Detecting Extratropical Cyclones of the Northern Hemisphere with Single Shot Detector

TL;DR

This paper introduces a deep learning model using Single Shot Detector to identify and classify extratropical cyclones in the Northern Hemisphere, demonstrating high accuracy especially for mature stages.

Contribution

The paper develops a novel workflow for labeling and preprocessing cyclone images and adapts SSD for multi-stage cyclone detection, which is a new application in meteorology.

Findings

Achieved 86.64% mean Average Precision for mature ETC detection.

Demonstrated effective multi-class classification of ETC stages.

Showed potential for future meteorological applications.

Abstract

In this paper, we propose a deep learning-based model to detect extratropical cyclones (ETCs) of northern hemisphere, while developing a novel workflow of processing images and generating labels for ETCs. We first label the cyclone center by adapting an approach from Bonfanti et.al. [1] and set up criteria of labeling ETCs of three categories: developing, mature, and declining stages. We then propose a framework of labeling and preprocessing the images in our dataset. Once the images and labels are ready to serve as inputs, we create our object detection model named Single Shot Detector (SSD) to fit the format of our dataset. We train and evaluate our model with our labeled dataset on two settings (binary and multiclass classifications), while keeping a record of the results. Finally, we achieved relatively high performance with detecting ETCs of mature stage (mean Average Precision is 86.64%), and an acceptable result for detecting ETCs of all three categories (mean Average Precision 79.34%). We conclude that the single-shot detector model can succeed in detecting ETCs of different stages, and it has demonstrated great potential in the future applications of ETC detection in other relevant settings.