基于机器视觉的扁平细长带钩零件检测系统研发

Development of Inspection System for Flat and Slender Parts with Hook Based on Machine Vision

某种扁平细长带钩零件的形状不规则、结构精细,但一致性要求极高,人工检测其变形的效率低、精准度差,为此研发了一种基于机器视觉的检验系统。首先通过设计环面自动移位系统提高零件移位速度,采用高精度相机和远心镜头获得高质量图像;其次利用模板匹配算法定位目标,对模板进行形态学操作,提取轮廓ROI作为检验标准,将其经旋转、平移变换拟合到零件图像上,用逐点扫描法检验轮廓一致性。系统测试表明:优化的算法设计具备过滤脏物颗粒及噪声的能力,在70枚/min的检测速度下,多次试验误检率为0,系统稳定可靠。

Some flat and slender parts with irregular shape and fine structure,but the consistency requirements are extremely high.The efficiency and accuracy of manual inspection are low.Therefore,an inspection system based on machine vision has been developed.Firstly,the efficiency of the inspection was improved by designing a toroidal transmission system,and the high-precision camera and telecentric lens were selected to obtain high-quality images.Secondly,the template matching algorithm was used to locate the target,morphological operation was carried out on the template,and the ROI of the contour was extracted as the inspection standard,which was fitted to the part image through rotation and translation transformation to check the contour consistency.System tests show that the optimized algorithm design has the ability to filter particles and noise.At the inspection rate of 70 pieces/min,the false detection rate is 0,and the system is stable and reliable.