☆ 4.7 Article

Quantifying uncertainty in soil moisture retrieval using a Bayesian neural network framework

COMPUTERS AND ELECTRONICS IN AGRICULTURE (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Materials Science, Characterization & Testing

Uncertainty quantification in super-resolution guided wave array imaging using a variational Bayesian deep learning approach

Homin Song et al.

Summary: A Bayesian deep learning approach is proposed to quantify and interpret uncertainties in super-resolution guided wave array imaging. The approach successfully quantifies two types of uncertainties: aleatoric uncertainty inherent in the data and epistemic uncertainty associated with the Bayesian deep learning model.

NDT & E INTERNATIONAL (2023)

添加到收藏夹

Article Geosciences, Multidisciplinary

DNN-Based Retrieval of Arctic Sea Ice Concentration From GNSS-R and Its Effects on the Synoptic-Scale Forecasting as Supplementary Observation Source

Lu Yang et al.

Summary: Using delay-Doppler maps of GNSS-R from TechDemoSat-1 and considering sea ice and ocean interaction, a deep neural network based method for retrieving Arctic sea ice concentration (SIC) is proposed. The retrieval method shows potential for future GNSS-R applications in Arctic missions. Compared to Hamburg University's SIC products, the root mean square errors (RMSE) of retrieved results in March and June 2016 are 0.0284 and 0.0415, respectively. When the retrieved GNSS-R SIC data are assimilated as supplementary passive microwave remote-sensing data, it improves the accuracy of Arctic SIC forecast, especially in edge regions of sea ice, with a maximum decrease of approximately 17% in the 24-hr forecast time and over 5% in 72-hr.

GEOPHYSICAL RESEARCH LETTERS (2023)

添加到收藏夹

Article Environmental Sciences

Soil moisture retrieval from Sentinel-1 using a first-order radiative transfer model-A case-study over the Po-Valley

Raphael Quast et al.

Summary: This paper presents the retrieval of high-resolution soil moisture data from Sentinel-1 C-band Synthetic Aperture Radar (SAR) backscatter measurements using a new bistatic radiative transfer modeling framework (RT1). The performance of the soil moisture retrievals is analyzed with respect to the ERA5-Land reanalysis dataset. The results demonstrate the potential of RT1 for the retrieval of high-resolution soil moisture data from SAR time series.

REMOTE SENSING OF ENVIRONMENT (2023)

添加到收藏夹

Article Environmental Sciences

Soil moisture dominates the variation of gross primary productivity during hot drought in drylands

Ruonan Qiu et al.

Summary: The frequency and severity of hot drought will increase in the future due to climate change and human activities, threatening terrestrial ecosystems and human societies. High vapor pressure deficit (VPD) and low soil moisture (SM) are the main characteristics of hot drought, leading to water stress on vegetation and exacerbating hydrological drought and ecosystem risks. However, our understanding of the effects of high VPD and low SM on vegetation productivity is limited.

SCIENCE OF THE TOTAL ENVIRONMENT (2023)

添加到收藏夹

Article Multidisciplinary Sciences

Soil moisture-atmosphere coupling accelerates global warming

Liang Qiao et al.

Summary: Soil moisture-atmosphere coupling induces non-linear warming via the 'warmer climate - drier soil' feedback, exerting an accelerating effect on global warming and extreme high temperatures. The projection shows that SA-driven warming will exceed 0.5°C over extratropical landmasses by the end of the 21st Century, increasing the likelihood of extreme high temperatures.

NATURE COMMUNICATIONS (2023)

添加到收藏夹

Article Engineering, Electrical & Electronic

Quasi-Global Assessment of Deep Learning-Based CYGNSS Soil Moisture Retrieval

M. M. Nabi et al.

Summary: A high-resolution global soil moisture product is crucial for hydrologic, meteorological, and agricultural applications. Utilizing GNSS signals and deep learning models can help reduce the uncertainty of soil moisture estimation under different conditions.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2023)

添加到收藏夹

Article Agriculture, Multidisciplinary

Developing machine learning models with multisource inputs for improved land surface soil moisture in China

Lei Wang et al.

Summary: Accurate and spatially continuous land surface soil moisture data is important for heat transfer, energy exchange, and agricultural dryness analysis. This study used five machine learning models to map regional soil moisture and evaluated the importance of input features. The random forest regression model performed the best, and geographical location was identified as the most important factor in the soil moisture estimation model.

