基于深度学习特征融合的天气雷达生物回波提取研究
作者:
作者单位:

1.南京信息工程大学江苏省大气环境与装备技术协同创新中心;2.中国气象局气象探测中心

基金项目:

国家重点研发计划(2018YFC1405703);国家自然科学基金(51875293)


Research on biological echo extraction of weather radar based on deep learning feature fusion
Author:
Affiliation:

1.Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment TechnologyCICAEET,Nanjing University of Information Science Technology,Nanjing;2.Meteorological Detection Center,China Meteorological Administration,Beijing

Fund Project:

National Key Research and Development Program of China;National Natural Science Foundation of China

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献
  • | |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    针对当前天气雷达提取生物回波的网络存在参数量大、效率低以及对大量样本的需求等问题,本文提出了一种改进的生物回波提取网络,命名为 MFF-PSPnet(Multi-Feature Fusion Pyramid Scene Parsing Network)。首先,该网络在 MobilenetV2 主干的基础上引入了注意力机制模块和边缘提取模块,然后对所有特征进行融合识别,本模型从生物回波主体提取和生物回波边缘细节刻画两个方面提升了网络的分割能力。最后,MFF-PSPnet 是一款轻量级网络,参数量相较于一般生物回波提取网络降低了 10.2%,相较于其他模型,该网络对样本容量的需求更低,更适应于小样本环境。通过本模型和其他模型在历史数据上的消融和对比实验,MFF-PSPnet 提取生物回波的准确率达到了 98.1%,IoU 达到了 94.8%,实验结果表明:在新一代中国天气雷达的历史数据中,本文改进的模型能够有效地提取出生物回波,并更好地适应小样本环境,可应用于移动端。这一成果为未来利用天气雷达进行生物习性研究和保护提供了有效的数据和技术支持。

    Abstract:

    The present network for extracting biological echoes from weather radar includes limitations such as a large number of parameters, low efficiency, and the requirement for a large number of samples. This paper offers an improved biological echo extraction network called MFF-PSPnet (Multi-Feature Fusion Pyramid Scene Parsing Network). This network introduces the attention mechanism module and edge extraction module based on the MobilenetV2 backbone, with all features fused for recognition. This modification increases the network's segmentation capabilities in two ways: biological echo main body extraction and biological echo edge detail characterisation. MFF-PSPnet is a lightweight network that uses 10.2% fewer parameters than the conventional biological echo extraction network. Compared to previous models, this network requires less sample capacity and is better suited to small sample situations. On the historical data set, MFF-PSPnet extracted biological echoes with an accuracy of 98.1% and an IoU of 94.8%. The experimental results show that in the historical data of the new generation of Chinese weather radar, the improved model in this work can efficiently extract biological echoes, adapt better to the limited sample environment, and be used to mobile devices. This finding gives useful data and technological support for the future application of weather radar for biological behavior study and protection.

