A FCN Approach to Blockage Correction in Radars

Wu, Hao and Liu, Qi and Liu, Xiaodong and Zhang, Yonghong and Xu, Xiaolong and Bilal, Muhammad (2022) A FCN Approach to Blockage Correction in Radars. In: Proceedings - 2021 IEEE International Conference on Dependable, Autonomic and Secure Computing, International Conference on Pervasive Intelligence and Computing, International Conference on Cloud and Big Data Computing and International Conference on Cybe. Proceedings - 2021 IEEE International Conference on Dependable, Autonomic and Secure Computing, International Conference on Pervasive Intelligence and Computing, International Conference on Cloud and Big Data Computing and International Conference on Cybe . Institute of Electrical and Electronics Engineers Inc., CAN, pp. 482-487. ISBN 9781665421744

Abstract

Doppler weather radar is the most widely used convection detector with the highest resolution in the ground. Echo reflectance data from the weather radar is the key reference for the meteorological department to carry out severe convective weather forecast and early warning, quantitative precipitation estimation(QPE) and quantitative precipitation forecast(QPF). However, in the process of radar detection, it is inevitable to be affected by obstacles, ground object echo interference, radar echo attenuation and other phenomena, resulting in poor data quality of detection results. Therefore, it is very important to correct the missing or disturbed data. On the other hand, with the rapid development of artificial intelligence technology in recent years, more and more meteorological researchers begin to introduce deep learning and other machine learning methods into the research of meteorological field such as weather radar data processing. In this paper, a deep convolutional encoder-decoder network is proposed to correct the beam blocking of weather radar. In this study, the correction of radar beam blockage is regarded as an image inpainting problem. It's the first trying to use deep learning to realize the correction of radar beam blockage. Experiment shows that the method proposed in this paper is significantly better than the traditional method in accuracy, error rate, false alarm rate and other aspects. The method can directly identify and correct the blocking area, and the operation procedure is simple compared traditional methods.