A Method of Weather Radar Echo Extrapolation Based on Convolutional Neural Networks

Weather radar echo extrapolation techniques possess wide application prospects in short-term forecasting (i.e., nowcasting). Traditional methods of radar echo extrapolation have difficulty obtaining long limitation period data and lack the utilization rate of radar. To solve this problem, this paper proposes a method of weather radar echo extrapolation based on convolutional neural networks (CNNs). To create a strong correlation among contiguous weather radar echo images from traditional CNNs, this method present a new CNN model: Recurrent Dynamic CNNs (RDCNN). RDCNN consists of a recurrent dynamic sub-network and a probability prediction layer, which constructs a cyclic structure in the convolution layer, improving the ability of RDCNN to process time-related images. Nanjing, Hangzhuo and Xiamen experimented with radar data, and compared with traditional methods, our method achieved higher accuracy of extrapolation and extended the limitation period effectively, meeting the requirements for application.

[1]  Pabitra Mitra,et al.  Leveraging Convolutions in Recurrent Neural Networks for Doppler Weather Radar Echo Prediction , 2017, ISNN.

[2]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[3]  L. Li,et al.  Nowcasting of Motion and Growth of Precipitation with Radar over a Complex Orography , 1995 .

[4]  Remko Uijlenhoet,et al.  Raindrop size distributions and radar reflectivity–rain rate relationships for radar hydrology , 2001 .

[5]  Lianwen Jin,et al.  DropSample: A New Training Method to Enhance Deep Convolutional Neural Networks for Large-Scale Unconstrained Handwritten Chinese Character Recognition , 2015, Pattern Recognit..

[6]  Yann LeCun,et al.  Large-scale Learning with SVM and Convolutional for Generic Object Categorization , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[7]  Ming Yang,et al.  DeepFace: Closing the Gap to Human-Level Performance in Face Verification , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[8]  Lukás Burget,et al.  Recurrent neural network based language model , 2010, INTERSPEECH.

[9]  D. Zrnic,et al.  Doppler Radar and Weather Observations , 1984 .

[10]  Xiang Zhang,et al.  OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks , 2013, ICLR.

[11]  Geoffrey E. Hinton,et al.  Generating Text with Recurrent Neural Networks , 2011, ICML.

[12]  Juanzhen Sun,et al.  Use of NWP for Nowcasting Convective Precipitation: Recent Progress and Challenges , 2014 .

[13]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[14]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[15]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[16]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Daniel Povey,et al.  Revisiting Recurrent Neural Networks for robust ASR , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[18]  Luca Maria Gambardella,et al.  Mitosis Detection in Breast Cancer Histology Images with Deep Neural Networks , 2013, MICCAI.

[19]  Tomoo Ushio,et al.  Precipitation Nowcasting with Three-Dimensional Space–Time Extrapolation of Dense and Frequent Phased-Array Weather Radar Observations , 2016 .