Permasalahan utama yang sering terjadi pada proses checkout di kasir adalah sulitnya barcode produk terbaca akibat kerusakan, lipatan, atau tertutup kemasan, sehingga memperlambat pelayanan. Untuk mengatasi hal tersebut, penelitian ini mengimplementasikan algoritma You Only Look Once versi 8s (YOLOv8s) sebagai solusi deteksi produk ritel berbasis visi komputer. Data penelitian berupa citra lima produk ritel yang diambil langsung di supermarket, dengan fokus pada objek yang terlihat minimal 70–85%. Model dilatih, divalidasi, dan diuji menggunakan metode deep learning dengan evaluasi performa berdasarkan precision, recall, dan mean Average Precision (mAP). Hasil pengujian menunjukkan nilai precision sebesar 0,89, recall 0,97, dan mAP50 rata rata 0,995 serta mAP50–95 sebesar 0,919. Nilai tersebut menunjukkan bahwa YOLOv8s mampu mendeteksi produk secara akurat dan konsisten, serta efektif dalam mengatasi keterbatasan barcode. Dengan demikian, algoritma ini dapat menjadi solusi untuk mempercepat proses checkout sekaligus meningkatkan efisiensi pelayanan ritel.

A common problem in retail checkout is the inability of barcode scanners to read damaged or folded barcodes, which causes delays in customer service. To address this issue, this study implements the You Only Look Once version 8s (YOLOv8s) algorithm as a computer vision-based product detection solution. The dataset consists of images of five retail products collected directly from a supermarket, focusing on objects with at least 70–85% visibility. The model was trained, validated, and tested using deep learning methods, and its performance was evaluated based on precision, recall, and mean Average Precision (mAP). Experimental results indicate a precision of 0.89, recall of 0.97, and an average mAP50 of 0.995 with mAP50–95 reaching 0.919. These findings demonstrate that YOLOv8s is highly accurate and reliable in detecting retail products, effectively overcoming barcode limitations. Therefore, this algorithm can serve as a practical solution to accelerate checkout processes and improve retail service efficiency.