Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging Systems
The industrial application of artificial intelligence (AI) has witnessed outstanding adoption due to its robust efficiency in recent times. Image fault detection and classification have also been implemented industrially for product defect detection, as well as for maintaining standards and optimizi...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/8/2449 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850155106354331648 |
|---|---|
| author | Joung-Hwan Yoon Chibuzo Nwabufo Okwuosa Nnamdi Chukwunweike Aronwora Jang-Wook Hur |
| author_facet | Joung-Hwan Yoon Chibuzo Nwabufo Okwuosa Nnamdi Chukwunweike Aronwora Jang-Wook Hur |
| author_sort | Joung-Hwan Yoon |
| collection | DOAJ |
| description | The industrial application of artificial intelligence (AI) has witnessed outstanding adoption due to its robust efficiency in recent times. Image fault detection and classification have also been implemented industrially for product defect detection, as well as for maintaining standards and optimizing processes using AI. However, there are deep concerns regarding the latency in the performance of AI for fault detection in glossy and curved surface products, due to their nature and reflective surfaces, which hinder the adequate capturing of defective areas using traditional cameras. Consequently, this study presents an enhanced method for curvy and glossy surface image data collection using a Basler vision camera with specialized lighting and KEYENCE displacement sensors, which are used to train deep learning models. Our approach employed image data generated from normal and two defect conditions to train eight deep learning algorithms: four custom convolutional neural networks (CNNs), two variations of VGG-16, and two variations of ResNet-50. The objective was to develop a computationally robust and efficient model by deploying global assessment metrics as evaluation criteria. Our results indicate that a variation of ResNet-50, ResNet-50<sub>224</sub>, demonstrated the best overall efficiency, achieving an accuracy of 97.97%, a loss of 0.1030, and an average training step time of 839 milliseconds. However, in terms of computational efficiency, it was outperformed by one of the custom CNN models, CNN<sub>6</sub>-240, which achieved an accuracy of 95.08%, a loss of 0.2753, and an average step time of 94 milliseconds, making CNN<sub>6</sub>-240 a viable option for computational resource-sensitive environments. |
| format | Article |
| id | doaj-art-4fefc30a36b74b9393ee54004b69d9df |
| institution | OA Journals |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-4fefc30a36b74b9393ee54004b69d9df2025-08-20T02:25:02ZengMDPI AGSensors1424-82202025-04-01258244910.3390/s25082449Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging SystemsJoung-Hwan Yoon0Chibuzo Nwabufo Okwuosa1Nnamdi Chukwunweike Aronwora2Jang-Wook Hur3Department of Mechanical Engineering (Department of Aeronautics, Mechanical and Electronic Convergence Engineering), Kumoh National Institute of Technology, 61 Daehak-ro, Gumi-si 39177, Gyeonsangbuk-do, Republic of KoreaDepartment of Mechanical Engineering (Department of Aeronautics, Mechanical and Electronic Convergence Engineering), Kumoh National Institute of Technology, 61 Daehak-ro, Gumi-si 39177, Gyeonsangbuk-do, Republic of KoreaDepartment of Mechanical Engineering (Department of Aeronautics, Mechanical and Electronic Convergence Engineering), Kumoh National Institute of Technology, 61 Daehak-ro, Gumi-si 39177, Gyeonsangbuk-do, Republic of KoreaDepartment of Mechanical Engineering (Department of Aeronautics, Mechanical and Electronic Convergence Engineering), Kumoh National Institute of Technology, 61 Daehak-ro, Gumi-si 39177, Gyeonsangbuk-do, Republic of KoreaThe industrial application of artificial intelligence (AI) has witnessed outstanding adoption due to its robust efficiency in recent times. Image fault detection and classification have also been implemented industrially for product defect detection, as well as for maintaining standards and optimizing processes using AI. However, there are deep concerns regarding the latency in the performance of AI for fault detection in glossy and curved surface products, due to their nature and reflective surfaces, which hinder the adequate capturing of defective areas using traditional cameras. Consequently, this study presents an enhanced method for curvy and glossy surface image data collection using a Basler vision camera with specialized lighting and KEYENCE displacement sensors, which are used to train deep learning models. Our approach employed image data generated from normal and two defect conditions to train eight deep learning algorithms: four custom convolutional neural networks (CNNs), two variations of VGG-16, and two variations of ResNet-50. The objective was to develop a computationally robust and efficient model by deploying global assessment metrics as evaluation criteria. Our results indicate that a variation of ResNet-50, ResNet-50<sub>224</sub>, demonstrated the best overall efficiency, achieving an accuracy of 97.97%, a loss of 0.1030, and an average training step time of 839 milliseconds. However, in terms of computational efficiency, it was outperformed by one of the custom CNN models, CNN<sub>6</sub>-240, which achieved an accuracy of 95.08%, a loss of 0.2753, and an average step time of 94 milliseconds, making CNN<sub>6</sub>-240 a viable option for computational resource-sensitive environments.https://www.mdpi.com/1424-8220/25/8/2449convolutional neural networkResNet-50VGG-16Dijkstra’s algorithmglossy surfacecurved surface |
| spellingShingle | Joung-Hwan Yoon Chibuzo Nwabufo Okwuosa Nnamdi Chukwunweike Aronwora Jang-Wook Hur Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging Systems Sensors convolutional neural network ResNet-50 VGG-16 Dijkstra’s algorithm glossy surface curved surface |
| title | Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging Systems |
| title_full | Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging Systems |
| title_fullStr | Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging Systems |
| title_full_unstemmed | Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging Systems |
| title_short | Optimizing Defect Detection on Glossy and Curved Surfaces Using Deep Learning and Advanced Imaging Systems |
| title_sort | optimizing defect detection on glossy and curved surfaces using deep learning and advanced imaging systems |
| topic | convolutional neural network ResNet-50 VGG-16 Dijkstra’s algorithm glossy surface curved surface |
| url | https://www.mdpi.com/1424-8220/25/8/2449 |
| work_keys_str_mv | AT jounghwanyoon optimizingdefectdetectiononglossyandcurvedsurfacesusingdeeplearningandadvancedimagingsystems AT chibuzonwabufookwuosa optimizingdefectdetectiononglossyandcurvedsurfacesusingdeeplearningandadvancedimagingsystems AT nnamdichukwunweikearonwora optimizingdefectdetectiononglossyandcurvedsurfacesusingdeeplearningandadvancedimagingsystems AT jangwookhur optimizingdefectdetectiononglossyandcurvedsurfacesusingdeeplearningandadvancedimagingsystems |