浏览全部资源
扫码关注微信
Received：21 June 2020，
Revised：24 August 2020，
Accepted：01 September 2020，
Published：16 May 2021
移动端阅览
目的
2
雷达回波外推是进行短临降水预测的一种重要方法，相较于传统的数值天气预报方法能够实现更快、更准确的预测。基于卷积长短期记忆网络（convolutional long short-term memory network，ConvLSTM）的回波外推算法的效果优于其他的深度学习外推算法，但是忽略了普通卷积运算在面对局部变化特征时的局限性，并且在外推过程中将损失函数简单定义为均方误差（mean squared error，MSE），忽略了外推图像与原始图像的分布相似性，容易导致信息丢失。为解决以上不足，提出了一种基于对抗型光流长短期记忆网络（deep convolutional generative adversarial flow based long short-term memory network，DCF-LSTM）的回波外推算法。
方法
2
首先，采用光流追踪局部特征的方式改进ConvLSTM，突破了一般卷积核面对局部变化特征的限制。然后，以光流长短期记忆网络（flow based long short-term memory network，FLSTM）作为基本模块构建外推模型。最后，引入对抗网络，与外推模型组成端到端的博弈系统DCF-LSTM，两者交替训练实现外推图像分布向原图像分布的拟合。
结果
2
在4种不同的反射率强度下进行了消融研究，并与3种主流的气象业务算法进行了对比。实验结果表明，DCF-LSTM在所有评价指标中表现最优，尤其在反射率为35 dBZ的条件下。
结论
2
由实验结果可知，引入光流法能够使模型具有更好的抗畸变性，引入深度卷积生成对抗网络（deep convolutional generative adversarial network，DCGAN）判别模块能进一步增加结果的准确性。本文提出的DCF-LSTM回波外推算法相比于其他算法，雷达外推准确率获得了进一步提升。
Objective
2
Radar echo extrapolation is an important method for short-term precipitation prediction. It can achieve faster and more accurate predictions compared with traditional methods
such as numerical weather forecast and optical flow method. Among them
numerical weather forecasting requires complex and meticulous simulations of physical equations in the atmosphere and then uses observation data as input to predict future weather conditions. The optical flow method is currently the mainstream method used by the meteorological department
but it has two inherent flaws. On the one hand
only two adjacent frames can be used to estimate the optical flow; on the other hand
the radar echo sequence cannot be fully used for prediction. Nevertheless
the radar echo extrapolation method based on deep learning can take full advantage of spatiotemporal sequence data to achieve faster and more accurate prediction. In addition
the echo extrapolation algorithm based on convolutional long short-term memory network (ConvLSTM) has been proved to be effective in real applications
and the effect is superior to other deep learning extrapolation algorithms. However
it ignores the limitations of ordinary convolution operations in the face of locally changing features
and in the extrapolation process
the loss function is simply defined as mean square error (MSE)
ignoring the distribution similarity between the extrapolated image and the original image
which is easy to cause information loss. To solve the above problems
an improved echo extrapolation algorithm based on adversarial long short-term memory network (LSTM) is proposed.
Method
2
First
in view of the local-invariance limitations of the traditional convolution kernel
we borrowed the idea of the dense optical flow method and constructed a two-dimensional instantaneous velocity field for all pixels to extract the motion information of each part of the object. Based on this idea
ConvLSTM is improved to form flow long short-term memory network (FLSTM)
which is an optical flow optimization extrapolation algorithm. The algorithm uses optical flow to track local features
breaking through the limitation of local invariance of general convolution kernels. Then
according to the characteristics of radar sequence data (high-dimensional spatiotemporal data)
the convolutional layer is used to extract effective spatial features to reduce spatial redundancy in the encoder
and then deconvolution is used in the decoder to amplify the generated decoded features to the size of the original image to form an output sequence. The convolutional layer and FLSTM are cross-stacked in depth to encode the input spatiotemporal sequence data into a fixed-length vector. The deconvolution and FLSTM are cross-stacked to decode the output sequence from the encoded vector. Finally
in order to obtain extrapolated images with higher accuracy
an adversarial generation network is introduced
and an extrapolation model forms an end-to-end game system deep convolutional generative adversarial flow-based long short-term memory network (DCF-LSTM). In this system
the generation network is the extrapolation model that tends to be stable after pre-training. Then
the pre-trained generation network continue to be alternately trained with the discriminator to further fit the extrapolated image distribution to the real image distribution
thereby improving the accuracy of the extrapolated image.
