Satellite Aerosol Optical Depth Retrieval Based on Fully Connected Neural Network (FCNN) and a Combine Algorithm of Simplified Aerosol Retrieval Algorithm and Simplified and Robust Surface Reflectance Estimation (SREMARA)

Fan, Yulong; Sun, Lin

doi:10.1109/jstars.2023.3281777

Cited by 7 publications

(1 citation statement)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast to LUT-based AOD calculations that require iteration or interpolation processes [17], the retrieval of the AOD using a trained machine learning model is instantaneous [22]. Benefiting from this advantage, many classical machine learning models have been applied to describe the nonlinear relationships between satellite measurements and the AOD, such as the fully connected neural network (FCNN), gradient boosting framework (LGBM) [23,24], support vector machine (SVM) [21], and backpropagation neural network (BPNN) [17]. These methods have the advantage of satellite aerosol retrieval but are affected by the training sample.…”

Section: Introductionmentioning

confidence: 99%

Wide and Deep Learning Model for Satellite-Based Real-Time Aerosol Retrievals in China

Luo,

Zou,

Zang

et al. 2024

Atmosphere

View full text Add to dashboard Cite

Machine learning methods have been recognized as rapid methods for satellite-based aerosol retrievals but have not been widely applied in geostationary satellites. In this study, we developed a wide and deep learning model to retrieve the aerosol optical depth (AOD) using Himawari-8. Compared to traditional deep learning methods, we embedded a “wide” modeling component and tested the proposed model across China using independent training (2016–2018) and test (2019) datasets. The results showed that the “wide” model improves the accuracy and enhances model interpretability. The estimates exhibited better accuracy (R2 = 0.81, root-mean-square errors (RMSEs) = 0.19, and within the estimated error (EE) = 63%) than those of the deep-only models (R2 = 0.78, RMSE = 0.21, within the EE = 58%). In comparison with extreme gradient boosting (XGBoost) and Himawari-8 V2.1 AOD products, there were also significant improvements. In addition to higher accuracy, the interpretability of the proposed model was superior to that of the deep-only model. Compared with other seasons, higher contributions of spring to the AOD concentrations were interpreted. Based on the application of the wide and deep learning model, the near-real-time variation of the AOD over China could be captured with an ultrafine temporal resolution.

show abstract

Section: Introductionmentioning

confidence: 99%

Wide and Deep Learning Model for Satellite-Based Real-Time Aerosol Retrievals in China

Luo,

Zou,

Zang

et al. 2024

Atmosphere

View full text Add to dashboard Cite

show abstract

A deep learning-based combination method of spatio-temporal prediction for regional mining surface subsidence

Xiao,

Tao,

et al. 2024

Sci Rep

View full text Add to dashboard Cite

In coal mining areas, surface subsidence poses significant risks to human life and property. Fortunately, surface subsidence caused by coal mining can be monitored and predicted by using various methods, e.g., probability integral method and deep learning (DL) methods. Although DL methods show promise in predicting subsidence, they often lack accuracy due to insufficient consideration of spatial correlation and temporal nonlinearity. Considering this issue, we propose a novel DL-based approach for predicting mining surface subsidence. Our method employs K-means clustering to partition spatial data, allowing the application of a gate recurrent unit (GRU) model to capture nonlinear relationships in subsidence time series within each partition. Optimization using snake optimization (SO) further enhances model accuracy globally. Validation shows our method outperforms traditional Long Short-Term Memory (LSTM) and GRU models, achieving 99.1% of sample pixels with less than 8 mm absolute error.

show abstract

Cloud detection sample generation algorithm for nighttime satellite imagery based on daytime data and machine learning application

Shi,

Fan,

Sun

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Highly accurate nighttime cloud detection in satellite imagery is challenging due to the absence of visible to near-infrared (0.38–3 μm, VNI) data, which is critical for distinguishing clouds from other ground features. Fortunately, Machine learning (ML) techniques can more effectively leverage the limited wavelength information and show high-accuracy cloud detection based on vast sample volume. However, accurately distinguishing cloud pixels solely through thermal infrared bands (8–14 μm, TIR) is challenging, and acquiring numerous, high-quality and representative samples of nighttime images for ML proves to be unattainable in practice. Given the thermal infrared radiation transmission process and the fact that daytime and nighttime have the same source of radiance in TIR, we propose a sample generation idea that uses daytime images to provide samples of nighttime cloud detection, which is different from the traditional sample construction methods (e.g., manually label, simulation and transfer learning method), and can obtain samples effectively. Based on this idea, nighttime cloud detection experiments were carried out for MODIS, GF-5 (02) and Himawari-8 satellites, respectively. The results were validated by the Lidar cloud product and manual labels and show that our nighttime cloud detection result has higher accuracy than MYD35 (78.17%) and the ML model trained by nighttime manual labels (75.86%). The accuracy of the three sensors is 82.19%, 88.71%, and 79.34%, respectively. Moreover, we validated and discussed the performance of our algorithm on various surface types (vegetation, urban, barren and water). The results revealed that the accuracy of the three sensors over barren was found to be poor and varied with surface types but overall high. Our study can provide a novel perspective on nighttime cloud detection of muti-spectral satellite imagery. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-024-78889-z.

show abstract

Satellite Aerosol Optical Depth Retrieval Based on Fully Connected Neural Network (FCNN) and a Combine Algorithm of Simplified Aerosol Retrieval Algorithm and Simplified and Robust Surface Reflectance Estimation (SREMARA)

Cited by 7 publications

References 75 publications

Wide and Deep Learning Model for Satellite-Based Real-Time Aerosol Retrievals in China

Wide and Deep Learning Model for Satellite-Based Real-Time Aerosol Retrievals in China

A deep learning-based combination method of spatio-temporal prediction for regional mining surface subsidence

Cloud detection sample generation algorithm for nighttime satellite imagery based on daytime data and machine learning application

Contact Info

Product

Resources

About