Concept and methodology for automated data preprocessing of object recognition algorithm training

Preparing required data for training object recognition algorithms represents a complex and time-consuming process, that must be avoided especially in industrial environments. The work presented in this paper aims to overcome this challenge through on-line machine learning algorithms, as foundation...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Procedia CIRP Ročník 104; s. 1791 - 1794
Hlavní autoři: Giosan, Stefan, Matei, Raul, Albota, Vlad-Calin, Constantinescu, Carmen
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 2021
Témata:
ISSN:2212-8271, 2212-8271
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Preparing required data for training object recognition algorithms represents a complex and time-consuming process, that must be avoided especially in industrial environments. The work presented in this paper aims to overcome this challenge through on-line machine learning algorithms, as foundation for further developments and validation. The concept and the developed and validated methodology rely on point clouds resulted from the image processing using a depth camera. The geometry and coordinates of the objects are derived from the point clouds, fact that enables the automation of data preprocessing steps (e.g. manually take the pictures, labelling images), optimizing logistics and production activities.
AbstractList Preparing required data for training object recognition algorithms represents a complex and time-consuming process, that must be avoided especially in industrial environments. The work presented in this paper aims to overcome this challenge through on-line machine learning algorithms, as foundation for further developments and validation. The concept and the developed and validated methodology rely on point clouds resulted from the image processing using a depth camera. The geometry and coordinates of the objects are derived from the point clouds, fact that enables the automation of data preprocessing steps (e.g. manually take the pictures, labelling images), optimizing logistics and production activities.
Author Giosan, Stefan
Matei, Raul
Albota, Vlad-Calin
Constantinescu, Carmen
Author_xml – sequence: 1
  givenname: Stefan
  surname: Giosan
  fullname: Giosan, Stefan
  organization: Fraunhofer IAO, Nobelstraße 12, Stuttgart 70569, Germany
– sequence: 2
  givenname: Raul
  surname: Matei
  fullname: Matei, Raul
  organization: Fraunhofer IAO, Nobelstraße 12, Stuttgart 70569, Germany
– sequence: 3
  givenname: Vlad-Calin
  surname: Albota
  fullname: Albota, Vlad-Calin
  email: vlad.albota@iao.fraunhofer.de
  organization: Fraunhofer IAO, Nobelstraße 12, Stuttgart 70569, Germany
– sequence: 4
  givenname: Carmen
  surname: Constantinescu
  fullname: Constantinescu, Carmen
  organization: Fraunhofer IAO, Nobelstraße 12, Stuttgart 70569, Germany
BookMark eNp9kMtOwzAQRS0EEqX0D1j4BxJs5-Fkg4QqXlIlNrC2HD9SR40nsg1S_x5XZcGK2cws5t65c27QpQdvELqjpKSEtvdTuQRQLpSMMFpSWlaEXaAVY5QVHeP08s98jTYxTiQXr0lF2QrJLXhlloSl13g2aQ8aDjAesYWA5VeCWSajsZZJ4iWY0ykTo_MjBothmIxKOBgFo3fJgcfyMEJwaT_jFKTzefEWXVl5iGbz29fo8_npY_ta7N5f3raPu0JVhKaC8bojTat6VSvOu2boqe6G3jaKKV13vJdDw7pGKcKNrWrCq962jWXUtta2ZKjWqD77qgAxBmPFEtwsw1FQIk6kxCTOpMSJlKBUZFJZ9nCWmZzt25kgonImM9Eu_5WEBve_wQ8pZXdk
Cites_doi 10.1371/journal.pone.0238802
10.1109/ICRA.2017.7989165
10.1109/SIBGRAPI.2016.017
10.1109/MRA.2010.936956
ContentType Journal Article
Copyright 2021
Copyright_xml – notice: 2021
DBID 6I.