COMPUTERS AND ELECTRONICS IN AGRICULTURE (2022)

添加到收藏夹

Article Agriculture, Multidisciplinary

Soil moisture content estimation in winter wheat planting area for multi-source sensing data using CNNR

Jiao Guo et al.

Summary: This study combines UWB radar and multispectral remote sensing data to construct a one-dimensional regression convolutional neural network model for rapid and accurate estimation of soil moisture content in farmland. The introduction of different vegetation indices significantly improves the accuracy of the model.

COMPUTERS AND ELECTRONICS IN AGRICULTURE (2022)

添加到收藏夹

Article Environmental Sciences

A Bayesian approach for remote sensing of chlorophyll-a and associated retrieval uncertainty in oligotrophic and mesotrophic lakes

Mortimer Werther et al.

Summary: In this study, Bayesian probabilistic neural networks (BNNs) were developed for estimating chlorophyll-a concentration (chla) in oligotrophic and mesotrophic lakes. The BNNs provided per-pixel uncertainty percentages and were evaluated through various assessments. The results showed that the BNNs achieved higher accuracy in oligotrophic waters, but the accuracy decreased as nutrient levels increased. The study also demonstrated the importance of uncertainty estimation in improving the quality of the chla result.

REMOTE SENSING OF ENVIRONMENT (2022)

添加到收藏夹

Article Environmental Sciences

Quasi-global machine learning-based soil moisture estimates at high spatio-temporal scales using CYGNSS and SMAP observations

Fangni Lei et al.

Summary: High-resolution global soil moisture mapping is crucial for meteorological, hydrological, and agricultural applications. This study proposes a machine learning approach to map quasi-global soil moisture using CYGNSS mission's bistatic reflectance observations integrated with remote sensing products and SMAP data, providing daily 9 km soil moisture mapping.

REMOTE SENSING OF ENVIRONMENT (2022)

添加到收藏夹

Article Engineering, Electrical & Electronic

Uncertainty Quantification for Deep Unrolling-Based Computational Imaging

Canberk Ekmekci et al.

Summary: This article proposes a learning-based image reconstruction framework that incorporates the observation model into the reconstruction task and quantifies uncertainties based on deep unrolling and Bayesian neural networks. The framework demonstrates the ability to characterize uncertainties in magnetic resonance imaging and computed tomography reconstruction problems. The research motivates future exploration of utilizing uncertainty information to develop more accurate and robust image reconstruction methods for imaging problems.

IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING (2022)

添加到收藏夹

Article Geosciences, Multidisciplinary

Quantifying the uncertainty of precipitation forecasting using probabilistic deep learning

Lei Xu et al.

Summary: Precipitation forecasting is important in weather science, and data-driven methods can improve accuracy. This study proposes a joint uncertainty modeling framework that estimates predictive uncertainty and achieves better accuracy and reduced uncertainty compared to existing methods.

HYDROLOGY AND EARTH SYSTEM SCIENCES (2022)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Uncertainty Quantification in Deep MRI Reconstruction

Vineet Edupuganti et al.

Summary: This study aims to quantify the uncertainty in image recovery with DL models by developing a probabilistic reconstruction scheme using VAEs. Empirical experiments show that adversarial losses introduce more uncertainty, while recurrent unrolled nets reduce prediction uncertainty and risk.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2021)

添加到收藏夹

Article Agriculture, Multidisciplinary

A synergic method of Sentinel-1 and Sentinel-2 images for retrieving soil moisture content in agricultural regions

Jiatan Liang et al.

Summary: The paper aims to accurately estimate soil moisture content in Dawen River basin by developing a synergic method of radar and optical data, as well as an on-site sampling scheme. Results showed that the coefficient of determination between the estimated vegetation moisture content and field observation was 0.655 from March to May in 2020.

COMPUTERS AND ELECTRONICS IN AGRICULTURE (2021)

添加到收藏夹

Review Environmental Sciences

A roadmap for high-resolution satellite soil moisture applications - confronting product characteristics with user requirements

Jian Peng et al.

Summary: Soil moisture observations from satellite data are important for various applications, and significant progress has been made in estimating soil moisture. However, there is a need for further development of high-resolution soil moisture products to meet the requirements of different disciplines.

