Improved YOLOv8n based helmet wearing inspection method
Abstract This paper proposes the YOLOv8n_H method to address issues regarding parameter redundancy, slow inference speed, and suboptimal detection precision in contemporary helmet-wearing target recognition algorithms. The YOLOv8 C2f module is enhanced with a new SC_Bottleneck structure, incorporati...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-84555-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594679916396544 |
|---|---|
| author | Xinying Chen Zhisheng Jiao Yuefan Liu |
| author_facet | Xinying Chen Zhisheng Jiao Yuefan Liu |
| author_sort | Xinying Chen |
| collection | DOAJ |
| description | Abstract This paper proposes the YOLOv8n_H method to address issues regarding parameter redundancy, slow inference speed, and suboptimal detection precision in contemporary helmet-wearing target recognition algorithms. The YOLOv8 C2f module is enhanced with a new SC_Bottleneck structure, incorporating the SCConv module, now termed SC_C2f, to mitigate model complexity and computational costs. Additionally, the original Detect structure is substituted with the PC-Head decoupling head, leading to a significant reduction in parameter count and an enhancement in model efficiency. Moreover, the original Detect structure is replaced by the PC-Head decoupling head, significantly reducing parameter count and enhancing model efficiency. Finally, regression accuracy and convergence speed are boosted by the dynamic non-monotonic focusing mechanism introduced through the WIoU boundary loss function. Experimental results on the expanded SHWD dataset demonstrate a 46.63% reduction in model volume, a 44.19% decrease in parameter count, a 54.88% reduction in computational load, and an improvement in mean Average Precision (mAP) to 93.8% compared to the original YOLOv8 algorithm. In comparison to other algorithms, the model proposed in this paper markedly reduces model size, parameter count, and computational load while ensuring superior detection accuracy. |
| format | Article |
| id | doaj-art-34b178e0903f4a7783468adcc1b96505 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-34b178e0903f4a7783468adcc1b965052025-01-19T12:23:26ZengNature PortfolioScientific Reports2045-23222025-01-0115111610.1038/s41598-024-84555-1Improved YOLOv8n based helmet wearing inspection methodXinying Chen0Zhisheng Jiao1Yuefan Liu2School of Computer and Communication Engineering, Dalian Jiaotong UniversitySchool of Computer and Communication Engineering, Dalian Jiaotong UniversitySchool of Computer and Communication Engineering, Dalian Jiaotong UniversityAbstract This paper proposes the YOLOv8n_H method to address issues regarding parameter redundancy, slow inference speed, and suboptimal detection precision in contemporary helmet-wearing target recognition algorithms. The YOLOv8 C2f module is enhanced with a new SC_Bottleneck structure, incorporating the SCConv module, now termed SC_C2f, to mitigate model complexity and computational costs. Additionally, the original Detect structure is substituted with the PC-Head decoupling head, leading to a significant reduction in parameter count and an enhancement in model efficiency. Moreover, the original Detect structure is replaced by the PC-Head decoupling head, significantly reducing parameter count and enhancing model efficiency. Finally, regression accuracy and convergence speed are boosted by the dynamic non-monotonic focusing mechanism introduced through the WIoU boundary loss function. Experimental results on the expanded SHWD dataset demonstrate a 46.63% reduction in model volume, a 44.19% decrease in parameter count, a 54.88% reduction in computational load, and an improvement in mean Average Precision (mAP) to 93.8% compared to the original YOLOv8 algorithm. In comparison to other algorithms, the model proposed in this paper markedly reduces model size, parameter count, and computational load while ensuring superior detection accuracy.https://doi.org/10.1038/s41598-024-84555-1YOLOv8nSCConvCA attention mechanismPC-Head decoupling headWIoU |
| spellingShingle | Xinying Chen Zhisheng Jiao Yuefan Liu Improved YOLOv8n based helmet wearing inspection method Scientific Reports YOLOv8n SCConv CA attention mechanism PC-Head decoupling head WIoU |
| title | Improved YOLOv8n based helmet wearing inspection method |
| title_full | Improved YOLOv8n based helmet wearing inspection method |
| title_fullStr | Improved YOLOv8n based helmet wearing inspection method |
| title_full_unstemmed | Improved YOLOv8n based helmet wearing inspection method |
| title_short | Improved YOLOv8n based helmet wearing inspection method |
| title_sort | improved yolov8n based helmet wearing inspection method |
| topic | YOLOv8n SCConv CA attention mechanism PC-Head decoupling head WIoU |
| url | https://doi.org/10.1038/s41598-024-84555-1 |
| work_keys_str_mv | AT xinyingchen improvedyolov8nbasedhelmetwearinginspectionmethod AT zhishengjiao improvedyolov8nbasedhelmetwearinginspectionmethod AT yuefanliu improvedyolov8nbasedhelmetwearinginspectionmethod |