An Intelligent IoT Framework for the Identification of Nutrition Value in Crops Using Convolutional Neural Networks

Abstract

Consumption of a balanced diet that meets all the nutritional requirements is very important especially at all stages of human life. Insufficiency in meeting these body requirements lead to serious illnesses and organ collapse leading to significant health conditions in adulthood. In this context, early assessment of nutrition content in crops leads to educate the consumers to make healthy food. To address the challenge, this study develops a self-assessing nutrition monitoring system based on the new Internet of Things (IoT) to enhance agricultural yield. The artificial intelligence assessment of the crops to monitor its nutritional content is important to ensure healthy and complete growth. The objective of this study is to design a self- an automized system that will detect the nutritional deficiencies in crops by scanning the images of leaves of the crops. Convolutional neural networks (CNN) are used to further process the images. This technique compares the captured image with the readily available dataset. Results of the deficiency are obtained when the captured image partially or completely matches with the already present data set images. The result is shown in the form of percentage values. This approach will be highly beneficial to farmers by ensuring crop productivity and decreased labor. Simulation results report that the proposed system is highly beneficial when compared to the existing monitoring systems.