Result
2
Experiments were carried out under four different reflectance intensities. The DCF-LSTM model is compared with the flow based ConvLSTM (FLSTM) and DC-LSTM
which is an optimized convolutional LSTM by integrating deep convolutional generative adversarial network (DCGAN)
and three mainstream meteorological business algorithms. The experimental results show that DCF-LSTM had the best performance under all intensity thresholds. Its probability of detection (POD) and critical success index (CSI) are higher than the other two methods
and it has the lowest false alarm rate (FAR) and mean square error (MSE)
especially when the reflectivity is 35 dBZ. The higher the value of POD and CSI
the better the model performance; the lower the FAR value
the more accurate the model. Compared with FLSTM
DCF-LSTM has a 0.012 higher POD
0.02 lower FAR
0.015 higher CSI
and 0.115 lower MSE. Compared with DC-LSTM
DCF-LSTM has 0.035 higher POD
0.03 lower FAR
0.034 higher CSI
and 0.274 lower MSE. In addition
compared with TrajGRU
ConvLSTM
and Flow methods
DCF-LSTM has a 0.018
0.047
and 0.099 higher POD; 0.015
0.036
and 0.083 higher CSI; and 0.012
0.034
and 0.087 lower FAR
respectively.
Conclusion
2
The experimental results show that the optical flow method can enable the model to learn the dynamic changes of local features in the radar sequence
breaking through the limitation of local invariance of the convolution operation and making the model more resistant to distortion. In addition
the introduction of DCGAN module for further game training prediction model can further increase the accuracy of the results. Compared with the three mainstream meteorological business algorithms
the DCF-LSTM echo extrapolation algorithm proposed in this study has further improved the accuracy of radar extrapolation.
Elsayed N, Maida A S and Bayoumi M. 2019. Reduced-gate convolutional LSTM architecture for next-frame video prediction using predictive coding//Proceedings of the International Joint Conference on Neural Networks. Budapest, Hungary: IEEE: 1-9[ DOI: 10.1109/IJCNN.2019.8852480 http://dx.doi.org/10.1109/IJCNN.2019.8852480 ]
Fang W, Zhang F H, Sheng V S and Ding Y W. 2018. A method for improving CNN-based image recognition using DCGAN. Computers, Materials and Continua, 57(1): 167-178[DOI:10.32604/cmc.2018.02356]
Farnebäck G. 2003. Two-frame motion estimation based on polynomial expansion//Proceedings of the Scandinavian Conference on Image Analysis. Halmstad, Sweden: Springer: 363-370[ DOI: 10.1007/3-540-45103-X_50 http://dx.doi.org/10.1007/3-540-45103-X_50 ]
Finn C, Goodfellow I and Levine S. 2016. Unsupervised learning for physical interaction through video prediction//Proceedings of the 30th International Conference on Neural Information Processing Systems. Barcelona, Spain: Curran Associates: 64-72
Gao F, Yang Y, Wang J, Sun J P, Yang E F and Zhou H Y. 2018. A deep convolutional generative adversarial networks (DCGANs)-based semi-supervised method for object recognition in synthetic aperture radar (SAR) images. Remote Sensing, 10(6): #846[DOI:10.3390/rs10060846]
Kim S, Hong S, Joh M S and Song S. 2017. DeepRain: ConvLSTM network for precipitation prediction using multichannel dar data[EB/OL]. [2020-06-21] . https://arxiv.org/pdf/1711.02316.pdf https://arxiv.org/pdf/1711.02316.pdf
