{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T17:07:27Z","timestamp":1790960847949,"version":"4.1.0"},"reference-count":45,"publisher":"Public Library of Science (PLoS)","issue":"12","license":[{"start":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T00:00:00Z","timestamp":1764892800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Tropical cyclones pose significant threats to coastal regions, and have a major negative influence on the environment and society. Precise cyclone identification and intensity estimation are crucial for effective early warning systems and disaster prevention. Traditional methods rely on manual interpretation and empirical models, often lacking efficiency and accuracy. This study proposes a deep learning framework that utilizes satellite image sequences for cyclone detection, classification, and intensity estimation. Unlike conventional models relying solely on spatial or manual features, the proposed hybrid architecture integrates Convolutional Neural Networks (CNNs) and ConvLSTM to learn spatiotemporal patterns jointly. Key innovations include the clustering-based cyclone region isolation method, sequence-level data augmentation, and the use of SMOTE to mitigate class imbalance. The proposed approach demonstrates substantial improvement in accuracy over baseline models, achieving 99.16% accuracy for binary classification using VGG16. An accuracy of 81.1\u2009\u00b1\u20094.33% across cyclone intensity levels, and an RMSE of 7.79\u2009\u00b1\u20091.27 knots in wind speed prediction using the ConvLSTM-based model. All models are evaluated using 5-fold cross-validation on CIMSS Tropical Data Archive and IMD Best-Track datasets. Overall, these results show an exciting potential for future use of deep learning for real time forecasting and early warning systems. Future work will also look to improve or increase model generalization, either through using ensemble learning, or potentially more complex architectures and larger datasets.<\/jats:p>","DOI":"10.1371\/journal.pone.0330705","type":"journal-article","created":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T18:24:25Z","timestamp":1764959065000},"page":"e0330705","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":4,"title":["ConvLSTM-based tropical cyclone intensity estimation and classification using satellite imagery over the North Indian ocean"],"prefix":"10.1371","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-2029-4666","authenticated-orcid":true,"given":"Manju","family":"M. S.","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Harsh","family":"Pateriya","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rajeev Kumar","family":"Gupta","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Deepak Singh","family":"Tomar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Punit","family":"Gupta","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7159-2651","authenticated-orcid":true,"given":"Asmir","family":"Butkovic","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,12,5]]},"reference":[{"issue":"3","key":"pone.0330705.ref001","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1038\/ngeo779","article-title":"Tropical cyclones and climate change","volume":"3","author":"TR Knutson","year":"2010","journal-title":"Nature Geosci"},{"issue":"05","key":"pone.0330705.ref002","article-title":"Hurricanes of the North Atlantic: climate and society","volume":"37","year":"2000","journal-title":"Choice Reviews Online"},{"key":"pone.0330705.ref003","unstructured":"Dvorak VF. A technique for the analysis and forecasting of tropical cyclone intensities from satellite pictures. 1973."