areas from satellite imagery can significantly improve detection efficiency.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Meteorology & Atmospheric Sciences
Yukari Sumi, Hirohiko Masunaga
JOURNAL OF THE ATMOSPHERIC SCIENCES
(2019)
Article
Geochemistry & Geophysics
Husi Letu, Takashi M. Nagao, Takashi Y. Nakajima, Jerome Riedi, Hiroshi Ishimoto, Anthony J. Baran, Huazhe Shang, Miho Sekiguchi, Maki Kikuchi
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Article
Environmental Sciences
Hirohiko Masunaga, Marc Schroeder, Fumie A. Furuzawa, Christian Kummerow, Elke Rustemeier, Udo Schneider
ENVIRONMENTAL RESEARCH LETTERS
(2019)
Article
Meteorology & Atmospheric Sciences
Hirohiko Masunaga, Brian E. Mapes
JOURNAL OF THE ATMOSPHERIC SCIENCES
(2020)
Article
Meteorology & Atmospheric Sciences
Jing Li, Ralph A. Kahn, Jing Wei, Barbara E. Carlson, Andrew A. Lacis, Zhanqing Li, Xichen Li, Oleg Dubovik, Teruyuki Nakajima
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2020)
Article
Geosciences, Multidisciplinary
Jeyavinoth Jeyaratnam, Zhengzhao Johnny Luo, Scott E. Giangrande, Die Wang, Hirohiko Masunaga
Summary: A novel satellite-based method has been developed and evaluated for estimating convective vertical velocity and convective mass flux, showing solid agreement with ground-based radar observations and distinct convective characteristics in different regions.
GEOPHYSICAL RESEARCH LETTERS
(2021)
Article
Meteorology & Atmospheric Sciences
Hirohiko Masunaga, Christopher E. Holloway, Hironari Kanamori, Sandrine Bony, Thorwald H. M. Stein
Summary: Convective self-aggregation is a notable feature in radiative-convective equilibrium simulations, and this study seeks observational signals of convective aggregation in cloud cluster life cycles. The study finds that in heavy precipitation regimes, cloud clusters gather into fewer members with increased high-cloud cover, indicating transient convective aggregation, which is less evident in light precipitation regimes. Energy budget analysis reveals that column moist static energy accumulates before precipitation peaks, primarily due to horizontal moisture advection, and radiative cooling is greater in aggregated composites but the role of radiative-convective feedback remains unclear.
JOURNAL OF CLIMATE
(2021)
Article
Meteorology & Atmospheric Sciences
Hao Wang, Tie Dai, Daisuke Goto, Qing Bao, Bian He, Yimin Liu, Toshihiko Takemura, Teruyuki Nakajima, Guangyu Shi
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2020)
Article
Meteorology & Atmospheric Sciences
Hanii Takahashi, Matthew Lebsock, Zhengzhao Johnny Luo, Hirohiko Masunaga, Cindy Wang
Summary: This paper introduces a convection tracking method based on the IMERG precipitation product, providing more complete records of convective storms' lifecycle and aiding in diagnosing cloud precipitation processes. By investigating the evolution of convective systems before and after peak precipitation, the study demonstrates the utility of IMERG-CT.
JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY
(2021)
Article
Environmental Sciences
Andung Bayu Sekaranom, Emilya Nurjani, Sandy Budi Wibowo, Hirohiko Masunaga
Summary: This study investigated the homogeneity of ice-scattering signals from TRMM Microwave Imagers and its impact on rain-rate estimation bias. Statistical analysis revealed that the homogeneity and organization of precipitation systems can influence the accuracy of rain-rate estimation. Adjustments based on ice-scattering signals at the grid level were found to improve satellite rain-rate estimation accuracy, particularly for inhomogeneous precipitation systems.
Article
Geosciences, Multidisciplinary
Masato Ito, Hirohiko Masunaga
Summary: In this study, A-Train observations and reanalysis data are analyzed to assess the physical processes responsible for the iris effect. The major findings suggest that upper-tropospheric stability plays a role in the changes of anvil cloud fraction, while precipitation efficiency has less control on it. The day and nighttime cloud radiative effects are expected to cancel out over a diurnal cycle, indicating a neutral cloud feedback.
GEOPHYSICAL RESEARCH LETTERS
(2022)
Article
Geosciences, Multidisciplinary
Yueming Cheng, Tie Dai, Junji Cao, Lin Chen, Daisuke Goto, Mayumi Yoshida, Teruyuki Nakajima, Guangyu Shi
Summary: This study presents the first simultaneous assimilation of aerosol optical thicknesses (AOTs) over East Asia from two next-generation geostationary satellites. A new data-control method is proposed to assimilate high quality aerosol products. The results show that the joint assimilation significantly improves aerosol analysis and forecast skill and reduces model biases, particularly in South China.
GEOPHYSICAL RESEARCH LETTERS
(2022)
Article
Meteorology & Atmospheric Sciences
Hirohiko Masunaga
Summary: In this study, the behavior of ITCZ convection near the eastern Pacific was investigated through analysis of satellite observations and reanalysis data. The study found that when precipitation peaks at the ITCZ center, there is a prominent positive peak in diabatic forcing, while when convection develops at the ITCZ edges, there is only a weak diabatic forcing but an import of moist static energy.
JOURNAL OF CLIMATE
(2023)
Article
Meteorology & Atmospheric Sciences
Masafumi Hirose, Shoichi Shige, Takuji Kubota, Fumie A. Furuzawa, Haruya Minda, Hirohiko Masunaga
Summary: The study improves the estimation of precipitation data by updating the low-level precipitation profiles and enhancing the detection of shallow storms. By applying corrections to these factors, precipitation increases by 8% and 11% over land and ocean, respectively, reducing discrepancies in data accuracy in specific regions.
JOURNAL OF THE METEOROLOGICAL SOCIETY OF JAPAN
(2021)
Article
Public, Environmental & Occupational Health
Xerxes Seposo, Kayo Ueda, Sang Seo Park, Kengo Sudo, Toshihiko Takemura, Teruyuki Nakajima
GLOBAL HEALTH ACTION
(2019)
