Detecting Extratropical Cyclones of the Northern Hemisphere with Single Shot Detector
This work addresses cyclone detection for meteorology and climate science, but it is incremental as it applies an existing object detection method to a new dataset with custom labeling.
The paper tackled the problem of detecting extratropical cyclones in the Northern Hemisphere by developing a deep learning-based model using a Single Shot Detector, achieving a mean Average Precision of 86.64% for mature cyclones and 79.34% for all three categories.
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.