Perangkat Lunak Pengenalan Jenis Roti Pada Toko Roti Dengan Menggunakan Model Deteksi Objek Yolov8

Putra, Ronaldo
Yoannita, Yoannita

MDP Student Conference
Vol 3 No 1 (2024): The 3rd MDP Student Conference 2024

Bakery shop is a business entity in the Food & Beverage sector that sells various types of sweet bread. The diversity of bread types poses a challenge for cashiers in recognizing one type of bread from another, leading to input errors in the cashier's computer. This research explores the use of YOLOv8 object detection to effectively identify different bread types. The dataset consists of 507 samples, including 164 Sosis Keju Brs, 157 Kulit Nangka Coklat, 103 Flosy Chicken, 75 Pizza Mini, and 80 Pizza Brs, trained over 100 epochs, resulting in a mean average precision (mAP) of 0.92%. The detection model performs well in recognizing bread types; however, it has some limitations, particularly when the bread is separated and not stacked on top of each other. A cashier software based on computer vision is also developed to assist cashiers in conducting transactions without the need for individual input.