Fully Convolutional Siamese Networks for Change Detection

This paper presents three fully convolutional neural network architectures which perform change detection using a pair of coregistered images. Most notably, we propose two Siamese extensions of fully convolutional networks which use heuristics about the current problem to achieve the best results in...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings - International Conference on Image Processing S. 4063 - 4067
Hauptverfasser: Caye Daudt, Rodrigo, Le Saux, Bertr, Boulch, Alexandre
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.10.2018
Schlagworte:
ISSN:2381-8549
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract This paper presents three fully convolutional neural network architectures which perform change detection using a pair of coregistered images. Most notably, we propose two Siamese extensions of fully convolutional networks which use heuristics about the current problem to achieve the best results in our tests on two open change detection datasets, using both RGB and multispectral images. We show that our system is able to learn from scratch using annotated change detection images. Our architectures achieve better performance than previously proposed methods, while being at least 500 times faster than related systems. This work is a step towards efficient processing of data from large scale Earth observation systems such as Copernicus or Landsat.
AbstractList This paper presents three fully convolutional neural network architectures which perform change detection using a pair of coregistered images. Most notably, we propose two Siamese extensions of fully convolutional networks which use heuristics about the current problem to achieve the best results in our tests on two open change detection datasets, using both RGB and multispectral images. We show that our system is able to learn from scratch using annotated change detection images. Our architectures achieve better performance than previously proposed methods, while being at least 500 times faster than related systems. This work is a step towards efficient processing of data from large scale Earth observation systems such as Copernicus or Landsat.
Author Boulch, Alexandre
Caye Daudt, Rodrigo
Le Saux, Bertr
Author_xml – sequence: 1
  givenname: Rodrigo
  surname: Caye Daudt
  fullname: Caye Daudt, Rodrigo
  organization: DTIS, ONERA, Université Paris-Saclay, FR-91123, Palaiseau, France
– sequence: 2
  givenname: Bertr
  surname: Le Saux
  fullname: Le Saux, Bertr
  organization: DTIS, ONERA, Université Paris-Saclay, FR-91123, Palaiseau, France
– sequence: 3
  givenname: Alexandre
  surname: Boulch
  fullname: Boulch, Alexandre
  organization: DTIS, ONERA, Université Paris-Saclay, FR-91123, Palaiseau, France
BookMark eNotj91KwzAYQKMouM09gHiTF2jNl58m36VUp4Ohgno9kvSrVrtG2s6xt1dxV-fmcOBM2UmXOmLsAkQOIPBqWS6fcinA5U4bKIw8YnO0DrRFtKIAecwmUjnInNF4xqbD8CHEr69gwnCxbds9L1P3ndrt2KTOt_y58RsaiD_QuEv958Dr1PPy3XdvxG9opPjnnbPT2rcDzQ-csdfF7Ut5n60e75bl9SprpIYxq2XlTXQhYq20lqbAyhCGAIGUszagsDZi5QuKiNIa7xWYyiIU3oRoajVjl__dhojWX32z8f1-fThVP4xWSSs
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICIP.2018.8451652
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Xplore
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
EISBN 9781479970612
1479970611
EISSN 2381-8549
EndPage 4067
ExternalDocumentID 8451652
Genre orig-research
GroupedDBID 29O
6IE
6IF
6IH
6IK
6IL
6IM
6IN
AAJGR
AAWTH
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IPLJI
M43
OCL
RIE
RIL
RIO
RNS
ID FETCH-LOGICAL-i241t-f2da5c8bc9f3442569d5e9bb1be3877b9077c9da6ec99275aa315d7916a5bc5f3
IEDL.DBID RIE
ISICitedReferencesCount 1282
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000455181504033&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:50:51 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i241t-f2da5c8bc9f3442569d5e9bb1be3877b9077c9da6ec99275aa315d7916a5bc5f3
PageCount 5
ParticipantIDs ieee_primary_8451652
PublicationCentury 2000
PublicationDate 2018-10
PublicationDateYYYYMMDD 2018-10-01
PublicationDate_xml – month: 10
  year: 2018
  text: 2018-10
PublicationDecade 2010
PublicationTitle Proceedings - International Conference on Image Processing
PublicationTitleAbbrev ICIP
PublicationYear 2018
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0020131
Score 2.6280277
Snippet This paper presents three fully convolutional neural network architectures which perform change detection using a pair of coregistered images. Most notably, we...
SourceID ieee
SourceType Publisher
StartPage 4063
SubjectTerms Cats
Change detection
Change detection algorithms
Computer architecture
Earth
Earth observation
fully convolutional networks
Image analysis
Machine learning
supervised machine learning
Training
Title Fully Convolutional Siamese Networks for Change Detection
URI https://ieeexplore.ieee.org/document/8451652
WOSCitedRecordID wos000455181504033&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NSwMxEB1q8eCpait-k4NH0-5Xms25WuylLKjQW8nHBBZkK-224L832Q0rghdvYSEszGby3mze5AE8oLbcKqMpKm1optKMysTENFeZiRPpAEraxmyCL5f5aiWKHjx2vTCI2IjPcOyHzVm-2ei9_1U2yb2rLHMb7hHnvO3V6oorf29MOLWMIzFZzBaFF27l4zDpl3tKAx7zwf9eewqjny48UnT4cgY9rM5hEGgjCUm5G4LwZeQXcbMOYR3JD_JaevkrkmUr894RR05J20pAnrBuBFjVCN7nz2-zFxocEWjpkLamNjGS6VxpYdPMZdtUGIZCqVhhmnOuXKXLtTByilqIhDMp05gZ7iigZEozm15Av9pUeAlkqqx2eK2zTHvXaccybCwjGTHjKGPEoisY-kisP9tLL9YhCNd_P76BEx_sVuV2C_16u8c7ONaHutxt75sv9Q1MR5YB
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3LSgMxFL2UKuiqaiu-zcKlaeeRNJN1tbRYh4IVuit5QkGm0k4L_r3JTKgIbtyFQAjcZOacm5ybA_BglGVWaoWNVBoTmRIsEh3jTBIdJ8IBlLCV2QTL82w-59MGPO5rYYwxlfjMdH2zusvXK7X1R2W9zLvKUvfDPaCEJHFdrbVPr_zLMeHeMo54bzwYT710K-uGYb_8Uyr4GLb-N_EJdH7q8NB0jzCn0DDFGbQCcUThs9y0gftE8gu5Ubuwk8QHelt6AaxBeS303iBHT1FdTICeTFlJsIoOvA-fZ4MRDp4IeOmwtsQ20YKqTCpuUxcD2ueaGi5lLE2aMSZdrssU16JvFOcJo0KkMdXMkUBBpaI2PYdmsSrMBaC-tMohtiJEed9pxzNsLCIRUe1IY0SjS2j7SCw-62cvFiEIV39338PRaPY6WUzG-cs1HPvA15q3G2iW6625hUO1K5eb9V21at9MRZlI
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%3Abook&rft.genre=proceeding&rft.title=Proceedings+-+International+Conference+on+Image+Processing&rft.atitle=Fully+Convolutional+Siamese+Networks+for+Change+Detection&rft.au=Caye+Daudt%2C+Rodrigo&rft.au=Le+Saux%2C+Bertr&rft.au=Boulch%2C+Alexandre&rft.date=2018-10-01&rft.pub=IEEE&rft.eissn=2381-8549&rft.spage=4063&rft.epage=4067&rft_id=info:doi/10.1109%2FICIP.2018.8451652&rft.externalDocID=8451652