Scholarlink

Agriculture contributes substantially to the economy of every country. Identification of plant conditions is the very essential rudiments for maintaining a country with overall developed agricultural economy. A healthy and productive farming sector, and the prevention of economical along with other resource by-product, depend on the prompt and effective discovery of plant conditions. Crop farmers lose a significant amount of capitalist each time due to various kinds of plant conditions. By early discovery of conditions in plant leaves, deep knowledge can be truly helpful to farmers in preventing crop failure. In order to identify crop infection, different plants (paddy, potato and citrus leaves) are analysed using CNN, VGG-16, and VGG-19, ResNet-50 model on dataset of 9112 plant-village photos. The delicacy rates of CNN, VGG-16, and VGG-19 also ResNet-50 were 93.60, 94.39, 97.15, and 96.98, singly. With a delicacy of 97.98, according to the study, ResNet-50 outperforms all other models. The ResNet model was chosen to be transformed into an intelligent online operation in order to assist with crop conditions in real time. By the analysis of plant flake images, the proposed web tool needs to help farmers identify plant conditions.

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

Helping Farmers in Identifying Diseases of Plants Using Machine Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Surendra Kumar Keshari,
  • Kanishka Sanserwal,
  • Divya,
  • Sakshi Aggarwal,
  • Veepin Kumar

摘要

Agriculture contributes substantially to the economy of every country. Identification of plant conditions is the very essential rudiments for maintaining a country with overall developed agricultural economy. A healthy and productive farming sector, and the prevention of economical along with other resource by-product, depend on the prompt and effective discovery of plant conditions. Crop farmers lose a significant amount of capitalist each time due to various kinds of plant conditions. By early discovery of conditions in plant leaves, deep knowledge can be truly helpful to farmers in preventing crop failure. In order to identify crop infection, different plants (paddy, potato and citrus leaves) are analysed using CNN, VGG-16, and VGG-19, ResNet-50 model on dataset of 9112 plant-village photos. The delicacy rates of CNN, VGG-16, and VGG-19 also ResNet-50 were 93.60, 94.39, 97.15, and 96.98, singly. With a delicacy of 97.98, according to the study, ResNet-50 outperforms all other models. The ResNet model was chosen to be transformed into an intelligent online operation in order to assist with crop conditions in real time. By the analysis of plant flake images, the proposed web tool needs to help farmers identify plant conditions.