A novel and optimized IoT –ML based plant classification, monitoring and prediction system

Abstract

With the increasing demand for modern technologies and automation, there is a need to maintain plants by adopting such modern automated technologies. Every plant has some parameters that must be considered for its survival. Thus, the given paper proposes a novel and optimized system that enables plants to communicate with the user through the use of the Internet of Things (IoT). The parameters associated with the plants should be monitored and classified to ensure that the plants are healthy. In the proposed system, plant requirements are monitored with the help of several sensors and the IoT. The data is collected via sensors related to environmental conditions and sent to an Android application on the user's smart phone. Following this, the collected data is used to classify whether the plant is healthy or unhealthy. The classification is done by using a machine learning classifier named Random Forest (RF), and the experimental results show that the proposed framework is able to achieve higher accuracy (89.85%), higher precision (88.37%), and higher recall (86.55%) than the existing classifiers used in the recent studies.