<![CDATA[Potato productivity is often disrupted by pest attacks, leading to reduced yields. This research implements an IoT-based early warning system for pest detection and automated pesticide spraying on potato plants in a greenhouse. The system utilizes the YOLOv5 algorithm to detect armyworms and aphids using a mobile phone camera, supported by a NodeMCU microcontroller, soil moisture sensors, and the Blynk platform for real-time monitoring. A rule-based decision tree approach is used to regulate automatic spraying when the pest detection confidence level exceeds 70%. Field trials in Cibeureum Village, West Java, demonstrated a detection accuracy of 85% with a response time of 3–5 seconds. These results demonstrate the ability to reduce excessive pesticide use compared to manual methods and increase pest control efficiency. This smart farming system also provides real-time notifications via the Blynk app, facilitating remote monitoring and control by farmers. This implementation supports the concept of precision agriculture by optimally utilizing pesticides, maintaining productivity, and promoting sustainable agricultural practices.]]>
Copyrights © 2025