    参考文献
    [1] Anisimova V I, Anisimov Y A. Characteristics of the Spring Migration of Some Bird Species on the Southeastern Coast of Lake Baikal[J]. Contemporary Problems of Ecology, 2023, 16(3): 295-302
    [2] Kranstauber B,Bouten W, van Gasteren H, et al.Ensemble predications are essential for accurate bird migration forecasts for conservation and flight safety[J]. Ecological solutions and evidence, 2022,3(3)
    [3] 曹垒, 孟凡娟, 赵青山. 基于前沿监测技术探讨 “大开发” 对鸟类迁徙及其栖息地的影响[J]. 中国科学院院刊, 2021, 36(4): 436-447 CAO Lei, MENG Fanjuan, ZHAO Qingshan. Understanding Effects of Large-scale Development on Bird Migration and Habitats Through Cutting Edge Avian Monitoring Techniques. Bulletin of Chinese Academy of Sciences, 2021, 36(4): 436-447
    [4] 柯杰龙,张羽,朱朋辉,等.基于改进Q-learning算法的输电线路拟声驱鸟策略研究[J].南 京信息工程大学学报(自然科学版),2022,14(5):579-586 KE Jielong, ZHANG Yu, ZHU Penghui, HUANG Chikun, WU Keting. Audio bird repelling strategy for transmission line based on improved Q-learning algorithm[J]. Journal of Nanjing University of Information Science & Technology, 2022,14(5):579-586
    [5] Dokter A M, Liechti F, Stark H, et al. Bird migration flight altitudes studied by a network of operational weather radars[J]. Journal of the Royal Society Interface, 2011, 8(54): 30-43
    [6] Gauthreaux Jr S A, Belser C G. Radar ornithology and biological conservation[J]. The Auk, 2003, 120(2): 266-277
    [7] Lack D. Migration across the North Sea studied by radar part 1. Survey through the year[J]. Ibis, 1959, 101(2): 209-234
    [8] Hu C, Kong S, Wang R, et al. Identification of migratory insects from their physical features using a decision-tree support vector machine and its application to radar entomology[J]. Scientific reports, 2018, 8(1): 5449
    [9] 朱轶明,马舒庆,杨玲,等.,2019.上海南汇WSR-88D双偏振天气雷达的生物回波识别与分析[J].气象与环境科学.42(03):118-128 Zhu, Y. M., Ma, S. Q., Yang, L., et al. 2019. Biological echo recognition and analysis of Shanghai Nanhui WSR-88D dual-polarization weather radar[J]. Meteorology and Environmental Science, 42(03), 118-128
    [10] Lin T Y, Winner K, Bernstein G, et al. MistNet: Measuring historical bird migration in the US using archived weather radar data and convolutional neural networks[J]. Methods in Ecology and Evolution, 2019, 10(11): 1908-1922
    [11] Shuaihang W,Cheng H,Kai C, et al. Animal Migration Patterns Extraction Based on Atrous-Gated CNN Deep Learning Model[J].Remote Sensing,2021,13(24)
    [12] Falconi M T, Marzano F S. Weather Radar Data Processing and Atmospheric Applications: An overview of tools for monitoring clouds and detecting wind shear[J]. IEEE Signal Processing Magazine, 2019, 36(4): 85-97
    [13] Gauthreaux S A , Belser C G .Displays of Bird Movements on the WSR-88D: Patterns and `Quantification*[J].Weather & Forecasting, 1998, 13(2):453-464.
    [14] Yang C, Guo H. A method of image semantic segmentation based on pspnet[J]. Mathematical Problems in Engineering, 2022, 2022
    [15] Yong L, Ma L, Sun D, et al. Application of MobileNetV2 to waste classification[J]. Plos one, 2023, 18(3): e0282336
    [16] 王李祺,张成,侯宇超,等.基于深度学习特征融合的遥感图像场景分类应用[J].南京信息工程大学学报(自然科学版),2023,15(3):346-356 WANG Liqi, ZHANG Cheng, HOU Yuchao, TAN Xiuhui, CHENG Rong, GAO Xiang, BAI Yanping. Remote sensing image scene classification based on deep learning feature fusion[J]. Journal of Nanjing University of Information Science & Technology, 2023,15(3):346-356
    [17] 刘忠洋,周杰,陆加新,等.基于注意力机制的多尺度特征融合图像去雨方法[J].南京信 息工程大学学报(自然科学版),2023,15(5):505-513 LIU Zhongyang, ZHOU Jie, LU Jiaxin, MIAO Zelin, SHAO Genfu, JIANG Kaiqiang, GAO Wei. Image rain removal via multi-scale feature fusion based on attention mechanism[J]. Journal of Nanjing University of Information Science & Technology, 2023,15(5):505-513
    [18] 陈栋,李明,李莉,陈淑文.基于双池化注意力机制的高光谱图像分类算法[J].南京信息 工程大学学报(自然科学版),2023,15(4):393-402 CHEN Dong, LI Ming, LI Li, CHEN Shuwen. Hyperspectral image classification based on double pool attention mechanism[J]. Journal of Nanjing University of Information Science & Technology, 2023,15(4):393-402
    [19] 胡凯,张颖超,钱蕾.最优小波尺度空间的图像边缘检测方法[J].南京信息工程大学学报(自然科学版),2011,(3):259-264 HU Kai, ZHANG Yingchao, QIAN Lei. Edge detection of optimal wavelet scale Space image[J]. Journal of Nanjing University of Information Science & Technology, 2011,(3):259-264
    [20] 王李祺,张成,侯宇超,等.基于深度学习特征融合的遥感图像场景分类应用[J].南京信息工程大学学报(自然科学版),2023,15(3):346-356 WANG Liqi, ZHANG Cheng, HOU Yuchao, TAN Xiuhui, CHENG Rong, GAO Xiang, BAI Yanping. Remote sensing image scene classification based on deep learning feature fusion[J]. Journal of Nanjing University of Information Science & Technology, 2023,15(3):346-356
    [21] Chen L C, Papandreou G, Kokkinos I, et al. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 40(4): 834-848
    相似文献
    引证文献
引用本文

陈奕伕,邓志良,吴东丽,刘云平,张静.基于深度学习特征融合的天气雷达生物回波提取研究[J].南京信息工程大学学报,,():

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-04-18
  • 最后修改日期:2024-08-12
  • 录用日期:2024-08-22

地址:江苏省南京市宁六路219号    邮编:210044

联系电话:025-58731025    E-mail:nxdxb@nuist.edu.cn

南京信息工程大学学报 ® 2025 版权所有  技术支持:北京勤云科技发展有限公司