},{"issue":"9","key":"pone.0330705.ref004","doi-asserted-by":"crossref","first-page":"1195","DOI":"10.1175\/BAMS-87-9-1195","article-title":"The Dvorak Tropical Cyclone Intensity Estimation Technique: A Satellite-Based Method that Has Endured for over 30 Years","volume":"87","author":"C Velden","year":"2006","journal-title":"Bull Amer Meteor Soc"},{"issue":"2","key":"pone.0330705.ref005","doi-asserted-by":"crossref","first-page":"88","DOI":"10.1016\/j.tcrr.2022.07.002","article-title":"Verification of tropical cyclones (TC) wind structure forecasts from global NWP models and ensemble prediction systems (EPSs)","volume":"11","author":"X Lu","year":"2022","journal-title":"Tropical Cyclone Research and Review"},{"issue":"17","key":"pone.0330705.ref006","doi-asserted-by":"crossref","first-page":"4188","DOI":"10.3390\/rs15174188","article-title":"Estimation of Tropical Cyclone Intensity via Deep Learning Techniques from Satellite Cloud Images","volume":"15","author":"B Tong","year":"2023","journal-title":"Remote Sensing"},{"key":"pone.0330705.ref007","doi-asserted-by":"crossref","DOI":"10.3389\/feart.2023.1285138","article-title":"Tropical cyclone intensity estimation through convolutional neural network transfer learning using two geostationary satellite datasets","volume":"11","author":"H Jung","year":"2024","journal-title":"Front Earth Sci"},{"issue":"2","key":"pone.0330705.ref008","doi-asserted-by":"crossref","first-page":"692","DOI":"10.1109\/TIP.2017.2766358","article-title":"Tropical Cyclone Intensity Estimation Using a Deep Convolutional Neural Network","volume":"27","author":"R Pradhan","year":"2018","journal-title":"IEEE Trans Image Process"},{"key":"pone.0330705.ref009","article-title":"Deep Learning Experiments for Tropical Cyclone Intensity Forecasts","author":"X Wenwei","year":"2021","journal-title":"Weather and Forecasting"},{"key":"pone.0330705.ref010","article-title":"Physics-augmented Deep Learning to Improve Tropical Cyclone Intensity and Size Estimation from Satellite Imagery","author":"J-Y Zhuo","year":"2021","journal-title":"Monthly Weather Review"},{"issue":"9","key":"pone.0330705.ref011","article-title":"Tropical cyclone intensity detection by geometric features of cyclone images and multilayer perceptron","volume":"1","author":"C Kar","year":"2019","journal-title":"SN Appl Sci"},{"key":"pone.0330705.ref012","doi-asserted-by":"crossref","first-page":"2070","DOI":"10.1109\/JSTARS.2021.3050767","article-title":"Tropical Cyclone Intensity Classification and Estimation Using Infrared Satellite Images With Deep Learning","volume":"14","author":"C-J Zhang","year":"2021","journal-title":"IEEE J Sel Top Appl Earth Observations Remote Sensing"},{"issue":"1","key":"pone.0330705.ref013","doi-asserted-by":"crossref","first-page":"5230","DOI":"10.1038\/s41598-025-89196-6","article-title":"A machine learning model for detecting and quantifying tropical cyclone related disturbance and recovery in estuaries","volume":"15","author":"MA Ibrar","year":"2025","journal-title":"Sci Rep"},{"issue":"10","key":"pone.0330705.ref014","doi-asserted-by":"crossref","first-page":"2165","DOI":"10.1007\/s12524-024-01929-8","article-title":"An End-to-End Deep Learning Framework for Cyclone Intensity Estimation in North Indian Ocean Region Using Satellite Imagery","volume":"52","author":"MK Mawatwal","year":"2024","journal-title":"J Indian Soc Remote Sens"},{"key":"pone.0330705.ref015","doi-asserted-by":"crossref","first-page":"12008","DOI":"10.1109\/JSTARS.2025.3563472","article-title":"TFA-Net: A Temporal Feature Aggregation Framework for Tropical Cyclone Intensity Estimation From Satellite Images","volume":"18","author":"Z Zhao","year":"2025","journal-title":"IEEE J Sel Top Appl Earth Observations Remote Sensing"},{"key":"pone.0330705.ref016","doi-asserted-by":"crossref","first-page":"9893","DOI":"10.1109\/JSTARS.2025.3556475","article-title":"Boost Cyclone Intensity Estimation via Processing Motion Information and Embedding Rotation Invariance","volume":"18","author":"X Cheng","year":"2025","journal-title":"IEEE J Sel Top Appl Earth Observations Remote Sensing"},{"key":"pone.0330705.ref017","doi-asserted-by":"crossref","first-page":"70095","DOI":"10.1109\/ACCESS.2025.3561355","article-title":"Tropical Cyclone Intensity Prediction Using Spatio-Temporal Data Fusion","volume":"13","author":"K Sattar","year":"2025","journal-title":"IEEE Access"},{"issue":"1","key":"pone.0330705.ref018","doi-asserted-by":"crossref","DOI":"10.1007\/s12040-022-02042-5","article-title":"Tropical Cyclone intensity prediction based on hybrid learning techniques","volume":"132","author":"P Varalakshmi","year":"2023","journal-title":"J