基于机器视觉的钢包头检测系统设计

Design of Steel Toecaps Inspection System Based on Machine Vision

为了降低工人劳动强度,提高检测速度和检测准确度,将机器视觉引入到钢包头在线检测,并设计了检测样机。首先分析钢包头变形特点,提出了基于8个变形敏感区域14项重要指标的检测模型和流程,重点介绍了具体图形算法,最后基于halcon软件编程测试了系统性能,并分析了运动对测量的影响、误差来源及其消除方法。实验结果表明:在线检测的最大误差小于0.2 mm,漏检率0‰,检测精度远高于人工检测,平均视觉检测时间为213.71 ms,整体检测速度约是人工检测的3.5倍,所设计系统可以满足流水线检测需要。

This paper presents a prototype online system for the visual inspection of steel toecaps. Using this system can not only reduce the labor intensity of workers, but also improve the inspection speed and accuracy. By analyzing the deform characteristics of the steel toecaps, an inspection method was put forward based on 14 important indexes of the 8 main deformation areas. The specific graphics algorithm was focused. The system performance was tested based on the Halcon software, and the influence of motion on measurement, error sources and their eliminating methods was analyzed.Experimental results show that the maximum error of the system is less than 0.2 mm and the missing rate of defective product is 0‰. The inspection accuracy is much higher than the manual inspection, and the average inspection speed is 213.71 ms,which is faster 3.5 times than manual inspection. The design of the steel toecaps online vision inspection system can meet the demand of industrial production.