Scholarlink

<p>With the advancement of smart power grids, UAV-based insulator inspection has been widely applied. However, traditional detection systems face two significant challenges: the complex environment surrounding transmission lines complicates detection efforts and UAVs have limited computational resources. To address these issues, we propose a Lightweight and Adaptive Model of You Only Look Once v11 (LAM-YOLOv11) for UAV-based inspection. In this study, we proposed a lightweight network architecture, by using LA-C2K3 module, multiple feature maps of different scales are extracted from input images. Second, we incorporate a lightweight attention module to enhance detection performance with minimal parameter cost. Finally, an adaptive robustness module is designed to effectively mitigate environmental noise. We validated the model on multiple data sets. The experimental results demonstrate that this model outperforms other models in both server and UAV environments, and can meet the requirements of UAV inspection.</p>

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

LAM-YOLOv11 for UAV transmission line inspection: overcoming environmental challenges with enhanced detection efficiency

  • 期刊论文

  • 发表日期：

    2025年2月14日

查看全文
  • Yang Xuan,
  • Xiao-Yu Zhang,
  • Chen Li,
  • Hui Wang,
  • Chaoxu Mu

摘要

With the advancement of smart power grids, UAV-based insulator inspection has been widely applied. However, traditional detection systems face two significant challenges: the complex environment surrounding transmission lines complicates detection efforts and UAVs have limited computational resources. To address these issues, we propose a Lightweight and Adaptive Model of You Only Look Once v11 (LAM-YOLOv11) for UAV-based inspection. In this study, we proposed a lightweight network architecture, by using LA-C2K3 module, multiple feature maps of different scales are extracted from input images. Second, we incorporate a lightweight attention module to enhance detection performance with minimal parameter cost. Finally, an adaptive robustness module is designed to effectively mitigate environmental noise. We validated the model on multiple data sets. The experimental results demonstrate that this model outperforms other models in both server and UAV environments, and can meet the requirements of UAV inspection.