Scholarlink

With the ongoing wave of digital transformation, automating service processes has become essential for improving both service quality and operational efficiency. This study introduces a YOLO-based model designed for monitoring car wash procedures in automotive service workshops. In this study, we propose Y4CWM, a real-time object detection system based on You Only Look Once (YOLO), a single-stage model designed for fast and accurate detection. Y4CWM enables automated recognition of different car wash stages, thereby supporting real-time service tracking, enhancing operational workflows, and ensuring consistent quality control. Image and video data were collected from the workshop’s surveillance cameras, annotated, and used to train models including YOLOv8, YOLOv9, YOLOv10, YOLOv11, YOLOv12, and Faster R-CNN. Experimental evaluation on a dataset of 7354 labeled images demonstrates that YOLOv11l achieves the highest mAP 50:95 of 0.94339, outperforming both other YOLO variants and Faster R-CNN, while lightweight models such as YOLOv8n also offer promising accuracy with significantly lower computation costs. The system Y4CWM not only helps standardize the car wash process but also improves the overall customer experience and optimizes service operations.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Y4CWM: A Real-Time YOLO-Based Deep Learning Framework for Smart Car Wash Stage Detection and Monitoring

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Vu Nguyen,
  • Luan N. T. Huynh,
  • Tham Vo

摘要

With the ongoing wave of digital transformation, automating service processes has become essential for improving both service quality and operational efficiency. This study introduces a YOLO-based model designed for monitoring car wash procedures in automotive service workshops. In this study, we propose Y4CWM, a real-time object detection system based on You Only Look Once (YOLO), a single-stage model designed for fast and accurate detection. Y4CWM enables automated recognition of different car wash stages, thereby supporting real-time service tracking, enhancing operational workflows, and ensuring consistent quality control. Image and video data were collected from the workshop’s surveillance cameras, annotated, and used to train models including YOLOv8, YOLOv9, YOLOv10, YOLOv11, YOLOv12, and Faster R-CNN. Experimental evaluation on a dataset of 7354 labeled images demonstrates that YOLOv11l achieves the highest mAP 50:95 of 0.94339, outperforming both other YOLO variants and Faster R-CNN, while lightweight models such as YOLOv8n also offer promising accuracy with significantly lower computation costs. The system Y4CWM not only helps standardize the car wash process but also improves the overall customer experience and optimizes service operations.