A Deep Learning Framework for the Detection of Tropical Cyclones From Satellite Images

Nair, Aravind; Srujan, K. S. S. Sai; Kulkarni, Sayali R.; Alwadhi, Kshitij; Jain, Navya; Kodamana, Hariprasad; Sandeep, S.; John, Viju O.

doi:10.1109/lgrs.2021.3131638

Cited by 15 publications

(7 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Extracting small buildings and closely located buildings is not investigated. umar would like you to join Abey Resume Template Aravind Nair et al, [5] proposed that tropical cyclones are the foremost dangerous weather systems that originate over the tropical oceans, with roughly 90 storms forming annually throughout the globe. Quick identification and tracking of Tropical Cyclones are crucial for advanced warning to sensitive locations.…”

Section: Related Workmentioning

confidence: 99%

Tropical Cyclone Tracking and Forecasting Using BiGRU [TCTFB]

Khandelwal

et al. 2022

Preprint

View full text Add to dashboard Cite

The purpose of this study is to identify the exact location of cyclones to avert cyclone-related damage. Knowing about natural disasters like cyclones in advance will help with planning and preparations as they can be extremely dangerous. Numerous methods have been developed in the past to forecast cyclones and gauge their severity. It is a difficult task that demands swiftness and effectiveness. On the INSAT-3D dataset, a hybrid model of CNN and Bidi-rectional GRU is created in this study to estimate the position of the next cyclone. The INSAT-3D satellite pictures’ IR and visible images are analyzed and segmented using K-means clustering. The model’s training time is 400 ms, as observed. There have been comparisons between a variety of prior methods, including CNN-LSTM and a hybrid model that combines CNN and Bidirectional-LSTM. The suggested model’s experimental results show a training loss of 58.6879.

show abstract

Section: Related Workmentioning

confidence: 99%

Tropical Cyclone Tracking and Forecasting Using BiGRU [TCTFB]

Khandelwal

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Segmentation of the shape and size of the detected TC in high‐resolution satellite images was also provided by Nair et al. (2022). To this extent, a pipeline consisting of a Mask Region‐Convolutional Neural Network (R‐CNN) detector, a wind speed filter and a CNN classifier was adopted to accurately detect TCs.…”

Section: Introductionmentioning

confidence: 99%

“…Similarly, a DL-based object detection approach was proposed in Wang et al (2020) with the aim of retrieving the TC center through segmentation, edge detection, circle fitting, and comprehensive decision of satellite infrared images. Segmentation of the shape and size of the detected TC in high-resolution satellite images was also provided by Nair et al (2022). To this extent, a pipeline consisting of a Mask Region-Convolutional Neural Network (R-CNN) detector, a wind speed filter and a CNN classifier was adopted to accurately detect TCs.…”

mentioning

confidence: 99%

An Ensemble Machine Learning Approach for Tropical Cyclone Localization and Tracking From ERA5 Reanalysis Data

Accarino,

Donno,

Immorlano

et al. 2023

Earth and Space Science

View full text Add to dashboard Cite

Tropical Cyclones (TCs) are counted among the most destructive phenomena that can be found in nature. Every year, globally an average of 90 TCs occur over tropical waters, and global warming is making them stronger and more destructive. The accurate localization and tracking of such phenomena have become a relevant and interesting area of research in weather and climate science. Traditionally, TCs have been identified in large climate data sets through the use of deterministic tracking schemes that rely on subjective thresholds. This study presents a Machine Learning (ML) ensemble approach for locating TCs center coordinates. The ensemble combines TCs center estimates of different ML models that agree about the presence of a TC in input data. ERA5 reanalysis data was used for model training and testing jointly with the International Best Track Archive for Climate Stewardship (IBTrACS) records. Compared to single models estimates, the ML ensemble approach was able to improve TCs localization in terms of Euclidean Distance with respect to the observed TCs locations from IBTrACS. Moreover, a hybrid tracking scheme was defined: starting from the individual TC center locations detected by the ML ensemble approach, a deterministic tracking algorithm was used for reconstructing TC trajectories. The hybrid tracking scheme was then compared with four deterministic trackers reported in literature, achieving a Probability of Detection and a False Alarm Rate of 71.49% and 23%, respectively, over 40 years of reanalysis data.

show abstract

“…In recent years, deep learning techniques have gained fast development and more and more efforts have been made to investigate TCs via such techniques for detecting, identifying and forecasting key parameters/features of TCs (Chen et al ., 2020; Nair et al ., 2022; Tong et al ., 2022b). Technically, identification of TC centre belongs to an object detection problem.…”

Section: Introductionmentioning

confidence: 99%

Identification of tropical cyclone centre based on satellite images via deep learning techniques

Long

Tong

et al. 2022

Intl Journal of Climatology

View full text Add to dashboard Cite

A tropical cyclone (TC) is a highly destructive natural disaster. Accurate identification of key parameters of TCs is prerequisite for most TC-related research and practices. The centre position is one of TC's basic parameters. However, comparison of TC best track data released by different meteorological institutes usually indicates a noticeable discrepancy for this parameter among varied data sources. In this study, efforts are made towards identifying the centre location of TCs via deep learning techniques, based on TC satellite cloud images (SCIs). Six deep learning models are analysed and compared. YOLOv4 model achieved a confidence of 99.84%, which is better than other models. In addition, we further explore the factors affecting the positioning accuracy of the YOLOv4 model and its application to the location identification of multiple TCs and the tracking of individual TCs. Results demonstrate that the YOLOv4 model has a probability exceeding 99% for identifying multiple TC locations and also performs well for single TC tracking.

show abstract

A Deep Learning Framework for the Detection of Tropical Cyclones From Satellite Images

Cited by 15 publications

References 30 publications

Tropical Cyclone Tracking and Forecasting Using BiGRU [TCTFB]

Tropical Cyclone Tracking and Forecasting Using BiGRU [TCTFB]

An Ensemble Machine Learning Approach for Tropical Cyclone Localization and Tracking From ERA5 Reanalysis Data

Identification of tropical cyclone centre based on satellite images via deep learning techniques

Contact Info

Product

Resources

About