Methodology for Severe Convective Cloud Identification Using Lightweight Neural Network Model Ensembling
Open Access
- 7 June 2024
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 16 (12) , 2070
Abstract
This study introduces an advanced ensemble methodology employing lightweight neural network models for identifying severe convective clouds from FY-4B geostationary meteorological satellite imagery. We have constructed a FY-4B based severe convective cloud dataset by a combination of algorithms and expert judgment. Through the ablation study of a model ensembling combination of multiple specialized lightweight architectures—ENet, ESPNet, Fast-SCNN, ICNet, and MobileNetV2—the optimal EFNet (ENet- and Fast-SCNN-based network) not only achieves real-time processing capabilities but also ensures high accuracy in severe weather detection. EFNet consistently outperformed traditional, heavier models across several key performance indicators: achieving an accuracy of 0.9941, precision of 0.9391, recall of 0.9201, F1 score of 0.9295, and computing time of 18.65 s over the test dataset of 300 images (~0.06 s per 512 × 512 pic). ENet shows high precision but misses subtle clouds, while Fast-SCNN has high sensitivity but lower precision, leading to misclassifications. EFNet’s ensemble approach balances these traits, enhancing overall predictive accuracy. The ensemble method of lightweight models effectively aggregates the diverse strengths of the individual models, optimizing both speed and predictive performance.Keywords
This publication has 63 references indexed in Scilit:
- Volcanic clouds detection applying machine learning techniques to GNSS radio occultationsGPS Solutions, 2024
- A deep learning ensemble approach for predicting tropical cyclone rapid intensificationAtmospheric Science Letters, 2023
- Recognition algorithm for deep convective clouds based on FY4ANeural Computing and Applications, 2022
- An ensemble method to forecast 24-h ahead solar irradiance using wavelet decomposition and BiLSTM deep learning networkEarth Science Informatics, 2021
- A review on deep learning techniques for cloud detection methodologies and challengesSignal, Image and Video Processing, 2021
- ICNet for Real-Time Semantic Segmentation on High-Resolution Images2018
- ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation2018
- Simultaneous observations of atmospheric tides from combined in situ and remote observations at Mars from the MAVEN spacecraftJournal of Geophysical Research: Planets, 2016
- A Tutorial on the Cross-Entropy MethodAnnals of Operations Research, 2005
- Using Canny's criteria to derive a recursively implemented optimal edge detectorInternational Journal of Computer Vision, 1987