AAFTH
AAYXX
CITATION
DOI 10.1016/j.procir.2021.11.302
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 2212-8271
EndPage 1794
ExternalDocumentID 10_1016_j_procir_2021_11_302
S2212827121012002
GroupedDBID 0R~
0SF
4.4
457
6I.
AACTN
AAEDT
AAEDW
AAFTH
AAIKJ
AALRI
AAXUO
ABMAC
ACGFS
ADBBV
ADEZE
AEXQZ
AFTJW
AGHFR
AITUG
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
EBS
EJD
FDB
HZ~
IXB
KQ8
M41
M~E
NCXOZ
O-L
O9-
OK1
RIG
ROL
SSZ
AAYWO
AAYXX
ACVFH
ADCNI
ADVLN
AEUPX
AFPUW
AIGII
AKBMS
AKRWK
AKYEP
CITATION
ID FETCH-LOGICAL-c301t-2748056c9c4c7785b91d8b9f5c2cd4879ab5285cc07ef340739f65f21f6ff60b3
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001483972200302&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2212-8271
IngestDate Wed Nov 05 20:53:44 EST 2025
Wed May 17 00:08:32 EDT 2023
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords Image processing
Object recognition
Machine Learning
Logistics
Language English
License This is an open access article under the CC BY-NC-ND license.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c301t-2748056c9c4c7785b91d8b9f5c2cd4879ab5285cc07ef340739f65f21f6ff60b3
OpenAccessLink https://dx.doi.org/10.1016/j.procir.2021.11.302
PageCount 4
ParticipantIDs crossref_primary_10_1016_j_procir_2021_11_302
elsevier_sciencedirect_doi_10_1016_j_procir_2021_11_302
PublicationCentury 2000
PublicationDate 2021
2021-00-00
PublicationDateYYYYMMDD 2021-01-01
PublicationDate_xml – year: 2021
  text: 2021
PublicationDecade 2020
PublicationTitle Procedia CIRP
PublicationYear 2021
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Tensorflow Object-Detection API-How does the Fine-Tuning of a model works?
Lu, Wang, Hua, Chen (bib0005) 2020; 15
Mahler (bib0003) 2017; 13
(accessed Jan. 15, 2021).
Socher, Huval, Bhat, Manning, Ng (bib0002) 2012; 1
J.P.S. Do Monte Lima and V. Teichrieb, “An efficient global point cloud descriptor for object recognition and pose estimation,” in Proceedings-2016 29th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2016, 2017, pp. 56–63, doi: 10.1109/SIBGRAPI.2016.017.
J.P.S. do M. Lima and V. Teichrieb, “An Efficient Global Point Cloud Descriptor for Object Recognition and Pose Estimation,” in 2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), 2016, pp. 56–63, doi: 10.1109/SIBGRAPI.2016.017.
Cousins, Gerkey, Conley, Garage (bib00010) 2010; 17
Rusu, Bradski, Thibaux, Hsu (bib0006) 2010
A. Zeng et al., “Multi-view self-supervised deep learning for 6D pose estimation in the Amazon Picking Challenge,” in 2017 IEEE International Conference on Robotics and Automation (ICRA), 2017, pp. 1383–1386, doi: 10.1109/ICRA.2017.7989165.
Shreyas Dixit, Chadaga, Savalgimath, Ragavendra Rakshith, Naveen Kumar (bib0008) 2019; 8
EdjeElectronics / TensorFlow-Object-Detection-API-Tutorial-Train-Multiple-Objects-Windows-10.
10.1016/j.procir.2021.11.302_bib00011
Shreyas Dixit (10.1016/j.procir.2021.11.302_bib0008) 2019; 8
10.1016/j.procir.2021.11.302_bib0001
Lu (10.1016/j.procir.2021.11.302_bib0005) 2020; 15
Cousins (10.1016/j.procir.2021.11.302_bib00010) 2010; 17
Mahler (10.1016/j.procir.2021.11.302_bib0003) 2017; 13
Rusu (10.1016/j.procir.2021.11.302_bib0006) 2010
10.1016/j.procir.2021.11.302_bib0004
Socher (10.1016/j.procir.2021.11.302_bib0002) 2012; 1
10.1016/j.procir.2021.11.302_bib0009
10.1016/j.procir.2021.11.302_bib0007
References_xml – reference: (accessed Jan. 15, 2021).
