Cross-view image geographic positioning method and device based on attention efficiency network

The invention provides a cross-view image geographic positioning method and device based on an attention efficiency network. The method comprises the following steps: step 1, constructing an attention efficiency network, wherein the attention efficiency network is composed of an OfficientNetV2 netwo...

Full description

Saved in:
Bibliographic Details
Main Authors WANG YING, LU HUAYUAN, LIU YAN, YANG CHUNFANG, QI BAOJUN, XU JINGQIAN, LI JIANGSHAN, ZHU MA, LIU JINGYU
Format Patent
LanguageChinese
English
Published 23.06.2023
Subjects
Online AccessGet full text

Cover

Loading…
Abstract The invention provides a cross-view image geographic positioning method and device based on an attention efficiency network. The method comprises the following steps: step 1, constructing an attention efficiency network, wherein the attention efficiency network is composed of an OfficientNetV2 network and a space attention module; 2, constructing a double-branch network as a cross-view image geographic positioning network model; wherein each branch is composed of the attention efficiency network; 3, training the cross-view image geographic positioning network model based on a designed loss function; 4, performing polar coordinate transformation on the to-be-queried ground image to obtain a corresponding satellite panoramic image; and step 5, inputting the to-be-queried ground image and the satellite panoramic image into the trained cross-view-angle image geographic positioning network model to obtain a positioning result. 本发明提供一种基于注意力效率网络的跨视角图像地理定位方法及装置。该方法包括:步骤1:构建注意力效率网络,所述注意力效率网路由EfficientNetV2网络和空间注意力模块组成
AbstractList The invention provides a cross-view image geographic positioning method and device based on an attention efficiency network. The method comprises the following steps: step 1, constructing an attention efficiency network, wherein the attention efficiency network is composed of an OfficientNetV2 network and a space attention module; 2, constructing a double-branch network as a cross-view image geographic positioning network model; wherein each branch is composed of the attention efficiency network; 3, training the cross-view image geographic positioning network model based on a designed loss function; 4, performing polar coordinate transformation on the to-be-queried ground image to obtain a corresponding satellite panoramic image; and step 5, inputting the to-be-queried ground image and the satellite panoramic image into the trained cross-view-angle image geographic positioning network model to obtain a positioning result. 本发明提供一种基于注意力效率网络的跨视角图像地理定位方法及装置。该方法包括:步骤1:构建注意力效率网络,所述注意力效率网路由EfficientNetV2网络和空间注意力模块组成
Author LU HUAYUAN
LIU JINGYU
XU JINGQIAN
LI JIANGSHAN
QI BAOJUN
LIU YAN
YANG CHUNFANG
ZHU MA
WANG YING
Author_xml – fullname: WANG YING
– fullname: LU HUAYUAN
– fullname: LIU YAN
– fullname: YANG CHUNFANG
– fullname: QI BAOJUN
– fullname: XU JINGQIAN
– fullname: LI JIANGSHAN
– fullname: ZHU MA
– fullname: LIU JINGYU
BookMark eNqNyk0KwjAQBtAsdOHfHcYDCA0FXUtRXLlyX2LytR20M6EJLd5eBQ_g6m3e0sxEBQtTV4OmtBsZE3HvWlALbQcXO_YUNXFmFZaWeuROAzkJFDCyB91dQiAVcjlDvo_QNOwZ4l8kyJMOj7WZN-6ZsPm5Mtvz6VZddohaI0Xn8Zl1dbV2X9riUNhj-c95A_e_PwI
ContentType Patent
DBID EVB
DatabaseName esp@cenet
DatabaseTitleList
