改进YOLOv5的闪电哨声波轻量化自动检测模型
P352; 提出一种改进YOLOv5(You-Only-Look-Once version 5)检测模型YOLOv5-Upgraded.为了更快定位真实边框,该模型将损失函数CIoU(Complete IoU)替换为SIoU(Scylla IoU);同时为了避免网络训练过程中梯度消失、梯度爆炸以及神经元坏死等现象,将激活函数SiLU(Sigmoid-weighted Linear Unit)替换为具有更好梯度流的Mish;在主干网络中插入注意力(Coordinate Attention,CA)机制,帮助模型更精准地识别闪电哨声波,大大降低了漏检率.基于张衡一号感应磁力仪(Search Coil...
Saved in:
| Published in | 空间科学学报 Vol. 44; no. 3; pp. 458 - 473 |
|---|---|
| Main Authors | , , , , , , |
| Format | Journal Article |
| Language | Chinese |
| Published |
中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院国家空间科学中心 北京 100190
2024
应急管理部国家自然灾害防治研究院 北京 100085 |
| Subjects | |
| Online Access | Get full text |
| ISSN | 0254-6124 |
| DOI | 10.11728/cjss2024.03.2023-0067 |
Cover
| Abstract | P352; 提出一种改进YOLOv5(You-Only-Look-Once version 5)检测模型YOLOv5-Upgraded.为了更快定位真实边框,该模型将损失函数CIoU(Complete IoU)替换为SIoU(Scylla IoU);同时为了避免网络训练过程中梯度消失、梯度爆炸以及神经元坏死等现象,将激活函数SiLU(Sigmoid-weighted Linear Unit)替换为具有更好梯度流的Mish;在主干网络中插入注意力(Coordinate Attention,CA)机制,帮助模型更精准地识别闪电哨声波,大大降低了漏检率.基于张衡一号感应磁力仪(Search Coil Magnetometer,SCM)数据,以 2.4 s时间窗口截取数据,经带通滤波、短时傅里叶变换得到 1126张时频图数据集,再经图像增强操作扩充至 7882张,其中 7091张作为训练集,791张作为测试集.实验结果表明,基于改进YOLOv5的模型平均精度均值为 99.09%,召回率为 96.20%,与YOLOv5s相比,分别提升了 2.75%和 5.07%,与基于时频图的YOLOv3模型相比,平均精度均值和召回率则分别提高了 5.89%和 9.62%.基于智能语音的LSTM(Long Short Term Memory Networks)闪电哨声波识别模型大小为 82.89 MB,YOLOv5-Upgraded仅为 13.78 MB,约节省 83.38%的内存资源.研究表明改进后的轻量化模型大大降低了闪电哨声波的漏检现象,在测试集中取得了较好结果,并且其轻量化特征易于部署到卫星设备,极大地提高了星载识别的可能性. |
|---|---|
| AbstractList | P352; 提出一种改进YOLOv5(You-Only-Look-Once version 5)检测模型YOLOv5-Upgraded.为了更快定位真实边框,该模型将损失函数CIoU(Complete IoU)替换为SIoU(Scylla IoU);同时为了避免网络训练过程中梯度消失、梯度爆炸以及神经元坏死等现象,将激活函数SiLU(Sigmoid-weighted Linear Unit)替换为具有更好梯度流的Mish;在主干网络中插入注意力(Coordinate Attention,CA)机制,帮助模型更精准地识别闪电哨声波,大大降低了漏检率.基于张衡一号感应磁力仪(Search Coil Magnetometer,SCM)数据,以 2.4 s时间窗口截取数据,经带通滤波、短时傅里叶变换得到 1126张时频图数据集,再经图像增强操作扩充至 7882张,其中 7091张作为训练集,791张作为测试集.实验结果表明,基于改进YOLOv5的模型平均精度均值为 99.09%,召回率为 96.20%,与YOLOv5s相比,分别提升了 2.75%和 5.07%,与基于时频图的YOLOv3模型相比,平均精度均值和召回率则分别提高了 5.89%和 9.62%.基于智能语音的LSTM(Long Short Term Memory Networks)闪电哨声波识别模型大小为 82.89 MB,YOLOv5-Upgraded仅为 13.78 MB,约节省 83.38%的内存资源.研究表明改进后的轻量化模型大大降低了闪电哨声波的漏检现象,在测试集中取得了较好结果,并且其轻量化特征易于部署到卫星设备,极大地提高了星载识别的可能性. |
