IOT-Based Pest Detection in Maize Plants Using Machine Learning

Smart Farming is an agricultural technique that utilizes internet of things (IoT) technology and artificial intelligence to monitor plant development. Recently, smart farming has been widely used to monitor corn to improve the quality and quantity of production. However, pests are still a major problem in corn smart farming. This is because the existing corn smart farming still has low pest detection accuracy. This research provides a solution to the problem by developing an IOT-based intelligent pest detection system for corn smart farming. The system developed uses a machine learning algorithm for detecting a pest that is tracked through a camera connected to an IOT device. The camera works based on the trigger given by a Passive Infrared sensor (PIR). The camera will capture image if there is an input from PIR sensor. Images that are captured by the camera will be processed by a controller for pest detection. The pest detector controller uses a machine learning algorithm to track the presence of pests. Then, a buzzer (connected to the controller using the Message Queuing Telemetry Transport (MQTT) protocol) will sound if the controller successfully detects pests. Rigorous experiments on this pest detection system have been carried out, and it has been proven that the detection accuracy is up to 81%.