王桥2,
张学民3,
杨德贺2,,,
王志国4,
张乐5,
申旭辉2,
泽仁志玛2
1. 防灾科技学院, 河北 廊坊 062541
2. 应急管理部国家自然灾害防治研究院, 北京 100085
3. 中国地震局地震预测研究所, 北京 100036
4. 清华大学, 北京 100084
5. 中国电信股份有限公司研究院, 北京 102209
基金项目: 中央直属高校基本科研业务经费(ZY20180122), 中国科技部国家重点研发计划(2018YFC1503502和2018YFC1503806和2018YFC1503501), 廊坊科技局科学研究与发展计划自筹经费项目(2020011025), 中央直属高校基本科研业务经费(2020011025), APSCO earthquake project II: integrating satellite and ground-based observation for earthquake precursors and signatures(亚太地震二期项目: 地震前兆特征的星地一体化观测研究)项目资助, ISSI-BJ(2019IT-33)项目资助
详细信息
作者简介: 袁静, 女, 1981年出生, 副教授, 博士, 2019年毕业于清华大学, 主要从事大数据挖掘技术方面的研究.E-mail: yuanjing20110824@sina.com
通讯作者: 杨德贺, 男, 1985年出生, 助理研究员, 博士, 2014年毕业于中国矿业大学(北京), 主要从事大数据挖掘技术方面的研究.E-mail: ydhmmm@163.com
中图分类号: P352收稿日期:2020-07-10
修回日期:2020-12-07
上线日期:2021-05-10
Advances in the automatic detection algorithms for lightning whistlers recorded by electromagnetic satellite data
YUAN Jing1,,WANG Qiao2,
ZHANG XueMin3,
YANG DeHe2,,,
WANG ZhiGuo4,
ZHANG Le5,
SHEN XuHui2,
Zeren Zima2
1. Institute of Disaster Prevention, Langfang Hebei 062541, China
2. National Institute of Natural Hazards, Ministry of Emergency Management of China, Beijing 100085, China
3. Institute of Earthquake Forecasting, China Earthquake Administration, Beijing 100036, China
4. Tsinghua University, Beijing 100084, China
5. China Telecom Research Institute, Beijing 102209, China
More Information
Corresponding author: YANG DeHe,E-mail:ydhmmm@163.com
MSC: P352--> Received Date: 10 July 2020
Revised Date: 07 December 2020
Available Online: 10 May 2021
摘要
摘要:闪电哨声波作为探索空间物理环境的重要媒介，淹没在海量的电磁卫星数据中.近年来随着计算机视觉和深度学习等人工智能技术的发展，从电磁卫星的存档数据中自动检测闪电哨声波的算法取得了一定的效果.本文对近年来闪电哨声波智能检测算法的文献进行了整理和总结.首先，阐述闪电哨声波在电磁卫星数据中呈现的时频特征和类型；然后，介绍了闪电哨声波智能检测算法的流程并从闪电哨声波的特征提取、分类和定位三个方面对主要的智能检测算法进行归纳、总结和评述；其次，简述了闪电哨声波智能检测模型的评价指标；接着，在张衡一号（ZH-1）卫星的磁场数据上对三种典型的闪电哨声波智能检测算法进行复现，并对三种算法的优缺点进行了较深入的分析；最后，对基于电磁卫星的闪电哨声波智能检测的研究领域进行总结和展望.
关键词: 电磁卫星/
闪电哨声波/
智能检测算法/
张衡一号卫星
Abstract:Lightning whistlers, found frequently in electromagnetic satellite observation, are the important media to study the plasmasphere of the earth space. With the increasing number of data observed from electromagnetic satellites, a considerable amount of time and human efforts are needed to detect lightning whistlers from these tremendous data. However, in recent years, with the development of artificial intelligence (AI) technologies such as computer vision and deep learning, algorithms for lightning whistlers automatic detection in the time-frequency profile of the electromagnetic satellites data have been conducted. This study analyzes and summarizes the existing automatic detection algorithms. Firstly, we describe the time-frequency characteristics and types of lightning whistlers recorded by electromagnetic satellites. Secondly, we introduce the automatic detection technique, which is composed of three steps involving extracting the feature of the object (the lightning whistler or noise), choosing the correct class label for the feature, and providing the boundaries of each object. Then, we analyze and summarize the existing research methods from three aspects including lightning whistler feature extraction, classification and positioning. Thirdly, we briefly describe the metrics to evaluate the mathematical model for automatic detection of lightning whistler. Fourthly, we apply three typical algorithms on the search coil magnetometer data of ZH-1 satellite to detect the lightning whistler automatically and conduct a deep analysis on the results. Lastly, we present existing problems and future possible research directions.
Key words:Electromagnetic satellites/
Lightning whistler/
Automatic detection/
ZH-1 satellite
PDF全文下载地址:
http://www.geophy.cn/data/article/export-pdf?id=44f03d99-013f-463d-9692-7a0973c583aa
