Scholarlink

In this work hybrid transfer learning approaches were applied to accurately classify the tomato leaf diseases. VGG16, VGG19, and InceptionV3 architectures models were tuned up for transfer learning on GitHub external-defects-in-tomatoes dataset for diagnosis of tomatoes leaf disease. On this dataset applied SMOTE oversampling techniques for solve class imbalance problem. To assess the hybrid model’s performance according to recall, Precision, accuracy, and f1-score. A finding from experiments suggest that the hybrid model demonstrated an enhanced accuracy rate of 96.8%. In order to mitigate production losses, the suggested system may be implemented in tomato fields for an early detection of diseases. Subsequently, this model is expanded to include additional plant diseases and afflicted classes.

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 Investigation of Deep Learning Techniques for Diagnosis of Tomatoes Leaf Disease with Image Classification

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sakshi Tripathi,
  • L. K. Vishwamitra,
  • Sanjeev Kumar Sharma,
  • Manoj Tyagi,
  • Bhanu P. Singh

摘要

In this work hybrid transfer learning approaches were applied to accurately classify the tomato leaf diseases. VGG16, VGG19, and InceptionV3 architectures models were tuned up for transfer learning on GitHub external-defects-in-tomatoes dataset for diagnosis of tomatoes leaf disease. On this dataset applied SMOTE oversampling techniques for solve class imbalance problem. To assess the hybrid model’s performance according to recall, Precision, accuracy, and f1-score. A finding from experiments suggest that the hybrid model demonstrated an enhanced accuracy rate of 96.8%. In order to mitigate production losses, the suggested system may be implemented in tomato fields for an early detection of diseases. Subsequently, this model is expanded to include additional plant diseases and afflicted classes.