Scholarlink

Accurate and timely detection of rice crop diseases is crucial for maintaining agricultural productivity and ensuring global food security. While traditional diagnostic methods are widely used, they are often time-consuming and prone to human error. Recent advancements in deep learning, especially transfer learning and advanced color texture analysis of disease images, offer promising solutions. This study utilizes transfer learning models like ResNet50, MobileNetV2, and MobileNet to classify rice crop diseases effectively. To further boost performance, we introduce a novel hybrid MobileNet-ResNet Concate Model, which integrates MobileNet’s efficiency with ResNet’s robust feature extraction capabilities. Fine-tuned on a comprehensive dataset of rice crop images spanning various disease stages, the model demonstrates exceptional results, achieving a training accuracy of \(99.50\%\) and a test accuracy of \(94.00\%\) . Moreover, it excels in key metrics such as precision, recall, and F1-score. Comparative analysis highlights the hybrid model’s superiority over traditional transfer learning approaches regarding generalization, robustness, and computational efficiency. With its high accuracy and rapid processing capabilities, the proposed model offers a scalable and practical solution for precision agriculture. It paves the way for transformative improvements in rice disease management, crop health, and yield optimization.

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

Disease Detection and Classification of Rice Leaves Using Transfer Learning with Concatenate

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Anand Kumar Jain,
  • Dakshita Sharma,
  • Neeta Nain

摘要

Accurate and timely detection of rice crop diseases is crucial for maintaining agricultural productivity and ensuring global food security. While traditional diagnostic methods are widely used, they are often time-consuming and prone to human error. Recent advancements in deep learning, especially transfer learning and advanced color texture analysis of disease images, offer promising solutions. This study utilizes transfer learning models like ResNet50, MobileNetV2, and MobileNet to classify rice crop diseases effectively. To further boost performance, we introduce a novel hybrid MobileNet-ResNet Concate Model, which integrates MobileNet’s efficiency with ResNet’s robust feature extraction capabilities. Fine-tuned on a comprehensive dataset of rice crop images spanning various disease stages, the model demonstrates exceptional results, achieving a training accuracy of \(99.50\%\) and a test accuracy of \(94.00\%\) . Moreover, it excels in key metrics such as precision, recall, and F1-score. Comparative analysis highlights the hybrid model’s superiority over traditional transfer learning approaches regarding generalization, robustness, and computational efficiency. With its high accuracy and rapid processing capabilities, the proposed model offers a scalable and practical solution for precision agriculture. It paves the way for transformative improvements in rice disease management, crop health, and yield optimization.