The Study of Sheet Metal Part Recognition Using Computer Vision Technology

• Main Authors: WENG, YI-CHIUN, 翁羿群
• Other Authors: SHEU, RUEY-KAI
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/4xueex

碩士 === 東海大學 === 資訊工程學系 === 107 === In traditional industry, to recognize a metal parts rely on manual identified the difference between layout and parts. After long time working, orders or markers used to lost due to environment of the factory or human behavior, it lead to decreasing the efficiency of the factory. For the purpose of increasing productivity and improve error rate, we use technique based on computer vision, image recognition and deep learning to carry out a metal sheets recognition system and distribute each parts to the correct station. Our research build an architecture which enable to expand as much data deal with factory can produce, we put a framework to transfer traditional factory into semi-automated production line.