REMOTE SENSING OF ENVIRONMENT (2021)

添加到收藏夹

Review Computer Science, Artificial Intelligence

A review of uncertainty quantification in deep learning: Techniques, applications and challenges

Moloud Abdar et al.

Summary: Uncertainty quantification (UQ) methods are essential in reducing uncertainties in optimization and decision making processes. Bayesian approximation and ensemble learning techniques are widely used types of UQ methods.

INFORMATION FUSION (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach

Yan Jia et al.

Summary: In this article, global soil moisture (SM) is estimated using machine learning (ML) regression aided by a preclassification strategy, resulting in significant improvements in SM estimations with different ML algorithms. The optimal XGBoost model is selected for SM prediction, achieving satisfactory daily and seasonal outcomes at a global scale with a high correlation coefficient value. The study also evaluates the extensive temporal and spatial variations in CYGNSS SM predictions, revealing that reflectivity plays a main role in SM estimation, followed by vegetation, and that roughness may become more important in extremely dry areas. The approach also compares SM predictions from SMAP and CYGNSS against in situ measurements, showing similar low-median unbiased RMSEs and a good performance of CYGNSS-based SM predictions.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2021)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Uncertainty quantification using Bayesian neural networks in classification: Application to biomedical image segmentation

Yongchan Kwon et al.

COMPUTATIONAL STATISTICS & DATA ANALYSIS (2020)

添加到收藏夹

Article Environmental Sciences

Machine Learning-Based CYGNSS Soil Moisture Estimates over ISMN sites in CONUS

Volkan Senyurek et al.

REMOTE SENSING (2020)

添加到收藏夹

Article Environmental Sciences

Comprehensive Evaluation of Using TechDemoSat-1 and CYGNSS Data to Estimate Soil Moisture over Mainland China

Ting Yang et al.

REMOTE SENSING (2020)

添加到收藏夹

Article Environmental Sciences

Pan-tropical soil moisture mapping based on a three-layer model from CYGNSS GNSS-R data

Qingyun Yan et al.

REMOTE SENSING OF ENVIRONMENT (2020)

添加到收藏夹

Article Environmental Sciences

Evaluations of Machine Learning-Based CYGNSS Soil Moisture Estimates against SMAP Observations

Volkan Senyurek et al.

REMOTE SENSING (2020)

添加到收藏夹

Article Robotics

A General Framework for Uncertainty Estimation in Deep Learning

Antonio Loquercio et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2020)

添加到收藏夹

Article Multidisciplinary Sciences

Stable classification with limited sample: transferring a 30-m resolution sample set collected in 2015 to mapping 10-m resolution global land cover in 2017

Peng Gong et al.

SCIENCE BULLETIN (2019)

添加到收藏夹

Article Optics

Reliable deep-learning-based phase imaging with uncertainty quantification

Yujia Xue et al.

OPTICA (2019)

添加到收藏夹

Article Engineering, Electrical & Electronic

Analysis of CYGNSS Data for Soil Moisture Retrieval

Maria Paola Clarizia et al.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2019)

添加到收藏夹

Article Environmental Sciences

Evaluating the Performance of Satellite-Derived Vegetation Indices for Estimating Gross Primary Productivity Using FLUXNET Observations across the Globe

Xiaojuan Huang et al.

REMOTE SENSING (2019)

添加到收藏夹

Article Geosciences, Multidisciplinary

Soil Moisture Sensing Using Spaceborne GNSS Reflections: Comparison of CYGNSS Reflectivity to SMAP Soil Moisture

C. C. Chew et al.

GEOPHYSICAL RESEARCH LETTERS (2018)

添加到收藏夹

Article Geosciences, Multidisciplinary

Use of Cyclone Global Navigation Satellite System (CyGNSS) Observations for Estimation of Soil Moisture

Hyunglok Kim et al.

GEOPHYSICAL RESEARCH LETTERS (2018)

添加到收藏夹

Article Astronomy & Astrophysics

Uncertainty quantification for radio interferometric imaging: II. MAP estimation

Xiaohao Cai et al.

MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY (2018)

添加到收藏夹

Article Agronomy

Crop yield prediction under soil salinity using satellite derived vegetation indices

Onur Satir et al.

FIELD CROPS RESEARCH (2016)

添加到收藏夹

Article Meteorology & Atmospheric Sciences

The global land data assimilation system

M Rodell et al.

BULLETIN OF THE AMERICAN METEOROLOGICAL SOCIETY (2004)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.