To　help　the　management　of　stores,　especially　the　stores　that　provide　self-check　digital　services,　and　to　better　understand　and　study　the　object-detection　algorithm　based　on　You　Only　Look　Once,　used　many　ways　to　train　and　get　a　YOLO　model　that　satisfy　the　need　of　detecting　multiple　products　in　images.　The　RPC　dataset　is　used　to　train　the　model,　part　of　which　is　used　in　the　training　process　to　generate　a　final　version,　the　result　performance　of　the　model　generally　shows　the　YOLO　algorithm　has　its　strengths　when　faced　with　images　of　retail　goods.　©　2024　The　Authors.