Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement
Strip surface defects have large intraclass and small interclass differences, resulting in the available detection techniques having either a low accuracy or very poor real-time performance. In order to improve the ability for capturing steel surface defects, the context fusion structure introduces...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/11/2440 |
| _version_ | 1797597664141901824 |
|---|---|
| author | Yiming Li Lixin He Min Zhang Zhi Cheng Wangwei Liu Zijun Wu |
| author_facet | Yiming Li Lixin He Min Zhang Zhi Cheng Wangwei Liu Zijun Wu |
| author_sort | Yiming Li |
| collection | DOAJ |
| description | Strip surface defects have large intraclass and small interclass differences, resulting in the available detection techniques having either a low accuracy or very poor real-time performance. In order to improve the ability for capturing steel surface defects, the context fusion structure introduces the local information of the shallow layer and the semantic information of the deep layer into multiscale feature maps. In addition, for filtering the semantic conflicts and redundancies arising from context fusion, a feature refinement module is introduced in our method, which further improves the detection accuracy. Our experimental results show that this significantly improved the performance. In particular, our method achieved 79.5% mAP and 71 FPS on the public NEU-DET dataset. This means that our method had a higher detection accuracy compared to other techniques. |
| first_indexed | 2024-03-11T03:08:45Z |
| format | Article |
| id | doaj.art-9421e7a1ec404c32af19cbc2046a5dc5 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T03:08:45Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-9421e7a1ec404c32af19cbc2046a5dc52023-11-18T07:44:59ZengMDPI AGElectronics2079-92922023-05-011211244010.3390/electronics12112440Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature RefinementYiming Li0Lixin He1Min Zhang2Zhi Cheng3Wangwei Liu4Zijun Wu5School of Artificial Intelligence and Big Data, Hefei University, Hefei 230601, ChinaSchool of Artificial Intelligence and Big Data, Hefei University, Hefei 230601, ChinaSchool of Artificial Intelligence and Big Data, Hefei University, Hefei 230601, ChinaSchool of Artificial Intelligence and Big Data, Hefei University, Hefei 230601, ChinaSchool of Artificial Intelligence and Big Data, Hefei University, Hefei 230601, ChinaSchool of Artificial Intelligence and Big Data, Hefei University, Hefei 230601, ChinaStrip surface defects have large intraclass and small interclass differences, resulting in the available detection techniques having either a low accuracy or very poor real-time performance. In order to improve the ability for capturing steel surface defects, the context fusion structure introduces the local information of the shallow layer and the semantic information of the deep layer into multiscale feature maps. In addition, for filtering the semantic conflicts and redundancies arising from context fusion, a feature refinement module is introduced in our method, which further improves the detection accuracy. Our experimental results show that this significantly improved the performance. In particular, our method achieved 79.5% mAP and 71 FPS on the public NEU-DET dataset. This means that our method had a higher detection accuracy compared to other techniques.https://www.mdpi.com/2079-9292/12/11/2440context fusionfeature refinementsteel defect detectionSSD |
| spellingShingle | Yiming Li Lixin He Min Zhang Zhi Cheng Wangwei Liu Zijun Wu Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement Electronics context fusion feature refinement steel defect detection SSD |
| title | Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement |
| title_full | Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement |
| title_fullStr | Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement |
| title_full_unstemmed | Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement |
| title_short | Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement |
| title_sort | improving the performance of the single shot multibox detector for steel surface defects with context fusion and feature refinement |
| topic | context fusion feature refinement steel defect detection SSD |
| url | https://www.mdpi.com/2079-9292/12/11/2440 |
| work_keys_str_mv | AT yimingli improvingtheperformanceofthesingleshotmultiboxdetectorforsteelsurfacedefectswithcontextfusionandfeaturerefinement AT lixinhe improvingtheperformanceofthesingleshotmultiboxdetectorforsteelsurfacedefectswithcontextfusionandfeaturerefinement AT minzhang improvingtheperformanceofthesingleshotmultiboxdetectorforsteelsurfacedefectswithcontextfusionandfeaturerefinement AT zhicheng improvingtheperformanceofthesingleshotmultiboxdetectorforsteelsurfacedefectswithcontextfusionandfeaturerefinement AT wangweiliu improvingtheperformanceofthesingleshotmultiboxdetectorforsteelsurfacedefectswithcontextfusionandfeaturerefinement AT zijunwu improvingtheperformanceofthesingleshotmultiboxdetectorforsteelsurfacedefectswithcontextfusionandfeaturerefinement |