Automation and integration of growth monitoring in plants (with disease prediction) and crop prediction

Abstract

In the age of climate change, even the farmers who possess indigenous knowledge face difficulties in making judicious decisions on crop health monitoring that leads to the failure of the crop which in turn results in the decline of crop production. Another reason for the decline in crop production can be the selection of unsuitable crops for cultivation and the inability of identifying the visible effect of the disease in a plant. The research paper deals with how the IoT technology helps in collecting information about conditions like temperature, humidity, pH, and rainfall by applying various machine learning algorithms to obtain the output. The research work deals with Decision Tree to predict the crop condition and suggests a suitable solution, it also suggests which crop can be grown. Using a public dataset of 87,000 RGB images of diseased and healthy plant leaves, we train a convolutional neural network to identify diseases that are categorized into 38 different classes. This paper consists of a proposed model that takes real-time data for predicting the kind of crop which is the most suitable and will monitor the crop conditions in the field through weather analysis and crop disease diagnosis on a massive scale for better production.