Scholarlink

Tomatoes, the world’s most popular crop, can be utilized in a variety of ways in any kitchen, regardless of cuisine. A number of diseases, such as Mosaic virus, Early Blight, Septoria leaf and Bacterial spot, have damaged the quality and abundance of the tomato crop. The ability of any plant to develop swiftly and produce more crops is dependent on its leaves. Finding ailments in plant leaves can be difficult for both farmers and researchers. Deep learning algorithms can be used to identify and categorise different diseases caused to tomato plant. The popular deep learning models for identifying tomato plant diseases are VGG16 and denseNet201.These models performs well on training data with an accuracy of 91.84% and 95.39% but their performance decreases while testing on new data. The proposed method is a concatenation hybrid model which is made up of denseNet201 and VGG16 trained on different data sets and verified on new data sets. The results show that the proposed model is working with efficiency on new data sets too.

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

An Efficient Classification of Tomato Plant Diseases Using Deep Learning

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Durga Venkata Sai Vysyaraju,
  • Dhanunjaya Pamu,
  • Siva Kumar Patheda,
  • Ravindra Babu Rayapati,
  • Jaya Ram Suru,
  • M. V. Nageswara Rao

摘要

Tomatoes, the world’s most popular crop, can be utilized in a variety of ways in any kitchen, regardless of cuisine. A number of diseases, such as Mosaic virus, Early Blight, Septoria leaf and Bacterial spot, have damaged the quality and abundance of the tomato crop. The ability of any plant to develop swiftly and produce more crops is dependent on its leaves. Finding ailments in plant leaves can be difficult for both farmers and researchers. Deep learning algorithms can be used to identify and categorise different diseases caused to tomato plant. The popular deep learning models for identifying tomato plant diseases are VGG16 and denseNet201.These models performs well on training data with an accuracy of 91.84% and 95.39% but their performance decreases while testing on new data. The proposed method is a concatenation hybrid model which is made up of denseNet201 and VGG16 trained on different data sets and verified on new data sets. The results show that the proposed model is working with efficiency on new data sets too.