Scholarlink

<p>To achieve intelligent and precise management of tea plantation cultivation, and to address the issue of rapid and accurate identification of tea diseases in natural environments, this paper proposes a tea disease detection model based on YOLOv11n. First, the MCAttn module is introduced to improve the C3K2 module, enhancing the model’s ability to extract features from multiple scales. Second, BIFPN is used instead of the original FPN + PAN structure to improve the model’s efficiency. Finally, the loss function is improved to reduce the problem of class imbalance in model recognition. Experimental results show that MBS-YOLOv11 achieves an mAP of 91.7%, representing an improvement of 2.9 percentage points compared to the baseline YOLOv11. This algorithm achieves a balance between lightweight design and real-time performance, and significantly improves the accuracy of tea disease detection. It provides an ideal solution for real-time disease detection in tea plantations.</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

MBS-YOLOv11: a lightweight network for real-time detection of tea plant diseases

  • 期刊论文

  • 发表日期：

    2026年4月2日

查看全文
  • Song Yang,
  • Demin Huang,
  • Dayang Liu,
  • Liangkuan Zhu

摘要

To achieve intelligent and precise management of tea plantation cultivation, and to address the issue of rapid and accurate identification of tea diseases in natural environments, this paper proposes a tea disease detection model based on YOLOv11n. First, the MCAttn module is introduced to improve the C3K2 module, enhancing the model’s ability to extract features from multiple scales. Second, BIFPN is used instead of the original FPN + PAN structure to improve the model’s efficiency. Finally, the loss function is improved to reduce the problem of class imbalance in model recognition. Experimental results show that MBS-YOLOv11 achieves an mAP of 91.7%, representing an improvement of 2.9 percentage points compared to the baseline YOLOv11. This algorithm achieves a balance between lightweight design and real-time performance, and significantly improves the accuracy of tea disease detection. It provides an ideal solution for real-time disease detection in tea plantations.