RANCANG BANGUN APLIKASI MOBILE BERBASIS YOLOv11 UNTUK DETEKSI PENYAKIT DAUN TANAMAN ANGGUR
DEVELOPMENT OF A YOLOv11-BASED MOBILE APPLICATION FOR GRAPE LEAF DISEASE DETECTION
Tanaman anggur merupakan komoditas populer yang bermanfaat bagi kesehatan, namun rentan terhadap penyakit daun seperti leaf blight, black rot, esca, dan downy mildew yang dapat merugikan petani. Penelitian ini bertujuan untuk mengembangkan sistem deteksi penyakit daun anggur dalam bentuk aplikasi seluler menggunakan model YOLOv11. Model dilatih dengan skema pembagian dataset yang berbeda (70:20:10, 80:10:10, dan 60:20:20) dengan empat optimizer Adam, AdamW, NAdam, dan RAdam. Hasil terbaik diperoleh pada skema 80:10:10 dengan optimizer NAdam, menghasilkan F1-Score 98,93%, precision 99,81%, recall 98,13%, dan mAP50-95 sebesar 91,63%. Pengujian aplikasi menunjukkan bahwa model mampu mendeteksi objek dalam keadaan tertentu sesuai skema pengujian. Penggunaan model YOLOv11 terbukti efektif dalam mendeteksi penyakit daun tanaman anggur dari dataset yang digunakan dan layak diimplementasikan pada perangkat seluler.Kata Kunci : Deteksi Penyakit Daun, Tanaman Anggur, Aplikasi Seluler, YOLOv11.
Grapevines are a widely cultivated horticultural commodity with various health benefits but are vulnerable to leaf diseases such as leaf blight, esca, and downy mildew, which can reduce productivity and harm farmers. This study aims to develop a grape leaf disease detection system in the form of a mobile application using the YOLOv11 model. The model was trained using three dataset split schemes (70:20:10, 80:10:10, and 60:20:20) and four optimizers: Adam, AdamW, NAdam, and RAdam. The best results were achived with 80:10:10 split scheme using the NAdam optimizer, producing an F1-Score of 98.93%, precision of 99.81%, recall of 98.13%, and mAP50-95 of 91.63%. Application testing demonstrated that the model successfully detected objects under various conditions according to the testing scheme. The YOLOv11 model proved in detecting grape leaf disease based on the dataset used and is suitable for implementation on mobile devices.Keyword : Leaf Disease Detection, Grapevine, Mobile Application, YOLOv11.