机器视觉的薄片零件尺寸检测系统

对基于机器视觉的薄片零件尺寸检测系统进行研究。针对计算机硬盘弹性臂薄片零件的尺寸检测,采用线阵工业相机扫描获取待检零件图像,根据扫描特性提出了标定算法,同时提出了一种轮廓矢量化算法。将经过滤波、标定、二值化、边缘提取和细化的图像进行轮廓矢量化得到待检零件的尺寸参数,并与从设计图DXF文件中读取的零件的相应尺寸参数进行对比,判断零件制造质量。实测结果表明,检测系统的轮廓矢量化精度达到1 pixel,检测精度达到1 μm,平均每个零件的检测时间为1 s,该检测系统是可行的。

Dimensional inspecting system of thin sheet parts based on machine vision

According to the dimensional inspection,taking the thin sheet part of Head Gimbal Assembly (HGA) of the computer hard disk as an object, a line type industrial camera is used to acquire the image of the inspected part by line scanning. The calibration algorithm is proposed based on the property of the line scanning, and a contour vectorization algorithm is developed. The dimensional parameters of inspecting part are gained adopting the contour vectorization algorithm in which the image has been processed by filtering, calibration, binarization, edge detection and thinning. The parametric errors are obtained to evaluate the quality of a thin sheet part comparing the dimensional parameters obtained from scanned image to the dimensional parameters obtained from reading the design DXF file. The experimental results indicate that the precision of contour vectorization can reach to 1 pixel, the inspection accuracy can reach to 1 μm, the average inspection time of every part is 1 s and the inspecting system is feasible.