CD Skripsi
Deteksi Benih Padi Menggunakan Metode You Only Look Once (Studi Kasus: Upt Psbtph Provinsi Riau)
Seeds circulating in Indonesia are required to have seed eligibility certification as regulated by PERMENTAN No. 12/2018. One of the processes to obtain seed eligibility certification is the purity analysis laboratory test. In this process, seed analysts sort out seed work samples that contain several components, such as pure seeds and non-pure seeds. The sorting is done by identifying the seeds in the seed working sample according to their morphological appearance. In this research, object detection of rice seeds was conducted using the "You Only Look Once (YOLO)" algorithm. YOLO is an algorithm that performs real-time object detection based on Convolutional Neural Network (CNN). The method used is transfer learning pre-trained model YOLOv5s which is one of the models of YOLO version 5. The data used is image data as many as 235 photos of rice seed work samples with two-class labeling, namely rice for pure seed objects and non-rice for objects that are not pure seeds. The evaluation results of the overall model performance are quite good with a precision value of 0.908, recall 0.808, [email protected] 0.859, and F1_Score 0.85. Keywords: Rice Seed, Object Detection, You Only Look Once, YOLOv5
Tidak tersedia versi lain