– volume: 17
  start-page: 12
  year: 2010
  end-page: 14
  ident: bib00010
  article-title: “Sharing Software with ROS [ROS Topics],”
  publication-title: IEEE Robot. Autom. Mag.
– reference: J.P.S. Do Monte Lima and V. Teichrieb, “An efficient global point cloud descriptor for object recognition and pose estimation,” in Proceedings-2016 29th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2016, 2017, pp. 56–63, doi: 10.1109/SIBGRAPI.2016.017.
– volume: 8
  start-page: 824
  year: 2019
  end-page: 829
  ident: bib0008
  article-title: “Evaluation and evolution of object detection techniques YOLO and R-CNN,”
  publication-title: Int. J. Recent Technol. Eng.
– reference: “EdjeElectronics / TensorFlow-Object-Detection-API-Tutorial-Train-Multiple-Objects-Windows-10.”
– reference: J.P.S. do M. Lima and V. Teichrieb, “An Efficient Global Point Cloud Descriptor for Object Recognition and Pose Estimation,” in 2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), 2016, pp. 56–63, doi: 10.1109/SIBGRAPI.2016.017.
– volume: 1
  start-page: 656
  year: 2012
  end-page: 664
  ident: bib0002
  article-title: “Convolutional-recursive deep learning for 3D object classification,”
  publication-title: Advances in Neural Information Processing Systems
– volume: 13
  year: 2017
  ident: bib0003
  article-title: “Dex-Net 2.0: Deep learning to plan Robust grasps with synthetic point clouds and analytic grasp metrics,”
  publication-title: Robotics: Science and Systems
– year: 2010
  ident: bib0006
  publication-title: Fast 3D recognition and pose using the Viewpoint Feature Histogram
– reference: A. Zeng et al., “Multi-view self-supervised deep learning for 6D pose estimation in the Amazon Picking Challenge,” in 2017 IEEE International Conference on Robotics and Automation (ICRA), 2017, pp. 1383–1386, doi: 10.1109/ICRA.2017.7989165.
– reference: “Tensorflow Object-Detection API-How does the Fine-Tuning of a model works?”
– volume: 15
  year: 2020
  ident: bib0005
  article-title: “Automatic point cloud registration algorithm based on the feature histogram of local surface,”
  publication-title: PLoS One
– volume: 15
  issue: 9
  year: 2020
  ident: 10.1016/j.procir.2021.11.302_bib0005
  article-title: “Automatic point cloud registration algorithm based on the feature histogram of local surface,”
  publication-title: PLoS One
  doi: 10.1371/journal.pone.0238802
– ident: 10.1016/j.procir.2021.11.302_bib0001
– volume: 13
  year: 2017
  ident: 10.1016/j.procir.2021.11.302_bib0003
  article-title: “Dex-Net 2.0: Deep learning to plan Robust grasps with synthetic point clouds and analytic grasp metrics,”
  publication-title: Robotics: Science and Systems
– ident: 10.1016/j.procir.2021.11.302_bib0007
  doi: 10.1109/ICRA.2017.7989165
– ident: 10.1016/j.procir.2021.11.302_bib00011
  doi: 10.1109/SIBGRAPI.2016.017
– ident: 10.1016/j.procir.2021.11.302_bib0009
– volume: 1
  start-page: 656
  year: 2012
  ident: 10.1016/j.procir.2021.11.302_bib0002
  article-title: “Convolutional-recursive deep learning for 3D object classification,”
  publication-title: Advances in Neural Information Processing Systems
– volume: 17
  start-page: 12
  issue: 2
  year: 2010
  ident: 10.1016/j.procir.2021.11.302_bib00010
  article-title: “Sharing Software with ROS [ROS Topics],”
  publication-title: IEEE Robot. Autom. Mag.
