Thickness measurement of immersion metal carbon slide based on image segmentation
The thickness of a metal-immersed carbon slide mounted on a train’s flow shoe was measured by using machine vision and deep learning. A method for measuring the thickness of carbon slide plate based on improved U<sup>2</sup>-Net is proposed. Aiming at the problem that the edge feature ex...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Croatian Metallurgical Society
2024-01-01
|
| Series: | Metalurgija |
| Subjects: | |
| Online Access: | https://hrcak.srce.hr/file/456163 |
| _version_ | 1797221266464178176 |
|---|---|
| author | A. Y. Zheng C. Y. Chang W. M. Liu S. G. Qiao |
| author_facet | A. Y. Zheng C. Y. Chang W. M. Liu S. G. Qiao |
| author_sort | A. Y. Zheng |
| collection | DOAJ |
| description | The thickness of a metal-immersed carbon slide mounted on a train’s flow shoe was measured by using machine vision and deep learning. A method for measuring the thickness of carbon slide plate based on improved U<sup>2</sup>-Net is proposed. Aiming at the problem that the edge feature extraction is not obvious, a new feature extraction module is designed. Efficient Channel Attention (ECA) mechanism and pool residual structure are used to make the network more suitable for metal-immersed carbon slide image segmentation. The experimental results show that the improved U2-Net network accuracy reaches 99,4 %, and the average absolute error is only 0,4 %. The thickness measurement accuracy of metallized carbon slide using improved U<sup>2</sup>-Net network reaches 0,5 mm. |
| first_indexed | 2024-04-24T13:02:42Z |
| format | Article |
| id | doaj.art-3bc8a8433c89486dbc73809b82569d24 |
| institution | Directory Open Access Journal |
| issn | 0543-5846 1334-2576 |
| language | English |
| last_indexed | 2024-04-24T13:02:42Z |
| publishDate | 2024-01-01 |
| publisher | Croatian Metallurgical Society |
| record_format | Article |
| series | Metalurgija |
| spelling | doaj.art-3bc8a8433c89486dbc73809b82569d242024-04-05T10:19:04ZengCroatian Metallurgical SocietyMetalurgija0543-58461334-25762024-01-01633-4451453Thickness measurement of immersion metal carbon slide based on image segmentationA. Y. Zheng0C. Y. Chang1W. M. Liu2S. G. Qiao3College of Mechanical Engineering, North China University of Science and Technology, Hebei, Tangshan, ChinaCollege of Mechanical Engineering, North China University of Science and Technology, Hebei, Tangshan, ChinaCollege of Mechanical Engineering, North China University of Science and Technology, Hebei, Tangshan, ChinaCollege of Mechanical Engineering, North China University of Science and Technology, Hebei, Tangshan, ChinaThe thickness of a metal-immersed carbon slide mounted on a train’s flow shoe was measured by using machine vision and deep learning. A method for measuring the thickness of carbon slide plate based on improved U<sup>2</sup>-Net is proposed. Aiming at the problem that the edge feature extraction is not obvious, a new feature extraction module is designed. Efficient Channel Attention (ECA) mechanism and pool residual structure are used to make the network more suitable for metal-immersed carbon slide image segmentation. The experimental results show that the improved U2-Net network accuracy reaches 99,4 %, and the average absolute error is only 0,4 %. The thickness measurement accuracy of metallized carbon slide using improved U<sup>2</sup>-Net network reaches 0,5 mm.https://hrcak.srce.hr/file/456163railimmersion metal carbon slideimage segmentationU<sup>2</sup>-Net |
| spellingShingle | A. Y. Zheng C. Y. Chang W. M. Liu S. G. Qiao Thickness measurement of immersion metal carbon slide based on image segmentation Metalurgija rail immersion metal carbon slide image segmentation U<sup>2</sup>-Net |
| title | Thickness measurement of immersion metal carbon slide based on image segmentation |
| title_full | Thickness measurement of immersion metal carbon slide based on image segmentation |
| title_fullStr | Thickness measurement of immersion metal carbon slide based on image segmentation |
| title_full_unstemmed | Thickness measurement of immersion metal carbon slide based on image segmentation |
| title_short | Thickness measurement of immersion metal carbon slide based on image segmentation |
| title_sort | thickness measurement of immersion metal carbon slide based on image segmentation |
| topic | rail immersion metal carbon slide image segmentation U<sup>2</sup>-Net |
| url | https://hrcak.srce.hr/file/456163 |
| work_keys_str_mv | AT ayzheng thicknessmeasurementofimmersionmetalcarbonslidebasedonimagesegmentation AT cychang thicknessmeasurementofimmersionmetalcarbonslidebasedonimagesegmentation AT wmliu thicknessmeasurementofimmersionmetalcarbonslidebasedonimagesegmentation AT sgqiao thicknessmeasurementofimmersionmetalcarbonslidebasedonimagesegmentation |