A Detection Method for Conveyor Belt Damage with Small Size and Low Contrast
[Purposes] An improved YOLOv4 detection model is proposed to solve the problems of missing detection and false detection when the existing models detect objects with small size and low contrast with the background. [Methods] In order to solve the problem of small size, first, the DDS unit is designe...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Editorial Office of Journal of Taiyuan University of Technology
2025-01-01
|
| Series: | Taiyuan Ligong Daxue xuebao |
| Subjects: | |
| Online Access: | https://tyutjournal.tyut.edu.cn/englishpaper/show-2374.html |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823856997527715840 |
|---|---|
| author | HAN Yajie HAO Xiaoli NIU Baoning XUE Jindong |
| author_facet | HAN Yajie HAO Xiaoli NIU Baoning XUE Jindong |
| author_sort | HAN Yajie |
| collection | DOAJ |
| description | [Purposes] An improved YOLOv4 detection model is proposed to solve the problems of missing detection and false detection when the existing models detect objects with small size and low contrast with the background. [Methods] In order to solve the problem of small size, first, the DDS unit is designed to replace the Res unit in the backbone network. By connecting features of different levels across layers, complete and rich multi-scale features can be obtained, and small-size damage detection can be completed. Second, the gradient harmonized mechanism is introduced into the classification loss function, and the weight of small-size damage is dynamically adjusted to make it fully trained. Aiming at the low contrast between damage and background, first, the coordinate attention mechanism is embedded in the deep network layer of the backbone network to enhance the model′s attention to damage characteristics and reduce the interference of background noise. Second, the accurate decoupled head is designed to improve detection accuracy by solving the contradiction between classification and location requirements for features. [Findings] Experimental results demonstrate that the mean average precision of this model is increased by 3.92% compared with that of YOLOv4, and the detection accuracy of small-size crack damage and low-contrast wear damage is improved by 4.32% and 4.24%, respectively, which effectively solves the problems of missed detection and false detection. |
| format | Article |
| id | doaj-art-99ea110bb9c94407b94aba71fdc0f4bd |
| institution | Kabale University |
| issn | 1007-9432 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Editorial Office of Journal of Taiyuan University of Technology |
| record_format | Article |
| series | Taiyuan Ligong Daxue xuebao |
| spelling | doaj-art-99ea110bb9c94407b94aba71fdc0f4bd2025-02-12T03:34:22ZengEditorial Office of Journal of Taiyuan University of TechnologyTaiyuan Ligong Daxue xuebao1007-94322025-01-0156114815610.16355/j.tyut.1007-9432.202207131007-9432(2025)01-0148-09A Detection Method for Conveyor Belt Damage with Small Size and Low ContrastHAN Yajie0HAO Xiaoli1NIU Baoning2XUE Jindong3College of Computer Science and Technology (College of Data Science), Taiyuan University of Technology, Taiyuan, Shanxi, ChinaCollege of Computer Science and Technology (College of Data Science), Taiyuan University of Technology, Taiyuan, Shanxi, ChinaCollege of Computer Science and Technology (College of Data Science), Taiyuan University of Technology, Taiyuan, Shanxi, ChinaState Grid Taiyuan Power Supply Company, Taiyuan, Shanxi, China[Purposes] An improved YOLOv4 detection model is proposed to solve the problems of missing detection and false detection when the existing models detect objects with small size and low contrast with the background. [Methods] In order to solve the problem of small size, first, the DDS unit is designed to replace the Res unit in the backbone network. By connecting features of different levels across layers, complete and rich multi-scale features can be obtained, and small-size damage detection can be completed. Second, the gradient harmonized mechanism is introduced into the classification loss function, and the weight of small-size damage is dynamically adjusted to make it fully trained. Aiming at the low contrast between damage and background, first, the coordinate attention mechanism is embedded in the deep network layer of the backbone network to enhance the model′s attention to damage characteristics and reduce the interference of background noise. Second, the accurate decoupled head is designed to improve detection accuracy by solving the contradiction between classification and location requirements for features. [Findings] Experimental results demonstrate that the mean average precision of this model is increased by 3.92% compared with that of YOLOv4, and the detection accuracy of small-size crack damage and low-contrast wear damage is improved by 4.32% and 4.24%, respectively, which effectively solves the problems of missed detection and false detection.https://tyutjournal.tyut.edu.cn/englishpaper/show-2374.htmldamage of conveyor beltyolov4dds unitgradient harmonized mechanismcoordinate attention mechanismaccurate decoupled head |
| spellingShingle | HAN Yajie HAO Xiaoli NIU Baoning XUE Jindong A Detection Method for Conveyor Belt Damage with Small Size and Low Contrast Taiyuan Ligong Daxue xuebao damage of conveyor belt yolov4 dds unit gradient harmonized mechanism coordinate attention mechanism accurate decoupled head |
| title | A Detection Method for Conveyor Belt Damage with Small Size and Low Contrast |
| title_full | A Detection Method for Conveyor Belt Damage with Small Size and Low Contrast |
| title_fullStr | A Detection Method for Conveyor Belt Damage with Small Size and Low Contrast |
| title_full_unstemmed | A Detection Method for Conveyor Belt Damage with Small Size and Low Contrast |
| title_short | A Detection Method for Conveyor Belt Damage with Small Size and Low Contrast |
| title_sort | detection method for conveyor belt damage with small size and low contrast |
| topic | damage of conveyor belt yolov4 dds unit gradient harmonized mechanism coordinate attention mechanism accurate decoupled head |
| url | https://tyutjournal.tyut.edu.cn/englishpaper/show-2374.html |
| work_keys_str_mv | AT hanyajie adetectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast AT haoxiaoli adetectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast AT niubaoning adetectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast AT xuejindong adetectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast AT hanyajie detectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast AT haoxiaoli detectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast AT niubaoning detectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast AT xuejindong detectionmethodforconveyorbeltdamagewithsmallsizeandlowcontrast |