Preprints
https://doi.org/10.5194/amt-2021-405
https://doi.org/10.5194/amt-2021-405

  23 Dec 2021

23 Dec 2021

Review status: this preprint is currently under review for the journal AMT.

Identification of tropical cyclones via deep convolutional neural network based on satellite cloud images

Biao Tong1, Xiangfei Sun2, Jiyang Fu1, Yuncheng He1, and Pakwai Chan3 Biao Tong et al.
  • 1Research Center for Wind Engineering and Engineering Vibration, Guangzhou University, Guangzhou, China
  • 2Guangdong Key Laboratory of Environmental Pollution and Health, School of Environment, Jinan University, Guangzhou, China
  • 3Hong Kong Observatory, Hong Kong, China

Abstract. Tropical Cyclones (TCs) are one of the most destructive natural disasters. For the prevention and mitigation of TC-induced disasters, real-time monitoring and prediction of TCs is essential. At present, satellite cloud images (SCIs) are utilized widely as a basic data source for such studies. Although great achievements have been made in this field, lack of concerns on the identification of TC fingerprint from SCIs have become a potential issue, since it is a prerequisite step for follow-up analyses. This paper presents a methodology which identifies TC fingerprint via Deep Convolutional Neural Network (DCNN) techniques based on SCIs of more than 200 TCs over the Northwest Pacific basin. Two DCNN models have been proposed and validated, which are able to identify the TCs from not only single-TC featured SCIs but also multi-TCs featured SCIs. Results show that both models can reach 96 % of identification accuracy. As the TC intensity strengthens, the accuracy becomes better. To explore how these models work, heat maps are further extracted and analyzed. Results show that all the fingerprint features are focused on clouds during the testing process. For the majority of TC images, the cloud features in TC’s main parts, i.e., eye, eyewall and primary rainbands, are most emphasized, reflecting a consistent pattern with the subjective method.

Biao Tong et al.

Status: open (until 28 Jan 2022)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on amt-2021-405', Anonymous Referee #1, 31 Dec 2021 reply
    • AC1: 'Reply on RC1', Biao Tong, 13 Jan 2022 reply

Biao Tong et al.

Biao Tong et al.

Viewed

Total article views: 286 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
237 45 4 286 1 0
  • HTML: 237
  • PDF: 45
  • XML: 4
  • Total: 286
  • BibTeX: 1
  • EndNote: 0
Views and downloads (calculated since 23 Dec 2021)
Cumulative views and downloads (calculated since 23 Dec 2021)

Viewed (geographical distribution)

Total article views: 245 (including HTML, PDF, and XML) Thereof 245 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 15 Jan 2022
Download
Short summary
In recent years, there has been numerous research on Tropical Cyclone (TC) observation based on satellite cloud images (SCIs), but most methods are limited by low efficiency and subjectivity. To overcome subjectivity and improve efficiency of traditional methods, this paper uses deep learning technology to do further research on fingerprint identification of TC. Results provide an automatic and objective method to distinguish TCs from SCIs, and bring great convenience for subsequent research.