Rainfall cloud segmentation method in Tibet based on DeepLab v3
ZHANG Yonghong1,2, LIU Hao2, TIAN Wei3, WANG Jiangeng4
1. Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing Jiangsu 210044, China; 2. School of Automation, Nanjing University of Information Science and Technology, Nanjing Jiangsu 210044, China; 3. School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing Jiangsu 210044, China; 4. School of Atmospheric Physics, Nanjing University of Information Science and Technology, Nanjing Jiangsu 210044, China
Abstract:Concerning the problem that the numerical prediction method is complex in modeling, the radar echo extrapolation method is easy to generate cumulative error and the model parameters are difficult to set in plateau area, a method for segmenting rainfall clouds in Tibet was proposed based on the improved DeepLab v3. Firstly, the convolutional layers and residual modules in the coding network were used for down-sampling. Then, the multi-scale sampling module was constructed by using the dilated convolution, and the attention mechanism module was added to extract deep high-dimensional features. Finally, the deonvolutional layers in the decoding network were used to restore the feature map resolution. The proposed method was compared with Google semantic segmentation network DeepLab v3 and other models on the validation set. The experimental results show that the method has better segmentation performance and generalization ability, has the rainfall cloud segmented more accurately, and the Mean intersection over union (Miou) reached 0.95, which is 15.54 percentage points higher than that of the original DeepLab v3. On small targets and unbalanced datasets, rainfall clouds can be segmented more accurately by this method, so that the proposed method can provide a reference for the rain cloud monitoring and early warning.
张永宏, 刘昊, 田伟, 王剑庚. 基于DeepLab v3的西藏地区降雨云团分割方法[J]. 计算机应用, 2020, 40(9): 2781-2788.
ZHANG Yonghong, LIU Hao, TIAN Wei, WANG Jiangeng. Rainfall cloud segmentation method in Tibet based on DeepLab v3. Journal of Computer Applications, 2020, 40(9): 2781-2788.
[1] 陈森发,张文红,张建坤,等. 短期降雨预测的随机微分模型[J]. 系统工程学报, 2004, 19(3):239-243.(CHEN S F, ZHANG W H, ZHANG J K, et al. Short-term rainfall prediction of stochastic differential models[J]. Journal of Systems Engineering, 2004, 19(3):239-243.) [2] 赵欣,邹良超,倪林. 基于有序聚类的模糊加权马尔可夫模型在降雨预测中的应用[J]. 江西农业学报, 2009, 21(2):110-113. (ZHAO X, ZOU L C, NI L. Application of fuzzy weighted Markov model based on ordered clustering to rainfall prediction[J]. Acta Agriculture Jiangxi, 2009, 21(2):110-113.) [3] 朱平,李生辰,肖建设,等. 天气雷达回波外推技术应用研究[J]. 气象, 2008, 34(7):3-9.(ZHU P, LI S C, XIAO J S,et al. Study on extrapolation technique of weather radar echo and its application to nowcasting[J]. Meteorological Monthly, 2008, 34(7):3-9.) [4] 卢乃锰,方翔,刘健,等. 气象卫星的云观测[J]. 气象, 2017, 43(3):257-267.(LU N M, FANG X, LIU J, et al. Understanding clouds by meteorological satellites[J]. Meteorological Monthly, 2017, 43(3):257-267.) [5] 李森,刘健文,刘玉玲. 基于FY2D静止卫星云图的强对流云团识别[J]. 气象水文海洋仪器, 2010, 27(2):72-78.(LI S, LIU J W, LIU Y L. Detection of strong convective clouds based on FY2D geostationary satellite[J]. Meteorological, Hydrological and Marine Instruments, 2010, 27(2):72-78.) [6] 赵文化,单海滨. 基于红外窗区与水汽通道对流云团识别方法研究[J]. 气象, 2018, 44(6):814-824.(ZHAO W H, SHAN H B. Study of convective cloud identification based on H2O/IRW observation[J]. Meteorological Monthly, 2018, 44(6):814-824.) [7] 周晓丽,杨昌军. 基于FY-2D的新疆区域强对流云识别[J]. 沙漠与绿洲气象, 2017, 11(2):82-87.(ZHOU X L, YANG C J. Identification of strong convective cloud in Xinjiang using FY-2D stationary meteorological satellite data[J]. Desert and Oasis Meteorology, 2017, 11(2):82-87.) [8] 吴晓京,朱小祥,毛紫阳,等. 风云二号气象卫星红外观测在云团降水监测中的应用[J]. 海洋气象学报, 2019, 39(3):1-10.(WU X J, ZHU X X, MAO Z Y, et al. Algorithm design of convective precipitation monitoring and early warning service using FY-2 infrared data[J]. Journal of Marine Meteorology, 2019, 39(3):1-10.) [9] WOLF D E, HALL D J, ENDLICH R M. Experiments in automatic cloud tracking using SMS-GOES data[J]. Journal of Applied Meteorology, 1977, 16(11):1219-1230. [10] MAHOVIĆ N S, ZEINER B. Application of meteosat SEVIRI channel difference 0.6μm-1.6μm in convective cells detection[J]. Atmospheric Research, 2009, 93(1/2/3):270-276. [11] 许玥,冯梦如,皮家甜,等. 基于深度学习模型的遥感图像分割方法[J]. 计算机应用, 2019, 39(10):2905-2914.(XU Y, FENG M R, PI J T, et al. Remote sensing image segmentation method based on deep learning model[J]. Journal of Computer Applications, 2019, 39(10):2905-2914.) [12] 周明非,汪西莉. 结合纹理去除的遥感图像分割[J]. 计算机应用, 2017, 37(11):3162-3167.(ZHOU M F, WANG X L. Remote sensing image segmentation with texture removal[J]. Journal of Computer Applications, 2017, 37(11):3162-3167.) [13] 夏勇,赵荣椿. 基于形态学多重分形的遥感图像多尺度分割[J]. 计算机应用, 2006, 26(9):2071-2073.(XIA Y, ZHAO R C. Multiscale segmentation of remote sensing images based on multifractal exponents[J]. Journal of Computer Applications, 2006, 26(9):2071-2073.) [14] 陈天华,郑司群,于峻川. 采用改进DeepLab网络的遥感图像分割[J]. 测控技术, 2018, 37(11):34-39.(CHEN T H, ZHENG S Q, YU J C. Remote sensing image segmentation based on improved DeepLab network[J]. Measurement and Control Technology, 2018, 37(11):34-39.) [15] 宋小娜,芮挺,王新晴. 结合语义边界信息的道路环境语义分割方法[J]. 计算机应用, 2019, 39(9):2505-2510.(SONG X N, RUI T, WANG X Q. Semantic segmentation method of road environment combined semantic boundary information[J]. Journal of Computer Applications, 2019, 39(9):2505-2510.) [16] 吴宗胜,傅卫平,韩改宁. 基于深度卷积神经网络的道路场景理解[J]. 计算机工程与应用, 2017, 53(22):8-15.(WU Z S, FU W P, HAN G N. Road scene understanding based on deep convolutional neural network[J]. Computer Engineering and Applications, 2017, 53(22):8-15.) [17] WANG P, CHEN P, YUAN Y, et al. Understanding convolution for semantic segmentation[C]//Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision. Piscataway:IEEE, 2018:1451-1460. [18] 郑菲,孟朝晖,郭闯世. 基于反卷积特征学习的图像语义分割算法[J]. 计算机系统应用, 2019, 28(1):149-157.(ZHENG F, MENG Z H, GUO C S. Image semantic segmentation algorithm based on deconvolution feature learning[J]. Computer System and Applications, 2019, 28(1):149-157.)