Article
Chemistry, Analytical
Ivan Brkic, Marko Sevrovic, Damir Medak, Mario Miler
Summary: The European Commission has published an EU Road Safety Framework to reduce road fatalities. This study utilized satellite imagery and the Yolo object detector to automatically detect school routes, crosswalks, and divided carriageways. The accuracy achieved for school zones and divided carriageways was 0.988 and 0.950, respectively, while the accuracy for crosswalks ranged from 0.957 to 0.988.
Article
Environmental Sciences
Varunan Theenathayalan, Shubha Sathyendranath, Gemma Kulk, Nandini Menon, Grinson George, Anas Abdulaziz, Nick Selmes, Robert J. W. Brewin, Anju Rajendran, Sara Xavier, Trevor Platt
Summary: A growing coastal population has led to increased pollution in coastal and inland water bodies in the tropics. The use of satellite technology, such as the MSI sensor on Sentinel-2, allows for continuous monitoring of water quality variables like chlorophyll-a and total suspended matter at high resolutions. In order to monitor water quality in the tropical VKW system in India, two regionally tuned satellite algorithms were developed and tested, and they outperformed existing algorithms.
Article
Environmental Sciences
Sarina Little, Tamlin M. Pavelsky, Faisal Hossain, Sheikh Ghafoor, Grant M. Parkins, Sarah K. Yelton, Megan Rodgers, Xiao Yang, Jean-Francois Cretaux, Catherine Hein, Mohammad Arman Ullah, Debolina Halder Lina, Hanne Thiede, Darren Kelly, Donald Wilson, Simon N. Topp
Summary: This study examines water storage variations in small lakes in four states in the USA, using lake level measurements gathered by citizen scientists and lake surface area measurements from optical satellite imagery. The results show that water storage variations between lake pairs are moderately positively correlated on average, with a substantial spread in the degree of correlation. The distance between lake pairs and the extent to which their volume changes are correlated exhibit a weak but statistically significant negative relationship.
Article
Engineering, Civil
Mohammad Danesh-Yazdi, Majid Bayati, Massoud Tajrishy, Behdad Chehrenegar
Summary: By utilizing a machine learning model combined with field data and satellite imagery, the study successfully quantified the relationship between water depth and surface reflectance in Lake Urmia. Significant changes in water depth were observed in the past few years, indicating a remarkable improvement in the restoration status of the lake.
JOURNAL OF HYDROLOGY
(2021)
Article
Environmental Sciences
Dejun Wan, Jiapeng Gao, Ruiting Song, Lei Song, Dongliang Ning
Summary: This study evaluates the reliability and uncertainty of using background soils to assess heavy metal pollution and risks in Chinese lakes, through comparing them with assessments based on background sediments. The results reveal large uncertainties in the assessments using background soils, with Hg and Cd showing relatively higher uncertainties. Despite limited human activities, the lakes still receive considerable heavy metal influx via regional atmospheric transport. The assessments indicate moderate to considerable ecological risks, primarily contributed by Hg and Cd.
Article
Engineering, Electrical & Electronic
Bin Cao, Ruru Deng, Shulong Zhu, Yongming Liu, Yeheng Liang, Longhai Xiong
Summary: This study introduces multiangular imagery into physics-based bathymetry to compensate for the shortage of bathymetric spectral bands, and proposes a selective bathymetric retrieval method to eliminate the negative effect of nonoptimal image data on depth retrieval. The method selectively uses multiangular image data to improve depth retrieval accuracy.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Environmental Sciences
Teerapong Panboonyuen, Chaiyut Charoenphon, Chalermchon Satirapod
Summary: This paper proposes a semantic segmentation method called MeViT for remote sensing image analysis. By integrating medium-resolution multi-branch architectures with vision transformers (ViTs), MeViT can learn semantically rich and spatially precise multi-scale representations. Extensive experiments on a dataset of Thailand scenes demonstrate that MeViT outperforms existing methods in semantic segmentation.
