Scholarlink

The domain of agriculture has witnessed a transformative shift with the advent of automated disease detection systems. This paper delves into the realm of employing deep learning techniques to detect rice leaf diseases efficiently. Our focus centers on four prominent deep learning architectures: Convolutional Neural Networks (CNN), VGG-19, LeNet, and MobileNet. These architectures stand as pillars in the pursuit of accurate classification across diverse rice leaf disease classes. The dataset encompassing nine distinctive disease categories and a healthy state forms the cornerstone of our study. The experimentation phase is marked by meticulous evaluation, with performance metrics such as accuracy, precision, recall, and the F1-score taking center stage. Each architecture’s prowess in capturing intricate patterns plays a significant role in shaping the classification landscape. In an era where technology’s role in agriculture burgeons, this study serves as a significant milestone. Moreover, our exploration paves the way for further research, fuelling the engine of innovation in automated disease detection for a secure and prosperous agricultural future.

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

Efficient Rice Leaf Disease Detection Based on Convolutional Neural Networks

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Gouranga Mandal,
  • Nageswara Sastry,
  • Nihal Ahmed,
  • Kashyap Jonnabathula

摘要

The domain of agriculture has witnessed a transformative shift with the advent of automated disease detection systems. This paper delves into the realm of employing deep learning techniques to detect rice leaf diseases efficiently. Our focus centers on four prominent deep learning architectures: Convolutional Neural Networks (CNN), VGG-19, LeNet, and MobileNet. These architectures stand as pillars in the pursuit of accurate classification across diverse rice leaf disease classes. The dataset encompassing nine distinctive disease categories and a healthy state forms the cornerstone of our study. The experimentation phase is marked by meticulous evaluation, with performance metrics such as accuracy, precision, recall, and the F1-score taking center stage. Each architecture’s prowess in capturing intricate patterns plays a significant role in shaping the classification landscape. In an era where technology’s role in agriculture burgeons, this study serves as a significant milestone. Moreover, our exploration paves the way for further research, fuelling the engine of innovation in automated disease detection for a secure and prosperous agricultural future.