FR3127061B1 - Method for generating training images for supervised learning of a defect detection model of a manufactured object - Google Patents

Method for generating training images for supervised learning of a defect detection model of a manufactured object Download PDF

Info

Publication number
FR3127061B1
FR3127061B1 FR2109672A FR2109672A FR3127061B1 FR 3127061 B1 FR3127061 B1 FR 3127061B1 FR 2109672 A FR2109672 A FR 2109672A FR 2109672 A FR2109672 A FR 2109672A FR 3127061 B1 FR3127061 B1 FR 3127061B1
Authority
FR
France
Prior art keywords
manufactured object
model
alteration
defect detection
supervised learning
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
FR2109672A
Other languages
French (fr)
Other versions
FR3127061A1 (en
Inventor
Mohamed Slim Werda
Glênio Simião Ramalho
Aurèle Guillotin
Khalid Kouiss
Michael Decottignies
Thibault Poline
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Clermont Auvergne Inp
Centre National de la Recherche Scientifique CNRS
Faurecia Sieges dAutomobile SAS
Universite Clermont Auvergne
Original Assignee
Clermont Auvergne Inp
Centre National de la Recherche Scientifique CNRS
Faurecia Sieges dAutomobile SAS
Universite Clermont Auvergne
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Clermont Auvergne Inp, Centre National de la Recherche Scientifique CNRS, Faurecia Sieges dAutomobile SAS, Universite Clermont Auvergne filed Critical Clermont Auvergne Inp
Priority to FR2109672A priority Critical patent/FR3127061B1/en
Publication of FR3127061A1 publication Critical patent/FR3127061A1/en
Application granted granted Critical
Publication of FR3127061B1 publication Critical patent/FR3127061B1/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/001—Industrial image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30164—Workpiece; Machine component

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

La présente divulgation concerne un procédé (10) de génération d’images d’apprentissage pour l’apprentissage supervisé d’un modèle de détection de défaut d’un objet manufacturé, comportant des étapes de : (S11) détermination d’un modèle 3D représentant l’objet manufacturé dépourvu de défaut, ledit modèle 3D comportant un maillage 3D définissant des faces décrivant une enveloppe extérieure dudit objet manufacturé, et des textures associées respectivement aux différentes faces du maillage 3D,(S12) détermination d’une altération du modèle 3D de l’objet manufacturé, ladite altération étant représentative d’un défaut à détecter,(S14) génération d’une pluralité d’images d’apprentissage représentant l’objet manufacturé en faisant varier la présence ou l’absence d’altération dans le modèle 3D de l’objet manufacturé, (S15) annotation de chaque image d’apprentissage en fonction de la présence ou de l’absence d’altération dans le modèle 3D de l’objet manufacturé. Figure de l’abrégé : Figure 1The present disclosure relates to a method (10) for generating training images for supervised learning of a defect detection model of a manufactured object, comprising steps of: (S11) determining a 3D model representing the manufactured object devoid of defects, said 3D model comprising a 3D mesh defining faces describing an exterior envelope of said manufactured object, and textures associated respectively with the different faces of the 3D mesh, (S12) determination of an alteration of the 3D model of the manufactured object, said alteration being representative of a defect to be detected, (S14) generation of a plurality of learning images representing the manufactured object by varying the presence or absence of alteration in the 3D model of the manufactured object, (S15) annotation of each training image according to the presence or absence of alteration in the 3D model of the manufactured object. Abstract Figure: Figure 1

FR2109672A 2021-09-15 2021-09-15 Method for generating training images for supervised learning of a defect detection model of a manufactured object Active FR3127061B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
FR2109672A FR3127061B1 (en) 2021-09-15 2021-09-15 Method for generating training images for supervised learning of a defect detection model of a manufactured object

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
FR2109672A FR3127061B1 (en) 2021-09-15 2021-09-15 Method for generating training images for supervised learning of a defect detection model of a manufactured object
FR2109672 2021-09-15

Publications (2)

Publication Number Publication Date
FR3127061A1 FR3127061A1 (en) 2023-03-17
FR3127061B1 true FR3127061B1 (en) 2024-01-12

Family

ID=78212302

Family Applications (1)

Application Number Title Priority Date Filing Date
FR2109672A Active FR3127061B1 (en) 2021-09-15 2021-09-15 Method for generating training images for supervised learning of a defect detection model of a manufactured object

Country Status (1)

Country Link
FR (1) FR3127061B1 (en)

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9965901B2 (en) * 2015-11-19 2018-05-08 KLA—Tencor Corp. Generating simulated images from design information
US20210201474A1 (en) * 2018-06-29 2021-07-01 Photogauge, Inc. System and method for performing visual inspection using synthetically generated images

Also Published As

Publication number Publication date
FR3127061A1 (en) 2023-03-17

Similar Documents

Publication Publication Date Title
KR20230124713A (en) Fault detection methods, devices and systems
WO2021052159A1 (en) Adversarial transfer learning-based face beauty prediction method and device
US11468538B2 (en) Segmentation and prediction of low-level temporal plume patterns
KR20210115245A (en) Intelligent dam management system based on digital twin
US11255821B2 (en) Method for the graphical representation and data presentation of weld inspection results
EP2755051A3 (en) Image formation apparatus and method for nuclear imaging
JP2021018788A (en) Learning and detection method of neural network model for flame determination, system performing the same, and recording medium
Yu et al. Detecting gear surface defects using background‐weakening method and convolutional neural network
Naddaf-Sh et al. Defect detection and classification in welding using deep learning and digital radiography
FR3060169B1 (en) DETECTION OF NERFS IN A SERIES OF ECHOGRAPHIC IMAGES
JPWO2021216343A5 (en)
FR3127061B1 (en) Method for generating training images for supervised learning of a defect detection model of a manufactured object
Yuan et al. Underwater image fish recognition technology based on transfer learning and image enhancement
Qi et al. A CNN-based method for concreate crack detection in underwater environments
CN116930192A (en) High-precision copper pipe defect detection method and system
KR20230065188A (en) Apparatus for SaaS platform for automatic creation of descriptive AI-based dam facility defect detection, quantification, captioning, drawing, and condition evaluation
Sutcliffe et al. Automatic defect recognition of single-v welds using full matrix capture data, computer vision and multi-layer perceptron artificial neural networks
CN116152594A (en) Labeling method for constructing weld X-ray film artificial intelligence training set
CN113139507B (en) Automatic capturing method and system for drainage pipeline defect photos
MX2022005355A (en) Targeted application of deep learning to automated visual inspection equipment.
CN105427295A (en) Welding seam based image identification method and image identification system
MX2021005938A (en) Systems and methods for facilitating clone selection.
CN111047595A (en) Real-time sea wave segmentation method and device based on self-adaptive threshold frame difference method
Blubaugh et al. OpenCV and Perception
WO2019024094A1 (en) Mura detection method and mura detection system

Legal Events

Date Code Title Description
PLFP Fee payment

Year of fee payment: 2

PLSC Publication of the preliminary search report

Effective date: 20230317

PLFP Fee payment

Year of fee payment: 3