Scholarlink

In this paper, we will see how good a convolutional neural network (CNN) works in identifying various diseases in a tomato plant and then classifying them accordingly by analyzing the images of its leaves. The specific deep learning model used here for this approach is DenseNet-121 (i.e., Dense Convolutional Neural Network-121), which is very efficient in image classification as it uses 121 layers of modification for each input, i.e., image, in order to train itself. Two open-source datasets, ‘PlantVillage’ and ‘TomatoLeaf’, are combined together to be used as the final dataset in training and validating the model. The overall classification precision achieved by the proposed model reaches an approximate value of 99.71%, which shows how effective and efficient the DenseNet-121 model can be in disease detection and classification for a tomato plant. This also shows that traditional methods like manual inspections should upgrade themselves in adapting to such deep learning methods in saving time and energy for disease detection in tomato plants with much better accuracy.

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

A Deep Learning Approach in Tomato Leaf Disease Identification and Classification using DenseNet-121 model

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Babul Sarkar,
  • Kuldip Kaityar

摘要

In this paper, we will see how good a convolutional neural network (CNN) works in identifying various diseases in a tomato plant and then classifying them accordingly by analyzing the images of its leaves. The specific deep learning model used here for this approach is DenseNet-121 (i.e., Dense Convolutional Neural Network-121), which is very efficient in image classification as it uses 121 layers of modification for each input, i.e., image, in order to train itself. Two open-source datasets, ‘PlantVillage’ and ‘TomatoLeaf’, are combined together to be used as the final dataset in training and validating the model. The overall classification precision achieved by the proposed model reaches an approximate value of 99.71%, which shows how effective and efficient the DenseNet-121 model can be in disease detection and classification for a tomato plant. This also shows that traditional methods like manual inspections should upgrade themselves in adapting to such deep learning methods in saving time and energy for disease detection in tomato plants with much better accuracy.