Scholarlink

<p>Unmanned aerial vehicles (UAVs) and artificial intelligence have recently been used to revolutionize precision agriculture since agricultural crop health can now be monitored in real time. In this paper, the researcher suggests a new disease-detecting structure using drones to combine multispectral imaging and a convolutional neural network (CNN) model that assists in the effective detection of diseases in tea leaves. In contrast to the traditional human visual inspection technologies, the suggested system utilizes vegetation indices (NDVI, GNDVI) utilized in conjunction with automated image processing and classification pipelines to identify disease symptoms in their initial stages. CNN architecture presents more than 92% accuracy which shows that it is more efficient in differentiating diseased and healthy leaves in field conditions. The paper provides a scalable, data driven solution to sustainable agricultural management by providing an end-to-end adaptable algorithmic workflow applicable to a variety of crops and terrains.</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

AI-driven precision farming: UAV-based multispectral and CNN framework for tea leaf disease detection

  • 期刊论文

  • 发表日期：

    2025年12月27日

查看全文
  • Rupanjal Debbarma,
  • Nirmalya Das,
  • Aditya Sankar Sengupta

摘要

Unmanned aerial vehicles (UAVs) and artificial intelligence have recently been used to revolutionize precision agriculture since agricultural crop health can now be monitored in real time. In this paper, the researcher suggests a new disease-detecting structure using drones to combine multispectral imaging and a convolutional neural network (CNN) model that assists in the effective detection of diseases in tea leaves. In contrast to the traditional human visual inspection technologies, the suggested system utilizes vegetation indices (NDVI, GNDVI) utilized in conjunction with automated image processing and classification pipelines to identify disease symptoms in their initial stages. CNN architecture presents more than 92% accuracy which shows that it is more efficient in differentiating diseased and healthy leaves in field conditions. The paper provides a scalable, data driven solution to sustainable agricultural management by providing an end-to-end adaptable algorithmic workflow applicable to a variety of crops and terrains.