| Abstract_FL | This project proposes an improved YOLOv5 detection algorithm YOLOv5 Upgraded.To address this issue,the study proposes an improved YOLOv5 detection algorithm called YOLOv5-Up-graded.The model takes into account the vector angle between the predicted edge and the real edge,The model replaces the loss function CIoU(Complete IoU)with SIoU(Scylla IoU);at the same time,in or-der to avoid phenomena such as gradient disappearance,gradient explosion,and neuron necrosis during network training,the activation function SiLU(Sigmoid-weighted Linear Unit)is replaced with Mish with better gradient flow;The CA attention mechanism is inserted into the backbone network to help the model identify the Lightning whistler waves more accurately and greatly reduce the missed detec-tion rate.The study is based on the VLF-band data of CSES Satellite SCM with 2.4 seconds time win-dow to intercept data,and 1126 time-frequency map data sets are obtained by band-pass filtering and short-time Fourier transform,and then expanded to 7882 images by image enhancement operations,of which 7091 are used as training set and 791 are used as test set.Experimentally,the average mean accu-racy(mAP)of the improved YOLOv5-based model is 99.09%and the Recall is 96.20%,which are im-proved by 2.75%and 5.07%compared with the plain YOLOv5s,and 5.89%and 9.62%compared with the time-frequency map-based YOLOv3 model.The size of LSTM based on the speech processing tech-nology lightning whistler waves recognition model is 82.89MB,while the YOLOv5-Upgraded model is on-ly 13.78 MB,saving about 83.38%of memory resources.It is shown that the model greatly reduces the leakage problem of Lightning whistler waves,achieves better results in test set,and its lightweight fea-tures are easy to deploy to satellite devices,which greatly improves the possibility of satellite recogni-tion. |
| Author | 冉子霖 孙晓英 泽仁志玛 路超 吕访贤 杨德贺 申旭辉 |
| AuthorAffiliation | 应急管理部国家自然灾害防治研究院 北京 100085;中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院国家空间科学中心 北京 100190 |
| AuthorAffiliation_xml | – name: 应急管理部国家自然灾害防治研究院 北京 100085;中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院国家空间科学中心 北京 100190 |
| Author_FL | LU Chao SHEN Xuhui YANG Dehe Lü Fangxian ZEREN Zhima SUN Xiaoying RAN Zilin |
| Author_FL_xml | – sequence: 1 fullname: LU Chao – sequence: 2 fullname: ZEREN Zhima – sequence: 3 fullname: YANG Dehe – sequence: 4 fullname: SUN Xiaoying – sequence: 5 fullname: Lü Fangxian – sequence: 6 fullname: RAN Zilin – sequence: 7 fullname: SHEN Xuhui |
| Author_xml | – sequence: 1 fullname: 路超 – sequence: 2 fullname: 泽仁志玛 – sequence: 3 fullname: 杨德贺 – sequence: 4 fullname: 孙晓英 – sequence: 5 fullname: 吕访贤 – sequence: 6 fullname: 冉子霖 – sequence: 7 fullname: 申旭辉 |
