Programa del congreso

This study presents the development of a machine vision system based on the YOLOv11 model for the automatic detection of burnt alfajor wafers during the post-baking stage. The aim is to implement an intelligent inspection tool that optimizes quality control by automating visual classification. A dataset was constructed using images captured under controlled conditions and labeled into two
categories: normal product and burnt product. The model was trained for 70 epochs, achieving an accuracy of 96.8%, sensitivity of 94.5%, F1 score of 95.6%, mAP50 = 1.00, and mAP50–95 = 0.87. The results demonstrate robust performance and adequate generalizability, proving the feasibility of integrating artificial intelligence techniques into quality control systems for the food industry, reducing manual intervention and improving operational efficiency.