Research of U-Net-Based CNN Architectures for Metal Surface Defect Detection
The quality, wear and safety of metal structures can be controlled effectively, provided that surface defects, which occur on metal structures, are detected at the right time. Over the past 10 years, researchers have proposed a number of neural network architectures that have shown high efficiency i...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Machines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1702/10/5/327 |
| _version_ | 1797498431275532288 |
|---|---|
| author | Ihor Konovalenko Pavlo Maruschak Janette Brezinová Olegas Prentkovskis Jakub Brezina |
| author_facet | Ihor Konovalenko Pavlo Maruschak Janette Brezinová Olegas Prentkovskis Jakub Brezina |
| author_sort | Ihor Konovalenko |
| collection | DOAJ |
| description | The quality, wear and safety of metal structures can be controlled effectively, provided that surface defects, which occur on metal structures, are detected at the right time. Over the past 10 years, researchers have proposed a number of neural network architectures that have shown high efficiency in various areas, including image classification, segmentation and recognition. However, choosing the best architecture for this particular task is often problematic. In order to compare various techniques for detecting defects such as “scratch abrasion”, we created and investigated U-Net-like architectures with encoders such as ResNet, SEResNet, SEResNeXt, DenseNet, InceptionV3, Inception-ResNetV2, MobileNet and EfficientNet. The relationship between training validation metrics and final segmentation test metrics was investigated. The correlation between the loss function, the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>D</mi><mi>S</mi><mi>C</mi></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>I</mi><mi>o</mi><mi>U</mi></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>e</mi><mi>c</mi><mi>a</mi><mi>l</mi><mi>l</mi></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>P</mi><mi>r</mi><mi>e</mi><mi>c</mi><mi>i</mi><mi>s</mi><mi>i</mi><mi>o</mi><mi>n</mi></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>F</mi><mn>1</mn></mrow></semantics></math></inline-formula> validation metrics and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>D</mi><mi>S</mi><mi>C</mi></mrow></semantics></math></inline-formula> test metrics was calculated. Recognition accuracy was analyzed as affected by the optimizer during neural network training. In the context of this problem, neural networks trained using the stochastic gradient descent optimizer with Nesterov momentum were found to have the best generalizing properties. To select the best model during its training on the basis of the validation metrics, the main test metrics of recognition quality (Dice similarity coefficient) were analyzed depending on the validation metrics. The ResNet and DenseNet models were found to achieve the best generalizing properties for our task. The highest recognition accuracy was attained using the U-Net model with a ResNet152 backbone. The results obtained on the test dataset were <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>D</mi><mi>S</mi><mi>C</mi><mo>=</mo><mn>0.9304</mn></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>I</mi><mi>o</mi><mi>U</mi><mo>=</mo><mn>0.9122</mn></mrow></semantics></math></inline-formula>. |
| first_indexed | 2024-03-10T03:33:17Z |
| format | Article |
| id | doaj.art-e640bab6fe04455e90901ef8c09cda2a |
| institution | Directory Open Access Journal |
| issn | 2075-1702 |
| language | English |
| last_indexed | 2024-03-10T03:33:17Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machines |
| spelling | doaj.art-e640bab6fe04455e90901ef8c09cda2a2023-11-23T11:52:21ZengMDPI AGMachines2075-17022022-04-0110532710.3390/machines10050327Research of U-Net-Based CNN Architectures for Metal Surface Defect DetectionIhor Konovalenko0Pavlo Maruschak1Janette Brezinová2Olegas Prentkovskis3Jakub Brezina4Department of Industrial Automation, Ternopil National Ivan Puluj Technical University, Rus’ka Str. 56, 46001 Ternopil, UkraineDepartment of Industrial Automation, Ternopil National Ivan Puluj Technical University, Rus’ka Str. 56, 46001 Ternopil, UkraineDepartment of Technology, Materials and Computer Supported Production, Faculty of Mechanical Engineering, Technical University of Košice, Mäsiarska 74, 04001 Košice, SlovakiaDepartment of Mobile Machinery and Railway