Liang X D, Lin L, Shen X H, Feng J S, Yan S C and Xing E P. 2017. Interpretable structure-evolving LSTM//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE: 2175-2184[ DOI: 10.1109/cvpr.2017.234 http://dx.doi.org/10.1109/cvpr.2017.234 ]
Radford A, Metz L and Chintala S. 2015. Unsupervised representation learning with deep convolutional generative adversarial networks[EB/OL]. [2020-06-21] . https://arxiv.org/pdf/1511.06434.pdf https://arxiv.org/pdf/1511.06434.pdf
Shi E, Li Q, Gu D Q and Zhao Z M. 2018. A method of weather radar echo extrapolation based on convolutional neural networks//Proceedings of the International Conference on Multimedia Modeling. Bangkok, Thailand: Springer: 16-28[ DOI: 10.1007/978-3-319-73603-7_2 http://dx.doi.org/10.1007/978-3-319-73603-7_2 ]
Shi X J,Chen Z R, Wang H, Yeung D Y, Wong W K and Woo W C. 2015. Convolutional LSTM network: a machine learning approach for precipitation nowcasting//Proceedings of the 28th International Conference on Neural Information Processing Systems. Quebec, Canada: MIT Press: 802-810
Shi X J, Gao Z H, Lausen L, Wang H, Yeung D Y, Wong W K and Woo W C. 2017. Deep learning for precipitation nowcasting: a benchmark and a new model//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, USA: Curran Associates: 5617-5627
Srivastava N, Mansimov E and Salakhutdinov R. 2015. Unsupervised learning of video representations using LSTMs//Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille, France: ACM: 843-852
Sutskever I, Vinyals O and Le Q V. 2014. Sequence to sequence learning with neural networks//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, Canada: MIT Press: 3104-3112
Villegas R, Yang J M, Hong S, Lin X Y and Lee H. 2017. Decomposing motion and content for natural video sequence prediction[EB/OL]. [2020-06-21] . https://arxiv.org/pdf/1706.08033.pdf https://arxiv.org/pdf/1706.08033.pdf
Wang Y B, Gao Z F, Long M S, Wang J M and Yu P S. 2018. PredRNN++: towards a resolution of the deep-in-time dilemma in spatiotemporal predictive learning[EB/OL]. [2020-06-21] . https://arxiv.org/pdf/1804.06300.pdf https://arxiv.org/pdf/1804.06300.pdf
Wang Y B, Zhang J J, Zhu H Y, Long M S, Wang J M and Yu P S. 2019. Memory in memory: a predictive neural network for learning higher-order non-stationarity from spatiotemporal dynamics//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, USA: IEEE: 9154-9162[ DOI: 10.1109/CVPR.2019.00937 http://dx.doi.org/10.1109/CVPR.2019.00937 ]
Werbos P J. 1990. Backpropagation through time: what it does and how to do it. Proceedings of the IEEE, 78(10): 1550-1560[DOI:10.1109/5.58337]
Wu K, Liang W and Wang S Q. 2018. 3D convolutional neural network for regional precipitation nowcasting. Journal of Image and Signal Processing, 7(4): 200-212
吴昆, 梁伟, 王书强. 2018. 基于3D卷积神经网络的区域降雨量预报. 图像与信号处理, 7(4): 200-212[DOI:10.12677/JISP.2018.74023]
Xu Z, Du J, Wang J J, Jiang C X and Ren Y. 2019. Satellite image prediction relying on GAN and LSTM neural networks//Proceedings of 2019 IEEE International Conference on Communications. Shanghai, China: IEEE: 1-6[ DOI: 10.1109/ICC.2019.8761462 http://dx.doi.org/10.1109/ICC.2019.8761462 ]
Xu Z F, Xiong J and Ge W Z. 2006. Application of genetic algorithm in optimizing the Z-R parameter to radar rainfall estimation. Plateau Meteorology, 25(4): 710-715
徐枝芳, 熊军, 葛文忠. 2006. 使用遗传算法优化雷达测量降水Z-R关系. 高原气象, 25(4): 710-715[DOI:10.3321/j.issn:1000-0534.2006.04.020]
相关作者
相关机构
京公网安备11010802024621