Earth Syst Sci"},{"key":"pone.0330705.ref019","first-page":"1","article-title":"A Multiscale and Multilayer Feature Extraction Network With Dual Attention for Tropical Cyclone Intensity Estimation","volume":"62","author":"Z Ma","year":"2024","journal-title":"IEEE Trans Geosci Remote Sensing"},{"issue":"1","key":"pone.0330705.ref020","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1007\/s00376-024-3191-1","article-title":"Vision Transformer for Extracting Tropical Cyclone Intensity from Satellite Images","volume":"42","author":"Y Tian","year":"2024","journal-title":"Adv Atmos Sci"},{"key":"pone.0330705.ref021","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2024.3494255","article-title":"Estimating Tropical Cyclone Intensity Using an STIA Model From Himawari-8 Satellite Images in the Western North Pacific Basin","volume":"62","author":"R Zhang","year":"2024","journal-title":"IEEE Trans Geosci Remote Sensing"},{"key":"pone.0330705.ref022","doi-asserted-by":"crossref","first-page":"107887","DOI":"10.1016\/j.atmosres.2024.107887","article-title":"Cross-basin incremental learning for tropical cyclone intensity estimation","volume":"315","author":"J Ding","year":"2025","journal-title":"Atmospheric Research"},{"issue":"2","key":"pone.0330705.ref023","doi-asserted-by":"crossref","first-page":"215","DOI":"10.3390\/rs16020215","article-title":"MT-GN: Multi-Task-Learning-Based Graph Residual Network for Tropical Cyclone Intensity Estimation","volume":"16","author":"Z Zhao","year":"2024","journal-title":"Remote Sensing"},{"key":"pone.0330705.ref024","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2024.3489674","article-title":"FHDTIE: Fine-Grained Heterogeneous Data Fusion for Tropical Cyclone Intensity Estimation","volume":"62","author":"G Xu","year":"2024","journal-title":"IEEE Trans Geosci Remote Sensing"},{"issue":"1","key":"pone.0330705.ref025","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1007\/s00376-024-3301-0","article-title":"Short-Term Rolling Prediction of Tropical Cyclone Intensity Based on Multi-Task Learning with Fusion of Deviation-Angle Variance and Satellite Imagery","volume":"42","author":"W Tian","year":"2024","journal-title":"Adv Atmos Sci"},{"issue":"9","key":"pone.0330705.ref026","doi-asserted-by":"crossref","first-page":"1860","DOI":"10.3390\/rs13091860","article-title":"NDFTC: A New Detection Framework of Tropical Cyclones from Meteorological Satellite Images with Deep Transfer Learning","volume":"13","author":"S Pang","year":"2021","journal-title":"Remote Sensing"},{"key":"pone.0330705.ref027","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2024.3441729","article-title":"SSN: A Novel CNN-Based Architecture for Classification of Tropical Cyclone Images From INSAT-3D","volume":"62","author":"S Pal","year":"2024","journal-title":"IEEE Trans Geosci Remote Sensing"},{"issue":"8","key":"pone.0330705.ref028","article-title":"Tropical cyclone intensity classification from infrared images of clouds over Bay of Bengal and Arabian Sea using machine learning classifiers","volume":"14","author":"C Kar","year":"2021","journal-title":"Arab J Geosci"},{"issue":"2","key":"pone.0330705.ref029","article-title":"Improving Typhoon Predictions by Integrating Data\u2010Driven Machine Learning Model With Physics Model Based on the Spectral Nudging and Data Assimilation","volume":"12","author":"Z Niu","year":"2025","journal-title":"Earth and Space Science"},{"issue":"6","key":"pone.0330705.ref030","doi-asserted-by":"crossref","first-page":"1965","DOI":"10.1007\/s12524-024-02101-y","article-title":"A Machine Learning Approach for Detecting Rapid Intensification in Tropical Cyclones","volume":"53","author":"T Sharma","year":"2025","journal-title":"J Indian Soc Remote Sens"},{"issue":"22","key":"pone.0330705.ref031","doi-asserted-by":"crossref","first-page":"11975","DOI":"10.1073\/pnas.1920849117","article-title":"Global increase in major tropical cyclone exceedance probability over the past four decades","volume":"117","author":"JP Kossin","year":"2020","journal-title":"Proc Natl Acad Sci U S A"},{"key":"pone.0330705.ref032","unstructured":"India Meteorological