对抗型长短期记忆网络的雷达回波外推算法  

Radar echo extrapolation algorithm based on adversarial long short-term memory network

作　　者：方巍[1,2,3] 庞林 张飞鸿 盛胜利 Fang Wei;Pang Lin;Zhang Feihong;Sheng Victor S(School of Computer and Software,Nanjing University of Information Science and Technology,Nanjing 210044,China;Key Laboratory of Computer Aided Design and Graphics,Zhejiang University,Hangzhou 310058,China;Jiangsu Key Laboratory of Computer Information Processing Technology,Soochow University,Suzhou 215000,China;Texas Tech University,Lubbock TX79409,United States)

机构地区：[1]南京信息工程大学计算机与软件学院,南京210044 [2]浙江大学计算机辅助设计与图形学国家重点实验室,杭州310058 [3]苏州大学江苏省计算机信息处理技术重点实验室,苏州215000 [4]德克萨斯理工大学,拉伯克TX79409,美国

出　　处：《中国图象图形学报》2021年第5期1067-1080,共14页Journal of Image and Graphics

基　　金：国家自然科学基金项目(42075007);教育部天诚汇智创新促教科研创新基金项目(2018A03038);浙江大学计算机辅助设计与图形学国家重点实验室开放课题项目(A1916);苏州大学江苏省计算机信息处理技术重点实验室开放课题项目(KJS1935)。

摘　　要：目的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。方法首先,采用光流追踪局部特征的方式改进Conv LSTM,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 d BZ的条件下。结论由实验结果可知,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 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.MethodFirst,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,Conv LSTM is improved to form flow long short-term memory network(FLSTM),which is an optical flow optimization extrapolation

关 键 词：雷达回波外推 卷积长短期记忆网络(ConvLSTM) 深度卷积生成对抗网络(DCGAN) 光流法 序列到序列结构 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]