DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Deep Learning based Defect Classification in X-ray Images of Weld Tubes (Masterarbeit)

Rajan, Sarvesh Sundar (2019) Deep Learning based Defect Classification in X-ray Images of Weld Tubes (Masterarbeit). DLR-Interner Bericht. DLR-IB-BT-AU-2019-209. Master's. Technische Universität Chemnitz. 94 S.

[img] PDF


n the scheme of Non Destructive Testing (NDT), defect detection is an important process. Traditional image processing techniques have successfully been used for defect recognition. Usage of machine learning techniques is still in the initial stages of development. Convolution Neural Networks (CNN) is widely used for object classification one such scenario is defect classification in weld tubes. With the advent of deep learning techniques such as transfer learning, we can transfer knowledge gained in one domain successfully into other. Pre-trained models successfully learn features from large scale datasets that can be used for in domains where there is lack of data. The aim of this work is to help a manual inspector in recognition of defects on the weld tubes. With a given set of images, we proceed by forming unique pipeline architecture for automatic defect ecognition. The research in this thesis focuses on extraction of welds using mage segmentation techniques, creating a dataset of defects and using it to on pre-trained Convolution Neural Networks of VGG16, VGG19, Inception V3 and ResNet101. We evaluate the models on different metrics finding the best suited model for the created dataset. Further a prototype sliding window solution is used to find defects over the extracted weld region. We also present the limitations of this approach and suggest what could be modified as part of the future scope.

Item URL in elib:https://elib.dlr.de/132639/
Document Type:Monograph (DLR-Interner Bericht, Master's)
Title:Deep Learning based Defect Classification in X-ray Images of Weld Tubes (Masterarbeit)
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:No
Number of Pages:94
Keywords:Convolution Neural Network, Image Segmentation, Transfer Learning, Template Matching
Institution:Technische Universität Chemnitz
Department:Dept. of Computer Science, Chair of Computer Engineering
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Aeronautics
HGF - Program Themes:fixed-wing aircraft
DLR - Research area:Aeronautics
DLR - Program:L AR - Aircraft Research
DLR - Research theme (Project):L - Structures and Materials (old)
Location: Augsburg
Institutes and Institutions:Institute of Structures and Design > Automation and Production Technology
Institute of Structures and Design
Deposited By: Schuster, Dr.-Ing. Alfons
Deposited On:17 Dec 2019 10:45
Last Modified:07 Aug 2023 17:57

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.