Lightweight defect detection algorithm of tunnel lining based on knowledge distillation
Abstract Due to the influence of construction quality, engineering geology and hydrological environment, defects such as dehollowing and insufficient compaction can occur in tunnels. Aiming at the problems of complex detection model, poor real-time performance and low accuracy of the current tunnel...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-11-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-77404-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850061793799438336 |
|---|---|
| author | Anfu Zhu Jiaxiao Xie Bin Wang Heng Guo Zilong Guo Jie Wang Lei Xu SiXin Zhu Zhanping Yang |
| author_facet | Anfu Zhu Jiaxiao Xie Bin Wang Heng Guo Zilong Guo Jie Wang Lei Xu SiXin Zhu Zhanping Yang |
| author_sort | Anfu Zhu |
| collection | DOAJ |
| description | Abstract Due to the influence of construction quality, engineering geology and hydrological environment, defects such as dehollowing and insufficient compaction can occur in tunnels. Aiming at the problems of complex detection model, poor real-time performance and low accuracy of the current tunnel lining defect detection methods, the study proposes a lightweight defect detection algorithm of tunnel lining based on knowledge distillation. Firstly, a high-precision teacher model based on yolov5s was constructed by constructing a C3CSFM module that combines residual structure and attention mechanism, a MDFPN network structure with multi-scale feature fusion and a reweighted RWNMS re-screening mechanism. Secondly, in the distillation process, the feature and output dimension results are fused to improve the detection accuracy, and the mask feature relationship is learned in the space and channel dimension to improve the real-time detection. Tests on the tunnel lining radar defect image dataset showed that the number of parameters of the improved model was reduced from 16.03 MB to 3.20 MB, a reduction of 80%, and the average accuracy was improved from 83.4 to 86.5%, an increase of 3.1%. On the basis of maintaining the structure and detection performance of the model, the lightweight degree of the model is greatly improved, and the high-precision and real-time detection of tunnel lining defects is realized. |
| format | Article |
| id | doaj-art-2c48d9ffa1864b07a8e27fc96ca723e0 |
| institution | DOAJ |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-2c48d9ffa1864b07a8e27fc96ca723e02025-08-20T02:50:07ZengNature PortfolioScientific Reports2045-23222024-11-0114111610.1038/s41598-024-77404-8Lightweight defect detection algorithm of tunnel lining based on knowledge distillationAnfu Zhu0Jiaxiao Xie1Bin Wang2Heng Guo3Zilong Guo4Jie Wang5Lei Xu6SiXin Zhu7Zhanping Yang8North China University of Water Resources and Electric PowerNorth China University of Water Resources and Electric PowerNorth China University of Water Resources and Electric PowerNorth China University of Water Resources and Electric PowerNorth China University of Water Resources and Electric PowerNorth China University of Water Resources and Electric PowerNorth China University of Water Resources and Electric PowerNorth China University of Water Resources and Electric PowerScience and Technology Research Institute of China Railway Zhengzhou Group Co., Ltd.Abstract Due to the influence of construction quality, engineering geology and hydrological environment, defects such as dehollowing and insufficient compaction can occur in tunnels. Aiming at the problems of complex detection model, poor real-time performance and low accuracy of the current tunnel lining defect detection methods, the study proposes a lightweight defect detection algorithm of tunnel lining based on knowledge distillation. Firstly, a high-precision teacher model based on yolov5s was constructed by constructing a C3CSFM module that combines residual structure and attention mechanism, a MDFPN network structure with multi-scale feature fusion and a reweighted RWNMS re-screening mechanism. Secondly, in the distillation process, the feature and output dimension results are fused to improve the detection accuracy, and the mask feature relationship is learned in the space and channel dimension to improve the real-time detection. Tests on the tunnel lining radar defect image dataset showed that the number of parameters of the improved model was reduced from 16.03 MB to 3.20 MB, a reduction of 80%, and the average accuracy was improved from 83.4 to 86.5%, an increase of 3.1%. On the basis of maintaining the structure and detection performance of the model, the lightweight degree of the model is greatly improved, and the high-precision and real-time detection of tunnel lining defects is realized.https://doi.org/10.1038/s41598-024-77404-8Tunnel detectionDeep learningModel compression algorithmsKnowledge distillation |
| spellingShingle | Anfu Zhu Jiaxiao Xie Bin Wang Heng Guo Zilong Guo Jie Wang Lei Xu SiXin Zhu Zhanping Yang Lightweight defect detection algorithm of tunnel lining based on knowledge distillation Scientific Reports Tunnel detection Deep learning Model compression algorithms Knowledge distillation |
| title | Lightweight defect detection algorithm of tunnel lining based on knowledge distillation |
| title_full | Lightweight defect detection algorithm of tunnel lining based on knowledge distillation |
| title_fullStr | Lightweight defect detection algorithm of tunnel lining based on knowledge distillation |
| title_full_unstemmed | Lightweight defect detection algorithm of tunnel lining based on knowledge distillation |
| title_short | Lightweight defect detection algorithm of tunnel lining based on knowledge distillation |
| title_sort | lightweight defect detection algorithm of tunnel lining based on knowledge distillation |
| topic | Tunnel detection Deep learning Model compression algorithms Knowledge distillation |
| url | https://doi.org/10.1038/s41598-024-77404-8 |
| work_keys_str_mv | AT anfuzhu lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT jiaxiaoxie lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT binwang lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT hengguo lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT zilongguo lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT jiewang lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT leixu lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT sixinzhu lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation AT zhanpingyang lightweightdefectdetectionalgorithmoftunnelliningbasedonknowledgedistillation |