  doi: 10.1109/MRA.2010.936956
– year: 2010
  ident: 10.1016/j.procir.2021.11.302_bib0006
  publication-title: Fast 3D recognition and pose using the Viewpoint Feature Histogram
– volume: 8
  start-page: 824
  issue: 2
  year: 2019
  ident: 10.1016/j.procir.2021.11.302_bib0008
  article-title: “Evaluation and evolution of object detection techniques YOLO and R-CNN,”
  publication-title: Int. J. Recent Technol. Eng.
– ident: 10.1016/j.procir.2021.11.302_bib0004
  doi: 10.1109/SIBGRAPI.2016.017
SSID ssj0000740312
Score 2.1297114
Snippet Preparing required data for training object recognition algorithms represents a complex and time-consuming process, that must be avoided especially in...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 1791
SubjectTerms Image processing
Logistics
Machine Learning
Object recognition
Title Concept and methodology for automated data preprocessing of object recognition algorithm training
URI https://dx.doi.org/10.1016/j.procir.2021.11.302
Volume 104
WOSCitedRecordID wos001483972200302&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
journalDatabaseRights – providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2212-8271
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000740312
  issn: 2212-8271
  databaseCode: M~E
  dateStart: 20120101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Nb9QwELVWhQMcEJ-iLSAfuFWuEm8S28dqxZcEFapK1VsU2zFstU2qbbbqiZ_S38pM7HgDrCp64BLtWoqzm3nyvBm_GRPyNku1hKgiYbLKHctsoZm2xjGZCmwIaG3ed7w5-SwOD-Xpqfo6mdwMtTBXC9E08vpaXfxXU8MYGBtLZ-9g7jgpDMBnMDpcwexw_SfDz3whYr8r4M-H9m2WernkqmuBogLJRGUoNgi48IUCQfvcakzL7EVVEUqVF9_b5bz7cR5Pkxjz2b7OACC2N_t0FCvFPszby5BX7Wq3ht8XeHIvHjgayREPFrr1DPZkUVmGtWLxhpknrx1K880q6FPOQ_FaSFX4yme_lnHwkExyf9rKfr1hbFiMk2y0nGLv1JFrxsVj47LvMxBn6HTMHLu88hR8wf404Ws3N2zt_-H9oiZxkLudlX6WEmeBAKmcYq_Se1xA6IXS0J_rHB7QL1gTcaMq_pehOrOXEP79czaznxGjOX5MHoVQhB54CD0hk7p5Sh6OGlQ-I1UAEwUw0RGYKICJRjBRBBP9DUy0ddSDiY7ARCOY6ACm5-Tb-3fHs48sHMrBDPiCjnGRSSDNRpnMCCFzrVIrtXK54cZC9KsqnXOZG5OI2k0z3Ah2Re546grnikRPX5Ctpm3ql4Qm0maJkcrxrMIuR0qqwpk81ba2GKlvEza8LLBJ33ulvM1K20QMb7QM_NHzwhJwcuudO3d80i55gN98Ku4V2eqWq_o1uW-uuvnl8k0Pkl8gV5a-
linkProvider ISSN International Centre
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=Concept+and+methodology+for+automated+data+preprocessing+of+object+recognition+algorithm+training&rft.jtitle=Procedia+CIRP&rft.au=Giosan%2C+Stefan&rft.au=Matei%2C+Raul&rft.au=Albota%2C+Vlad-Calin&rft.au=Constantinescu%2C+Carmen&rft.date=2021&rft.issn=2212-8271&rft.eissn=2212-8271&rft.volume=104&rft.spage=1791&rft.epage=1794&rft_id=info:doi/10.1016%2Fj.procir.2021.11.302&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_procir_2021_11_302
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2212-8271&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2212-8271&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2212-8271&client=summon