Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques
The non-destructive testing methods offer great benefit in detecting and classifying the weld defects. Among these, infrared (IR) thermography stands out in the inspection, characterization, and analysis of the defects from the camera image sequences, particularly with the recent advent of deep lear...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/13/6455 |
| _version_ | 1827655525403197440 |
|---|---|
| author | Domenico Buongiorno Michela Prunella Stefano Grossi Sardar Mehboob Hussain Alessandro Rennola Nicola Longo Giovanni Di Stefano Vitoantonio Bevilacqua Antonio Brunetti |
| author_facet | Domenico Buongiorno Michela Prunella Stefano Grossi Sardar Mehboob Hussain Alessandro Rennola Nicola Longo Giovanni Di Stefano Vitoantonio Bevilacqua Antonio Brunetti |
| author_sort | Domenico Buongiorno |
| collection | DOAJ |
| description | The non-destructive testing methods offer great benefit in detecting and classifying the weld defects. Among these, infrared (IR) thermography stands out in the inspection, characterization, and analysis of the defects from the camera image sequences, particularly with the recent advent of deep learning. However, in IR, the defect classification becomes a cumbersome task because of the exposure to the inconsistent and unbalanced heat source, which requires additional supervision. In light of this, authors present a fully automated system capable of detecting defective welds according to the electrical resistance properties in the inline mode. The welding process is captured by an IR camera that generates a video sequence. A set of features extracted by such video feeds supervised machine learning and deep learning algorithms in order to build an industrial diagnostic framework for weld defect detection. The experimental study validates the aptitude of a customized convolutional neural network architecture to classify the malfunctioning weld joints with mean accuracy of 99% and median f1 score of 73% across five-fold cross validation on our locally acquired real world dataset. The outcome encourages the integration of thermographic-based quality control frameworks in all applications where fast and accurate recognition and safety assurance are crucial industrial requirements across the production line. |
| first_indexed | 2024-03-09T22:08:54Z |
| format | Article |
| id | doaj.art-e32d97d3c2734dc887c572a8ce394517 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T22:08:54Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-e32d97d3c2734dc887c572a8ce3945172023-11-23T19:36:57ZengMDPI AGApplied Sciences2076-34172022-06-011213645510.3390/app12136455Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning TechniquesDomenico Buongiorno0Michela Prunella1Stefano Grossi2Sardar Mehboob Hussain3Alessandro Rennola4Nicola Longo5Giovanni Di Stefano6Vitoantonio Bevilacqua7Antonio Brunetti8Department of Electrical and Information Engineering (DEI), Polytechnic University of Bari, 70126 Bari, ItalyDepartment of Electrical and Information Engineering (DEI), Polytechnic University of Bari, 70126 Bari, ItalyComau S.p.A., Via Rivalta 30, 10095 Grugliasco, ItalyDepartment of Electrical and Information Engineering (DEI), Polytechnic University of Bari, 70126 Bari, ItalyComau S.p.A., Via Rivalta 30, 10095 Grugliasco, ItalyApulian Bioengineering s.r.l., Via delle Violette 14, 70026 Modugno, ItalyComau S.p.A., Via Rivalta 30, 10095 Grugliasco, ItalyDepartment of Electrical and Information Engineering (DEI), Polytechnic University of Bari, 70126 Bari, ItalyDepartment of Electrical and Information Engineering (DEI), Polytechnic University of Bari, 70126 Bari, ItalyThe non-destructive testing methods offer great benefit in detecting and classifying the weld defects. Among these, infrared (IR) thermography stands out in the inspection, characterization, and analysis of the defects from the camera image sequences, particularly with the recent advent of deep learning. However, in IR, the defect classification becomes a cumbersome task because of the exposure to the inconsistent and unbalanced heat source, which requires additional supervision. In light of this, authors present a fully automated system capable of detecting defective welds according to the electrical resistance properties in the inline mode. The welding process is captured by an IR camera that generates a video sequence. A set of features extracted by such video feeds supervised machine learning and deep learning algorithms in order to build an industrial diagnostic framework for weld defect detection. The experimental study validates the aptitude of a customized convolutional neural network architecture to classify the malfunctioning weld joints with mean accuracy of 99% and median f1 score of 73% across five-fold cross validation on our locally acquired real world dataset. The outcome encourages the integration of thermographic-based quality control frameworks in all applications where fast and accurate recognition and safety assurance are crucial industrial requirements across the production line.https://www.mdpi.com/2076-3417/12/13/6455industrial quality controlweld defect detectionintelligent diagnostic systemsthermographymachine learningdeep learning |
| spellingShingle | Domenico Buongiorno Michela Prunella Stefano Grossi Sardar Mehboob Hussain Alessandro Rennola Nicola Longo Giovanni Di Stefano Vitoantonio Bevilacqua Antonio Brunetti Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques Applied Sciences industrial quality control weld defect detection intelligent diagnostic systems thermography machine learning deep learning |
| title | Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques |
| title_full | Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques |
| title_fullStr | Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques |
| title_full_unstemmed | Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques |
| title_short | Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques |
| title_sort | inline defective laser weld identification by processing thermal image sequences with machine and deep learning techniques |
| topic | industrial quality control weld defect detection intelligent diagnostic systems thermography machine learning deep learning |
| url | https://www.mdpi.com/2076-3417/12/13/6455 |
| work_keys_str_mv | AT domenicobuongiorno inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT michelaprunella inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT stefanogrossi inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT sardarmehboobhussain inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT alessandrorennola inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT nicolalongo inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT giovannidistefano inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT vitoantoniobevilacqua inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques AT antoniobrunetti inlinedefectivelaserweldidentificationbyprocessingthermalimagesequenceswithmachineanddeeplearningtechniques |