Transport, Vilnius Gediminas Technical University, Plytinės g. 27, LT-10101 Vilnius, LithuaniaDepartment of Technology, Materials and Computer Supported Production, Faculty of Mechanical Engineering, Technical University of Košice, Mäsiarska 74, 04001 Košice, SlovakiaThe quality, wear and safety of metal structures can be controlled effectively, provided that surface defects, which occur on metal structures, are detected at the right time. Over the past 10 years, researchers have proposed a number of neural network architectures that have shown high efficiency in various areas, including image classification, segmentation and recognition. However, choosing the best architecture for this particular task is often problematic. In order to compare various techniques for detecting defects such as “scratch abrasion”, we created and investigated U-Net-like architectures with encoders such as ResNet, SEResNet, SEResNeXt, DenseNet, InceptionV3, Inception-ResNetV2, MobileNet and EfficientNet. The relationship between training validation metrics and final segmentation test metrics was investigated. The correlation between the loss function, the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>D</mi><mi>S</mi><mi>C</mi></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>I</mi><mi>o</mi><mi>U</mi></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>e</mi><mi>c</mi><mi>a</mi><mi>l</mi><mi>l</mi></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>P</mi><mi>r</mi><mi>e</mi><mi>c</mi><mi>i</mi><mi>s</mi><mi>i</mi><mi>o</mi><mi>n</mi></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>F</mi><mn>1</mn></mrow></semantics></math></inline-formula> validation metrics and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>D</mi><mi>S</mi><mi>C</mi></mrow></semantics></math></inline-formula> test metrics was calculated. Recognition accuracy was analyzed as affected by the optimizer during neural network training. In the context of this problem, neural networks trained using the stochastic gradient descent optimizer with Nesterov momentum were found to have the best generalizing properties. To select the best model during its training on the basis of the validation metrics, the main test metrics of recognition quality (Dice similarity coefficient) were analyzed depending on the validation metrics. The ResNet and DenseNet models were found to achieve the best generalizing properties for our task. The highest recognition accuracy was attained using the U-Net model with a ResNet152 backbone. The results obtained on the test dataset were <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>D</mi><mi>S</mi><mi>C</mi><mo>=</mo><mn>0.9304</mn></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>I</mi><mi>o</mi><mi>U</mi><mo>=</mo><mn>0.9122</mn></mrow></semantics></math></inline-formula>.https://www.mdpi.com/2075-1702/10/5/327surface defect detectionvisual inspection technologyimage segmentationCNN optimizerstrip surfacemetallurgy |
| spellingShingle | Ihor Konovalenko Pavlo Maruschak Janette Brezinová Olegas Prentkovskis Jakub Brezina Research of U-Net-Based CNN Architectures for Metal Surface Defect Detection Machines surface defect detection visual inspection technology image segmentation CNN optimizer strip surface metallurgy |
| title | Research of U-Net-Based CNN Architectures for Metal Surface Defect Detection |
| title_full | Research of U-Net-Based CNN Architectures for Metal Surface Defect Detection |
| title_fullStr | Research of U-Net-Based CNN Architectures for Metal Surface Defect Detection |
| title_full_unstemmed | Research of U-Net-Based CNN Architectures for Metal Surface Defect Detection |
| title_short | Research of U-Net-Based CNN Architectures for Metal Surface Defect Detection |
| title_sort | research of u net based cnn architectures for metal surface defect detection |
| topic | surface defect detection visual inspection technology image segmentation CNN optimizer strip surface metallurgy |
| url | https://www.mdpi.com/2075-1702/10/5/327 |
| work_keys_str_mv | AT ihorkonovalenko researchofunetbasedcnnarchitecturesformetalsurfacedefectdetection AT pavlomaruschak researchofunetbasedcnnarchitecturesformetalsurfacedefectdetection AT janettebrezinova researchofunetbasedcnnarchitecturesformetalsurfacedefectdetection AT olegasprentkovskis researchofunetbasedcnnarchitecturesformetalsurfacedefectdetection AT jakubbrezina researchofunetbasedcnnarchitecturesformetalsurfacedefectdetection |