Scholarlink

This study aims to enhance the defect detection process for photovoltaic (PV) panels by reducing manual workload and improving operational efficiency. The proposed system employs YOLOv5, an advanced object detection algorithm capable of accurately identifying various PV panel defects. A mobile application, developed using the Android Studio platform, integrates the trained YOLOv5 model to enable real-time, on-site detection. Images of PV panels, captured via the camera embedded in the worker’s VR headset, are processed by the application to automatically identify and localize defects, with results displayed in real time on the worker’s mobile device. Furthermore, a web-based interface is developed to display real-time image data and detection results transmitted from the field via the VR headset, allowing remote personnel to monitor the condition of PV panels and respond promptly to identified issues. This approach not only improves the accuracy and timeliness of defect detection but also facilitates remote monitoring, thereby optimizing inspection and maintenance workflows.

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

Photovoltaic Defect Recognition Based on the YOLOv5 Object Detection Algorithm

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Shuhui Wen,
  • Kai Zheng,
  • Yaohui Chang,
  • Mengda Zhang,
  • Xudong Zhao,
  • Zheng Wang,
  • Yazhou Fan,
  • Shuhuan Wen,
  • Jiaqi Li

摘要

This study aims to enhance the defect detection process for photovoltaic (PV) panels by reducing manual workload and improving operational efficiency. The proposed system employs YOLOv5, an advanced object detection algorithm capable of accurately identifying various PV panel defects. A mobile application, developed using the Android Studio platform, integrates the trained YOLOv5 model to enable real-time, on-site detection. Images of PV panels, captured via the camera embedded in the worker’s VR headset, are processed by the application to automatically identify and localize defects, with results displayed in real time on the worker’s mobile device. Furthermore, a web-based interface is developed to display real-time image data and detection results transmitted from the field via the VR headset, allowing remote personnel to monitor the condition of PV panels and respond promptly to identified issues. This approach not only improves the accuracy and timeliness of defect detection but also facilitates remote monitoring, thereby optimizing inspection and maintenance workflows.