Department. Cyclone Best Track Data. 2023. https:\/\/mausam.imd.gov.in"},{"issue":"1","key":"pone.0330705.ref033","doi-asserted-by":"crossref","DOI":"10.1186\/s40537-019-0197-0","article-title":"A survey on Image Data Augmentation for Deep Learning","volume":"6","author":"C Shorten","year":"2019","journal-title":"J Big Data"},{"key":"pone.0330705.ref034","article-title":"Very deep convolutional networks for large-scale image recognition","author":"K Simonyan","year":"2014","journal-title":"Computer Vision and Pattern Recognition"},{"issue":"18","key":"pone.0330705.ref035","doi-asserted-by":"crossref","first-page":"8972","DOI":"10.3390\/app12188972","article-title":"Deep Residual Learning for Image Recognition: A Survey","volume":"12","author":"M Shafiq","year":"2022","journal-title":"Applied Sciences"},{"key":"pone.0330705.ref036","doi-asserted-by":"crossref","unstructured":"Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the Inception Architecture for Computer Vision. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. 2818\u201326. doi: 10.1109\/cvpr.2016.308","DOI":"10.1109\/CVPR.2016.308"},{"issue":"1","key":"pone.0330705.ref037","doi-asserted-by":"crossref","DOI":"10.1186\/s12864-019-6413-7","article-title":"The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation","volume":"21","author":"D Chicco","year":"2020","journal-title":"BMC Genomics"},{"key":"pone.0330705.ref038","first-page":"226","article-title":"A density-based algorithm for discovering clusters in large spatial databases with noise","author":"M Ester","year":"1996","journal-title":"Knowledge Discovery and Data Mining"},{"issue":"8","key":"pone.0330705.ref039","doi-asserted-by":"crossref","first-page":"837","DOI":"10.1109\/34.531803","article-title":"Texture features for browsing and retrieval of image data","volume":"18","author":"BS Manjunath","year":"1996","journal-title":"IEEE Trans Pattern Anal Machine Intell"},{"issue":"6","key":"pone.0330705.ref040","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1109\/TPAMI.1986.4767851","article-title":"A Computational Approach to Edge Detection","author":"J Canny","year":"1986","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"5","key":"pone.0330705.ref041","doi-asserted-by":"crossref","first-page":"691","DOI":"10.1007\/s13351-022-2016-3","article-title":"An Empirical Model of Tropical Cyclone Intensity Forecast in the Western North Pacific","volume":"36","author":"C Ma","year":"2022","journal-title":"J Meteorol Res"},{"key":"pone.0330705.ref042","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1613\/jair.953","article-title":"SMOTE: Synthetic Minority Over-sampling Technique","volume":"16","author":"NV Chawla","year":"2002","journal-title":"jair"},{"key":"pone.0330705.ref043","doi-asserted-by":"crossref","first-page":"50982","DOI":"10.1109\/ACCESS.2021.3065939","article-title":"Designing Deep-Based Learning Flood Forecast Model With ConvLSTM Hybrid Algorithm","volume":"9","author":"M Moishin","year":"2021","journal-title":"IEEE Access"},{"key":"pone.0330705.ref044","first-page":"161","article-title":"Performance Analysis of InceptionV3, VGG16, and Resnet50 Models for Crevices Recognition on Surfaces","volume-title":"Lecture Notes in Networks and Systems","author":"Ch Mukesh","year":"2024"},{"issue":"8","key":"pone.0330705.ref045","doi-asserted-by":"crossref","first-page":"14213","DOI":"10.1109\/TNNLS.2024.3511575","article-title":"Fully Tensorized Lightweight ConvLSTM Neural Networks for Hyperspectral Image Classification","volume":"36","author":"T-Y Ma","year":"2025","journal-title":"IEEE Trans Neural Netw Learn Syst"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0330705","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T18:24:35Z","timestamp":1764959075000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0330705"}},"subtitle":[],"editor":[{"given":"Bijay","family":"Halder,","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,12,5]]},"references-count":45,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2025,12,5]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0330705","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,5]]}}}