Article
Computer Science, Artificial Intelligence
Mayank Dixit, Kuldeep Chaurasia, Vipul Kumar Mishra
Summary: This paper introduces a novel deep learning architecture for building extraction from Sentinel-2 satellite images, which outperformed existing models in literature when tested on satellite images from three densely populated cities in India.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Meteorology & Atmospheric Sciences
Ming Lei, Tianhai Cheng, Xiaoyang Li, Shuaiyi Shi, Xin Zuo, Hong Guo, Yu Wu
Summary: This paper uses deep learning techniques combined with high-resolution satellite image data to successfully detect NH3 point sources at regional scale, providing powerful technical support for NH3 emission regulation and control.
ATMOSPHERIC RESEARCH
(2021)
Article
Geosciences, Multidisciplinary
Andrew L. Mullen, Jennifer D. Watts, Brendan M. Rogers, Mark L. Carroll, Clayton D. Elder, Jonas Noomah, Zachary Williams, Jordan A. Caraballo-Vega, Allison Bredder, Eliza Rickenbaugh, Eric Levenson, Sarah W. Cooley, Jacqueline K. Y. Hung, Greg Fiske, Stefano Potter, Yili Yang, Charles E. Miller, Susan M. Natali, Thomas A. Douglas, Ethan D. Kyzivat
Summary: Small water bodies, such as ponds, have a significant impact on Earth System processes, but detecting and monitoring them using satellite imagery has been challenging. A new approach using high-resolution optical satellite imagery and deep learning methods allows for mapping seasonal changes in pond and lake areas. This method has various applications including assessing water resources, land cover change, wildlife management, and biogeochemical modeling.
GEOPHYSICAL RESEARCH LETTERS
(2023)
Article
Environmental Sciences
Daniel M. Simms, Alex M. Hamer, Irmgard Zeiler, Lorenzo Vita, Toby W. Waine
Summary: Understanding the relationship between land use and opium production is crucial for monitoring and developing effective counter narcotics policy in Afghanistan. This study develops a generalized model for automatic agricultural land classification using satellite imagery. The model shows high accuracy and robustness, and can be used in other regions with similar landscapes.
Article
Environmental Sciences
Miao Liu, Li Wang, Fangdao Qiu
Summary: This study developed a two-step model to retrieve dissolved oxygen concentration in Lake Taihu using MODIS Aqua images. By utilizing machine learning methods, dissolved oxygen can be estimated from parameters such as water temperature, water clarity, and chlorophyll-a. The research revealed spatial and temporal variations in dissolved oxygen in Lake Taihu, which are influenced by factors such as water quality and air temperature.
FRONTIERS IN ENVIRONMENTAL SCIENCE
(2022)
Article
Environmental Sciences
Zhili Zhang, Meng Lu, Shunping Ji, Huafen Yu, Chenhui Nie
Summary: Accurately extracting water bodies from high-resolution remote sensing imagery is a major challenge. This study addresses the difficulty in identifying water body boundaries and semantic inconsistency in feature fusion by designing a novel multi-feature extraction and fusion module. Extensive experiments demonstrate that the proposed method achieves state-of-the-art segmentation performance and robustness in challenging water body extraction scenarios.
Article
Remote Sensing
John Brandt, Fred Stolle
Summary: This study presents a globally consistent method to identify trees with canopy diameters greater than 3 m using medium-resolution optical and radar imagery. By training a fully convolutional network, the proposed model achieves over 75% user's and producer's accuracy in hectares with low to medium tree cover, and 95% accuracy in hectares with dense tree cover. The method can increase the accuracy of monitoring tree presence in areas with sparse and scattered tree cover by up to 20%, and reduce commission and omission error in mountainous and very cloudy regions by nearly half.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2021)
Article
Meteorology & Atmospheric Sciences
Mohammed Ombadi, Phu Nguyen, Soroosh Sorooshian, Kuo-lin Hsu
Summary: The study utilizes information-theoretic measures to analyze the relationship between satellite infrared imagery and precipitation, determining that the correlation varies by region and season, and that the dependence between infrared and precipitation shows diminishing returns with increasing temporal or spatial aggregation. These findings can guide the development of operational algorithms for estimating precipitation using satellite infrared imagery.
ATMOSPHERIC RESEARCH
(2021)