Database_xml – sequence: 1
  dbid: EVB
  name: esp@cenet
  url: http://worldwide.espacenet.com/singleLineSearch?locale=en_EP
  sourceTypes: Open Access Repository
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
Chemistry
Sciences
Physics
DocumentTitleAlternate 基于注意力效率网络的跨视角图像地理定位方法及装置
ExternalDocumentID CN116310701A
GroupedDBID EVB
ID FETCH-epo_espacenet_CN116310701A3
IEDL.DBID EVB
IngestDate Fri Jul 19 13:57:05 EDT 2024
IsOpenAccess true
IsPeerReviewed false
IsScholarly false
Language Chinese
English
LinkModel DirectLink
MergedId FETCHMERGED-epo_espacenet_CN116310701A3
Notes Application Number: CN202211489821
OpenAccessLink https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20230623&DB=EPODOC&CC=CN&NR=116310701A
ParticipantIDs epo_espacenet_CN116310701A
PublicationCentury 2000
PublicationDate 20230623
PublicationDateYYYYMMDD 2023-06-23
PublicationDate_xml – month: 06
  year: 2023
  text: 20230623
  day: 23
PublicationDecade 2020
PublicationYear 2023
RelatedCompanies INFORMATION ENGINEERING UNIVERSITY
RelatedCompanies_xml – name: INFORMATION ENGINEERING UNIVERSITY
Score 3.6098487
Snippet The invention provides a cross-view image geographic positioning method and device based on an attention efficiency network. The method comprises the following...
SourceID epo
SourceType Open Access Repository
SubjectTerms CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
GYROSCOPIC INSTRUMENTS
MEASURING
MEASURING DISTANCES, LEVELS OR BEARINGS
NAVIGATION
PHOTOGRAMMETRY OR VIDEOGRAMMETRY
PHYSICS
SURVEYING
TESTING
Title Cross-view image geographic positioning method and device based on attention efficiency network
URI https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20230623&DB=EPODOC&locale=&CC=CN&NR=116310701A
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV3dS8MwED_m_HzTqej8IIL0rbg16T4eirh0YwjrhkzZ20ibZJtgN2xF8K_3km3OF31NILQHd7-75He_A7jVWiCuU-rWmjJ2me8LV9RlzfWoZpoyKSrCNAr3olr3mT2O_FEBXte9MFYn9NOKI6JHJejvuY3Xi80lVmi5ldldPMOl-X1nGITOqjo2-bRHnbAVtAf9sM8dzgMeOdFTUMW8AyudSvVhC7ZNGm109tsvLdOVsvgNKZ1D2BngaWl-BIWvaQn2-XryWgn2eqsH7xLsWoZmkuHiyguzYxhzA22uudQnszcMCGSyHGU-nSVkzcJCSCLL8dBEpJJIZUICMaAlyTwlRlbTEh2JsiISpgOTpEtO-AncdNpD3nXxm8c_BhrzaPN79BSK6TxVZ0Ckx7RMlPYEjZnEOqeudZM1lJ80VKUu43Mo_31O-b_NCzgwxjaUKY9eQjF__1BXCM55fG2t-g0LFJYj
link.rule.ids 230,309,786,891,25594,76903
linkProvider European Patent Office
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV3fT8IwEL4g_sA3RYnir5qYvS3C2jF4WIx0EFQYxKDhbenWVjBxEJkx8a-3LSC-6GubNO0ld1-_9rs7gCspmcJ1jO1ag8c2cV1mM4_XbAdLIjHhrMJ0onAvrHWeyP3IHeXgdZULY-qEfpriiMqjEuXvmYnXs_UjVmC0lfPreKKGpjftoR9YS3as79MOtoKm3xr0gz61KPVpaIWPflXdOxTTqVRvN2DTU5TQUKXnps5Kmf2GlPYebA3Uamm2D7mvcREKdNV5rQg7veWHdxG2jUIzmavBpRfODyCiGtps_aiPJm8qIKCXRSvz8SRBKxWWgiS0aA-NWMoRFzokIA1aHE1TpMtqGqEjEqaIhM7AROlCE34Il-3WkHZstefox0ARDdfHwyXIp9NUHAHiDpE8EdJhOCZc8RxPygapCzepi4rH42Mo_71O-b_JCyh0hr1u1L0LH05gVxtey6ccfAr57P1DnCmgzuJzY-FvNeWZDQ
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%3Apatent&rft.title=Cross-view+image+geographic+positioning+method+and+device+based+on+attention+efficiency+network&rft.inventor=WANG+YING&rft.inventor=LU+HUAYUAN&rft.inventor=LIU+YAN&rft.inventor=YANG+CHUNFANG&rft.inventor=QI+BAOJUN&rft.inventor=XU+JINGQIAN&rft.inventor=LI+JIANGSHAN&rft.inventor=ZHU+MA&rft.inventor=LIU+JINGYU&rft.date=2023-06-23&rft.externalDBID=A&rft.externalDocID=CN116310701A