딥러닝 기반 객체 검출 기술을 이용한 환봉 자동 라벨 부착 시스템

Abstract

In this paper, we propose a system that automatically attaches a label to the tip of a round bar using camera-robot coordinate transform technology and deep learning-based object detection technology to solve the problems of existing rule-based machine vision technology. According to our experiments, when the method proposed in this project was applied to the automatic round bar attachment system, it was confirmed that the label attachment accuracy was within 2 mm in most cases.