Desenvolvimento de modelos de visão computacional e aprendizado de máquina para balança inteligente de baixo custo.

Abstract

This work proposes the development of a computer vision and machine learning system applied to a low-cost smart scale. The system is designed to automate the process of object identification and weighing, using image processing techniques to capture and analyze the visual characteristics of objects placed on the scale. The main focus of this study is the development of the artificial intelligence models that will integrate with the scale. Several machine learning algorithms were analyzed through performance metrics, resulting in the selection of the Random Forest model, which presented the best results. The model was validated by comparing it with the YOLO framework, widely recognized for its efficiency in object detection.