IOT based Automatic Irrigation System with Plant Disease Recognition and Weather Monitoring

Abstract

Over the past two decades, there has been rapid growth in terms of technology in the agricultural sector. Yield rates can be reduced by several factors. This includes the issue of over or under irrigation and the need for dynamic irrigation based on soil properties and humidity and temperature conditions. In this we use the Raspberry Pi 3, soil moisture sensor, relay and pump for automatic irrigation purposes. Disease is another major contributing factor to the deterioration of yields. Photos of plant leaves are captured and processed and compared to diseased plant leaf images in the database. Based on the comparison, a list of suspected plant diseases is provided to the person using the CNN classifier as a probability. Continuous weather monitoring will contribute to better yields of crops and rainfall forecast. Automatic irrigation will automatically stop if rain is detected. There is a provision for violation of automatic functions if required and the irrigation can be done manually and the irrigation is automatically stopped if rainfall is detected. A website has been developed to control these functions for the ease of the user.