Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206675,
        author = {Prof. Daya Naik and Dhanyashree and Pratham P Shetty and Preethi C G and Vrushali A Poojary},
        title = {Autonomous AI-Driven Drone System for Arecanut Yield Estimation and Disease Classification},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {no},
        pages = {144-150},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206675},
        abstract = {Growing arecanut at scale is harder than it looks. Farmers spend days walking through dense plantations trying to gauge how healthy the crop is, how much yield to expect, and whether disease has begun to take hold — often without reliable answers. This paper describes a system we built to change that. We developed an autonomous drone-based framework that flies over arecanut plantations, captures aerial imagery, and uses deep learning to detect inflorescences, classify disease, and estimate yield — all without requiring manual inspection on the ground. At the core of the system is a U-Net segmentation model that pinpoints inflorescence regions at the pixel level, working alongside two YOLOv8 detection models that count individual fruits, assess ripeness, and flag disease. A Tello drone ties it all together: it streams live video, identifies inflorescences in real time, flies toward them, and captures the high-resolution frames needed for detailed analysis. A FastAPI backend handles the heavy lifting on the server side, while a Streamlit dashboard gives users a clear, interactive view of everything the system finds. In our experiments, the framework performed consistently well — accurately segmenting clusters, counting fruits, detecting disease, and generating reliable yield estimates across varied field conditions. We believe this kind of integrated, automated solution can genuinely ease the burden on farmers and help them make better decisions with real data.},
        keywords = {Precision agriculture, autonomous drones, arecanut yield estimation, semantic segmentation, object detection, deep learning},
        month = {July},
        }