Home // INTELLI 2018, The Seventh International Conference on Intelligent Systems and Applications // View article

A 3D Convolutional Neural Network for Anomalous Propagation Identification

Authors:
Hansoo Lee
Jonggeun Kim
Sungshin Kim

Keywords: Pattern recognition; Deep learning; 3D convolutional neural network; Anomalous propagation; Radar data analysis.

Abstract:
Radar is one of the most popular and widely used weather observation devices because of its better performance compared to other remote sensing devices. However, the observation results of the radar unavoidably contain unwanted signals, called non-precipitation echoes, which include anomalous propagation. These represent a negative influence, especially in the quantitative precipitation estimation. Therefore, it is essential to remove the anomalous propagation in the radar data for accurate weather forecasting. In this paper, we implemented a three-dimensional convolutional neural network for classifying the anomalous propagation in the radar data. Without considering feature engineering, which is difficult and mostly hand-crafted, we were able to obtain improved performance in the classification with actual occurrence cases of the echo.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2018

Publication date: June 24, 2018

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-646-0

Location: Venice, Italy

Dates: from June 24, 2018 to June 28, 2018