大气科学学报2025，Vol.48Issue(3)：389-403,15.DOI:10.13878/j.cnki.dqkxxb.20250306001
ImageNet数据能否帮助改进基于深度学习的云图分类准确率?
Can ImageNet data improve deep learning-based cloud image classification accuracy?
摘要
Abstract
Clouds play a vital role in the earth-atmosphere system.Accurate cloud classification is essential for improving regional weather forecasts and understanding the global energy budget.However,precise and objective identification of ground-based cloud images remain challenging,primarily due to the limited availability of stand-ardized cloud image datasets.This constraint hampers the further development of deep learning models for cloud classification.To address this issue,we propose a methodological hypothesis:can pre-training deep learning models on large scale,non-meteorological datasets enhance the accuracy of cloud classification,followed by fine-tuning with domain-specific cloud imagery?To test this hypothesis,we implement three deep learning architec-tures-two convolutional neural networks(ResNet50 and MobileNet-V2)and a self-attention-based Vision Transformer(ViT)—to perform ground-based cloud classification.We conduct a comparative analysis of models trained solely on cloud image datasets and those pre-trained on the ImageNet dataset before being fine-tuning with cloud data.Our results highlight the impact of pre-training strategies across different architectures.Even without pre-training,ResNet50 and MobileNet-V2 achieve strong baseline performance,with average F1 scores of 0.85 and 0.87,respectively.Notably,the ViT model shows significant improvement with pre-training:the F1 score in-creases from 0.79 to 0.96-a 21.5％enhancement-demonstrating the importance of large-scale pre-training for architectures reliant on spatial feature extraction.Analysis of misclassified cases reveals that deep learning models primarily rely on spatial characteristics to distinguish cloud types.This suggests that incorporating auxiliary mete-orological parameters-such as cloud-base height and thickness-as embedded features may further enhance model interpretability and performance.The performance gains from pre-training are largely attributed to improved edge detection and morphological pattern recognition,which are especially beneficial for complex architectures like ViT.In addition to these theoretical contributions,this study achieves practical implementation by deploying a stable cloud classification model on a mobile platform(available at http://43.142.162.19:5174/).This applica-tion supports real-time cloud-type identification via photo uploads and provides educational content,thereby pro-moting public engagement with atmospheric science and demonstrating the real-world applicability of deep learn-ing for ground-based cloud observation. Future research will focus on integrating cloud physical properties-such as thermodynamic parameters and radiative characteristics-into deep learning models.Fusing physical constraints with visual features may enhance classification robustness,reduce data requirements,and improve interpretability,paving the way for explainable AI systems in atmospheric sciences.In conclusion,this study establishes deep learning as an effective approach for au-tomated cloud classification and underscores the critical role of pre-training,especially for advanced architectures.The mobile deployment further bridges meteorological research and public outreach,demonstrating the dual scien-tific and educational value of AI-powered cloud classification systems.关键词
迁移学习/云图分类/ViT模型/预训练模型/非常规气象数据Key words
transfer learning/cloud classification/ViT model/pre-trained models/unconventional meteorological data引用本文复制引用
季焱,叶灵熙,黄智勇,彭婷,高智伟,孔德璇,吉璐莹,朱寿鹏,智协飞..ImageNet数据能否帮助改进基于深度学习的云图分类准确率?[J].大气科学学报,2025,48(3):389-403,15.基金项目
中国南方电网有限责任公司科技项目(YNKJXM20222172) （YNKJXM20222172）
云南省中青年学术和技术带头人后备人才项目(202105AC160014) （202105AC160014）
国家自然科学基金项目(42475151) （42475151）
贵州省气象局省市联合科研基金项目(黔气科合SS[2023]38号) （黔气科合SS[2023]38号）
贵州省基础研究计划(自然科学)面上项目(黔科合基础-zk[2025]面上319) （自然科学）
无锡学院引进人才科研启动专项 （）
