IoT-Based Agriculture Farm Monitoring System and Development of Plant Leaf Disease Detection Using Deep Neural Network

Throughout history, agriculture or farming has played a vital role in human civilization by providing people's fundamental necessities and significant contributions to a country's economy. Although small-scale smart farming has started in a few nations, it has yet to be adopted scientifically globally. The rapid growth of the world population along with urbanization has rendered modern agriculture increasingly essential day by day. Towards multi-domain modernization, the Internet of Things (IoT) technology is a developing paradigm that aims to link many smart physical parameters in smart agriculture. In this paper, an IoT-based smart agricultural farm monitoring system prototype has been proposed which has been designed considering different environmental parameters. Additionally, image processing-based plant disease detection model has been developed in two different methods. The system incorporates various sensors, actuators, and a microcontroller to monitor and control temperature, humidity, soil moisture, and rainfall. It also includes a water pump motor for automated irrigation where a Wi-Fi module allows remote monitoring and controlled by a developed smartphone app. To validate, the proposed hardware model was simulated using Electronic Design Automation (EDA) tools. There have been notable outcomes observed on temperature, humidity, soil moisture, and other monitoring parameters data reading from LCD. Also, deep learning-based image processing algorithm has been proposed using Feature extraction and Convolutional Neural Network (CNN) methods and achieved a high accuracy of about 96%-98% in plant disease detection. The overall proposed system offers improved resource utilization and smart agricultural practices for expanded functionalities and long-term sustainability to ensure food security for the growing population.