Scholarlink

This chapter explores the integration of Internet of Things (IoT) technology and deep learning models to enable real-time monitoring and disease detection in greenhouse environments. It outlines two primary objectives: developing an affordable IoT-based monitoring system and implementing deep learning algorithms for plant disease identification. Utilizing sensors connected to cost-effective IoT platforms, such as Arduino, essential parameters-including temperature, humidity, light intensity, and soil moisture-are continuously monitored to maintain optimal growing conditions. The seamless data transmission capabilities of IoT enhance the real-time processing of this information, allowing for automatic control of actuators to ensure an ideal greenhouse environment and mitigate problems like rotten roots. A deep learning model, built on the DenseNet121 architecture and trained on a comprehensive plant disease dataset, is embedded within Node-RED to facilitate immediate disease identification, achieving an impressive accuracy of 97.11% across 16,012 images spanning ten classes. This integrated system not only promotes crop health and increases yield but also offers a scalable and economically viable solution for future agricultural practices.

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

Real-Time Monitoring and Disease Detection in Greenhouses Using IoT and Deep Learning

  • 章节

  • 发表日期：

    2025年

查看全文
  • Sweta Jain,
  • Reenu Rajpoot

摘要

This chapter explores the integration of Internet of Things (IoT) technology and deep learning models to enable real-time monitoring and disease detection in greenhouse environments. It outlines two primary objectives: developing an affordable IoT-based monitoring system and implementing deep learning algorithms for plant disease identification. Utilizing sensors connected to cost-effective IoT platforms, such as Arduino, essential parameters-including temperature, humidity, light intensity, and soil moisture-are continuously monitored to maintain optimal growing conditions. The seamless data transmission capabilities of IoT enhance the real-time processing of this information, allowing for automatic control of actuators to ensure an ideal greenhouse environment and mitigate problems like rotten roots. A deep learning model, built on the DenseNet121 architecture and trained on a comprehensive plant disease dataset, is embedded within Node-RED to facilitate immediate disease identification, achieving an impressive accuracy of 97.11% across 16,012 images spanning ten classes. This integrated system not only promotes crop health and increases yield but also offers a scalable and economically viable solution for future agricultural practices.