| BookMark | eNotjz9Lw0AcQG-oYK39Cq5Oib_7m7tRiv8gkEUHp3Jp7sRUUvBQOzoIBSs4WBQUBO0SOgnSQUT8NEnTb2FFp7e9x1tBtayXGYTWMPgYB0RudFLnCBDmA_UXpB6ACGqoDoQzT2DCllHTueMYgBBBmaR1pMrRR_X9dBiF0TmfPV7NHyaz0bS4y4vxW_n-Wn19zge3xc19NZgU13k5viynwzJ_KZ6Hq2jJ6hNnmv9soIPtrf3WrhdGO3utzdBzGIj0LMOJIrLDBcXWcI4FxlbF2nKjwQTAFUk4VlYYHRvBE2GtjaVioDGhJrG0gdb_vBc6szo7aqe9s9NsUWx3026_H_8OAwWQ9Ae1r14M |
| ClassificationCodes | P352 |
| ContentType | Journal Article |
| Copyright | Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| Copyright_xml | – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| DBID | 2B. 4A8 92I 93N PSX TCJ |
| DOI | 10.11728/cjss2024.03.2023-0067 |
| DatabaseName | Wanfang Data Journals - Hong Kong WANFANG Data Centre Wanfang Data Journals 万方数据期刊 - 香港版 China Online Journals (COJ) China Online Journals (COJ) |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Astronomy & Astrophysics |
| DocumentTitle_FL | Lightweight Automatic Detection Model for Lightning Whistle Waves Based on Improved YOLOv5 |
| EndPage | 473 |
| ExternalDocumentID | kjkxxb202403008 |
| GrantInformation_xml | – fundername: 国家自然科学基金 funderid: (41874174) |
| GroupedDBID | -01 2B. 4A8 5XA 5XB 92H 92I 93N ABJNI ACGFS ALMA_UNASSIGNED_HOLDINGS CCEZO CCVFK CW9 GROUPED_DOAJ PSX TCJ TGT U1G U5K |
| ID | FETCH-LOGICAL-s1028-f41d928c5631fe551611f9baf5ea0e70592d519f6eabe65d6fffb8940a123edf3 |
| ISSN | 0254-6124 |
| IngestDate | Thu May 29 03:55:02 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 3 |
| Keywords | 张衡一号卫星 CSES Lightning whistler waves YOLOv5 Lightweight 自动检测 闪电哨声波 轻量化 Automatic detection |
| Language | Chinese |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-s1028-f41d928c5631fe551611f9baf5ea0e70592d519f6eabe65d6fffb8940a123edf3 |
| PageCount | 16 |
| ParticipantIDs | wanfang_journals_kjkxxb202403008 |
| PublicationCentury | 2000 |
| PublicationDate | 2024 |
| PublicationDateYYYYMMDD | 2024-01-01 |
| PublicationDate_xml | – year: 2024 text: 2024 |
| PublicationDecade | 2020 |
| PublicationTitle | 空间科学学报 |
| PublicationTitle_FL | Chinese Journal of Space Science |
| PublicationYear | 2024 |
| Publisher | 中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院国家空间科学中心 北京 100190 应急管理部国家自然灾害防治研究院 北京 100085 |
| Publisher_xml | – name: 应急管理部国家自然灾害防治研究院 北京 100085 – name: 中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院大学应急管理科学与工程学院 北京 100049%中国科学院国家空间科学中心 北京 100190 |
| SSID | ssib002263483 ssib051375011 ssj0039410 ssib000862484 |
| Score | 2.3395922 |
| Snippet | P352; 提出一种改进YOLOv5(You-Only-Look-Once version 5)检测模型YOLOv5-Upgraded.为了更快定位真实边框,该模型将损失函数CIoU(Complete IoU)替换为SIoU(Scylla... |
| SourceID | wanfang |
