Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep Learning
Aiming at the problem of the appearance scratch detection of the smart meter HPLC communication module, an advanced pixel-level crack detection architecture Res-DU-Net is proposed, which is characterized by the use of deep convolutional neural network technology. First, Res-DU-Net achieves pixel-lev...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | Chinese |
| Published: |
Harbin University of Science and Technology Publications
2022-06-01
|
| Series: | Journal of Harbin University of Science and Technology |
| Subjects: | |
| Online Access: | https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2097 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839614034448482304 |
|---|---|
| author | SUN Kai ZHAI Xiao-hui ZHAO Ji-fu SUN Yan-ling DONG Xian-guang WANG Yan |
| author_facet | SUN Kai ZHAI Xiao-hui ZHAO Ji-fu SUN Yan-ling DONG Xian-guang WANG Yan |
| author_sort | SUN Kai |
| collection | DOAJ |
| description | Aiming at the problem of the appearance scratch detection of the smart meter HPLC communication module, an advanced pixel-level crack detection architecture Res-DU-Net is proposed, which is characterized by the use of deep convolutional neural network technology. First, Res-DU-Net achieves pixel-level slip detection through operations such as convolution, residual structure, cavity convolution, and cascade, forming a U-shaped model structure. Secondly, the Adam algorithm was used to train and verify the appearance image of the smart meter communication module. Finally, Res-DU-Net has an accuracy rate of 0.985 and a recall rate of 0.987 when the learning rate is 0.001 and the loss rate is 0.020. Experimental results prove that Res-DU-Net has more advantages than traditional methods, Fully Convolutional Network (FCN) and U-Net in pixel-level scratch detection. |
| format | Article |
| id | doaj-art-9d6a06f5a7104d41ac8f18cbc4c33a73 |
| institution | Matheson Library |
| issn | 1007-2683 |
| language | zho |
| publishDate | 2022-06-01 |
| publisher | Harbin University of Science and Technology Publications |
| record_format | Article |
| series | Journal of Harbin University of Science and Technology |
| spelling | doaj-art-9d6a06f5a7104d41ac8f18cbc4c33a732025-07-26T09:27:31ZzhoHarbin University of Science and Technology PublicationsJournal of Harbin University of Science and Technology1007-26832022-06-012703667210.15938/j.jhust.2022.03.009Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep LearningSUN Kai0ZHAI Xiao-hui1ZHAO Ji-fu2SUN Yan-ling3DONG Xian-guang4WANG Yan5Marketing Service Center (Metering Center), State Grid, Shandong Electric Power Company, Ji′nan 250000, ChinaMarketing Service Center (Metering Center), State Grid, Shandong Electric Power Company, Ji′nan 250000, ChinaMarketing Service Center (Metering Center), State Grid, Shandong Electric Power Company, Ji′nan 250000, ChinaMarketing Service Center (Metering Center), State Grid, Shandong Electric Power Company, Ji′nan 250000, ChinaMarketing Service Center (Metering Center), State Grid, Shandong Electric Power Company, Ji′nan 250000, ChinaMarketing Service Center (Metering Center), State Grid, Shandong Electric Power Company, Ji′nan 250000, ChinaAiming at the problem of the appearance scratch detection of the smart meter HPLC communication module, an advanced pixel-level crack detection architecture Res-DU-Net is proposed, which is characterized by the use of deep convolutional neural network technology. First, Res-DU-Net achieves pixel-level slip detection through operations such as convolution, residual structure, cavity convolution, and cascade, forming a U-shaped model structure. Secondly, the Adam algorithm was used to train and verify the appearance image of the smart meter communication module. Finally, Res-DU-Net has an accuracy rate of 0.985 and a recall rate of 0.987 when the learning rate is 0.001 and the loss rate is 0.020. Experimental results prove that Res-DU-Net has more advantages than traditional methods, Fully Convolutional Network (FCN) and U-Net in pixel-level scratch detection.https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2097hplcscratch detectionresidual structurecavity convolution |
| spellingShingle | SUN Kai ZHAI Xiao-hui ZHAO Ji-fu SUN Yan-ling DONG Xian-guang WANG Yan Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep Learning Journal of Harbin University of Science and Technology hplc scratch detection residual structure cavity convolution |
| title | Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep Learning |
| title_full | Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep Learning |
| title_fullStr | Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep Learning |
| title_full_unstemmed | Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep Learning |
| title_short | Scratch Detection on the Appearance of Smart Meter HPLC Communication Module Based on Deep Learning |
| title_sort | scratch detection on the appearance of smart meter hplc communication module based on deep learning |
| topic | hplc scratch detection residual structure cavity convolution |
| url | https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2097 |
| work_keys_str_mv | AT sunkai scratchdetectionontheappearanceofsmartmeterhplccommunicationmodulebasedondeeplearning AT zhaixiaohui scratchdetectionontheappearanceofsmartmeterhplccommunicationmodulebasedondeeplearning AT zhaojifu scratchdetectionontheappearanceofsmartmeterhplccommunicationmodulebasedondeeplearning AT sunyanling scratchdetectionontheappearanceofsmartmeterhplccommunicationmodulebasedondeeplearning AT dongxianguang scratchdetectionontheappearanceofsmartmeterhplccommunicationmodulebasedondeeplearning AT wangyan scratchdetectionontheappearanceofsmartmeterhplccommunicationmodulebasedondeeplearning |