Journal of Experimental Agriculture International

Bird pests cause significant agricultural losses by damaging crops during critical growth stages, resulting in reduced yield and economic instability for farmers. Traditional deterrent methods such as scarecrows, hawk kites, chemical repellents, flashing lights, and gun firing have shown limited effectiveness due to bird habituation and high labor dependency. To address these limitations, this study presents the development of a solar-powered automated bird repellent system integrating computer vision, IoT connectivity, and Arduino-based control. The proposed system consists of two primary modules: (i) a real-time bird detection module using an IP camera and machine learning-based image processing, and (ii) an automated repellent module that generates predator sounds through an MP3 playback system and megaphone upon detection. The system is powered by a solar panel with battery backup, ensuring uninterrupted field operation. Experimental validation was conducted in an open agricultural field over five consecutive days, comparing a control plot and a protected test plot. Results demonstrated a 74% reduction in bird visits in the protected area, with detection accuracy of 92%, a false alarm rate of 8%, and a response time of 2–3 seconds. The automated system significantly reduced manual labor (~90%) and provided approximately 75% crop protection efficiency within a 5 m radius coverage area. The findings confirm that the proposed IoT-based solution offers an effective, eco-friendly, and scalable approach for precision agriculture and sustainable crop protection.