Abstract
Image data classification using machine learning is an effective method for detecting atmospheric phenomena. However, extreme weather events with a small number of cases cause a decrease in classification prediction accuracy owing to the imbalance in data between the target class and the other classes. To build a highly accurate classification model, I held a data analysis competition to determine the best classification performance for two classes of cloud image data, specifically tropical cyclones including precursors and other classes. For the top models in the competition, minority data oversampling, majority data undersampling, ensemble learning, deep layer neural networks, and cost-effective loss functions were used to improve the classification performance of the imbalanced data. In particular, the best model of 209 submissions succeeded in improving the classification capability by 65.4% over similar conventional methods in a measure of the low false alarm ratio.
Highlights
In recent years, deep learning, which is machine learning using multilayered neural networks, has attracted much attention in various research and industrial fields as a technology that greatly exceeds the performance of conventional methods
Deep convolutional neural networks, which are specialized for image recognition, are highly efficient at extracting spatial feature patterns (Krizhevsky et al 2012)
4 Results and discussion we show the classification performance of all submissions, including the top model of the competition described in the previous section
Summary
Deep learning, which is machine learning using multilayered neural networks, has attracted much attention in various research and industrial fields as a technology that greatly exceeds the performance of conventional methods. Deep convolutional neural networks, which are specialized for image recognition, are highly efficient at extracting spatial feature patterns (Krizhevsky et al 2012). One of the simplest tasks that a convolutional neural network can perform is the classification of image categories. Image classification has been applied to detect hurricanes, fronts, and atmospheric rivers (Liu et al 2016), tropical cyclones (Kim et al 2017), and precursors of previous studies have reported interesting results by using classifications, the imbalance in the amount of data between classes is still an issue. Matsuoka et al (2018) classified 50,000 positive and 1,000,000 negative example images, with a balance of 20 times. This class imbalance is known to cause a decrease in classification performance This class imbalance is known to cause a decrease in classification performance (e.g. Sun et al 2009)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.