基于机器视觉提高实习安全性的金工实习教学改革探索

金工实习是高等院校工科教育中重要的工程训练环节之一,在提高学生工程实践能力上发挥着不可取代的作用。而大小事故的发生一直是影响金工实习教学质量最重要的因素之一,也是当前确保实习安全亟待解决的问题之一。该文提出了一种将机器视觉应用于金工实习的教学改革方案,并通过YOLO v3和U-Net神经网络初步实现金工实习过程中刀具和零件的识别。通过该方案,可以利用机器视觉对不规范操作、危险操作及时发出警报,引导学生正确操作,提高金工实习的安全性。

Exploration on reform of metalworking practice teaching based on machine vision to improve safety of practice

Metalworking practice is one of the important engineering training links in engineering education in colleges and universities,which plays an irreplaceable role in improving students’engineering practice ability.The occurrence of accidents has always been one of the most important factors affecting the quality of metalworking practice teaching and also one of the problems to be solved to ensure the safety of practice.In this paper,a teaching reform scheme of applying machine vision to metalworking practice is proposed,and the recognition of cutting tools and parts in the process of metalworking practice is preliminarily realized by using YOLO v3 and U-Net neural network.Through this scheme,machine vision can be used to give an alarm in time for irregular operation and dangerous operation,guide students to operate correctly and improve the safety of metalworking practice.