| SourceType | Aggregation Database |
| StartPage | 458 |
| Title | 改进YOLOv5的闪电哨声波轻量化自动检测模型 |
| URI | https://d.wanfangdata.com.cn/periodical/kjkxxb202403008 |
| Volume | 44 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw5R3LahRBcAjx4kV8Ep_swfYSRmemH9N97M7OEkTNJYF4CjO700pCNuAmEnPyIASM4MGgoCBoLktOguQgIn7NbpJ_8GBVz2x2MBGjVy-7TXVVdT2a7qqefnjedWqzjOcq9XPeinwYJSPcBGB9mBu5UDDFpxGeHb57T0zOsNuzfHZk9Edl19LKcnazuXbkuZJ_8SrAwK94SvYvPHvAFABQBv_CL3gYfo_lY5IIohgxiiSSmAZR5v7UnanHnCQxUZpIRhJFVEy0dhDAhCpOFCVaYkFTYgJkYgASOSZ1YgxSyZjIBuLICaIEVsmCD0C0IxdILgtyABoHkUSHrokEINXAFwXQihhdimSYg4BUDl_XiRaVgnCt8EF3cKIBesMVoLJS46QHuROGokvHDm0RYwMyAaNUcVW91N0kyBHZAZ0eojgRlHK4Ck2Fqhtiwur6SDRcGXXtSiRCtY2T5LBiCpmhiQHCsLaqMxgFfIM6B6CzI0-cHYFJQKQ4iqFrFLRD8nhA5SQ91Oj4wI2xIwO7ToyHLmkjEf_fxC9xOMpoxPF6ZaUJ6FiS_kaisHiRtpxWI858iOtZNQYo7iAtxzpamdBZ8bBAGRuy4tmdw2FHHOFZmuZ8p4NdEK9Nhn_qYyw4DLQOtr8uzC-srmaICVMs3lRwIorjkFcWhAaLDUxWg29B2TAZ4CGFaD882NBGFQuLFeVSv_I-BJTt1pGSuXOKbZu2H1RC6unT3qkyF67pYmA7442sPTzrjekOfp1bWnxSu1Fz5WLxtXPOU_3NL3vf3xVj3O7bZ_tvtnc3d3qvur2tT_3PH_e-fd1ff9l78Xpvfbv3vNvfetrf2eh3P_Teb5z3ZhrJ9MSkXz794ncw4_EtC1sqkk0uaGhz_JgfhlZlqeV5GuQx5IRRC3JPK_I0ywVvCWttJhULUojE85alF7zR9lI7H_NqzSBo5ixqhjHNmYUMpUVpbvHiRKV4loqLXq20wlw5tHfmfnHQpT-jXPZOYrlYnL3ijS4_WsmvQrqynF1zXv0JA3rd2Q |
| linkProvider | Directory of Open Access Journals |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E6%94%B9%E8%BF%9BYOLOv5%E7%9A%84%E9%97%AA%E7%94%B5%E5%93%A8%E5%A3%B0%E6%B3%A2%E8%BD%BB%E9%87%8F%E5%8C%96%E8%87%AA%E5%8A%A8%E6%A3%80%E6%B5%8B%E6%A8%A1%E5%9E%8B&rft.jtitle=%E7%A9%BA%E9%97%B4%E7%A7%91%E5%AD%A6%E5%AD%A6%E6%8A%A5&rft.au=%E8%B7%AF%E8%B6%85&rft.au=%E6%B3%BD%E4%BB%81%E5%BF%97%E7%8E%9B&rft.au=%E6%9D%A8%E5%BE%B7%E8%B4%BA&rft.au=%E5%AD%99%E6%99%93%E8%8B%B1&rft.date=2024&rft.pub=%E4%B8%AD%E5%9B%BD%E7%A7%91%E5%AD%A6%E9%99%A2%E5%A4%A7%E5%AD%A6%E5%BA%94%E6%80%A5%E7%AE%A1%E7%90%86%E7%A7%91%E5%AD%A6%E4%B8%8E%E5%B7%A5%E7%A8%8B%E5%AD%A6%E9%99%A2+%E5%8C%97%E4%BA%AC+100049%25%E4%B8%AD%E5%9B%BD%E7%A7%91%E5%AD%A6%E9%99%A2%E5%A4%A7%E5%AD%A6%E5%BA%94%E6%80%A5%E7%AE%A1%E7%90%86%E7%A7%91%E5%AD%A6%E4%B8%8E%E5%B7%A5%E7%A8%8B%E5%AD%A6%E9%99%A2+%E5%8C%97%E4%BA%AC+100049%25%E4%B8%AD%E5%9B%BD%E7%A7%91%E5%AD%A6%E9%99%A2%E5%9B%BD%E5%AE%B6%E7%A9%BA%E9%97%B4%E7%A7%91%E5%AD%A6%E4%B8%AD%E5%BF%83+%E5%8C%97%E4%BA%AC+100190&rft.issn=0254-6124&rft.volume=44&rft.issue=3&rft.spage=458&rft.epage=473&rft_id=info:doi/10.11728%2Fcjss2024.03.2023-0067&rft.externalDocID=kjkxxb202403008 |
| thumbnail_s | http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fkjkxxb%2Fkjkxxb.jpg |