摘要
设计采用十字线结构光的双相机视觉测量系统,构建由2个高精度工业相机、2个十字激光器和环形光源等组成的图像采集装置.在传动箱加工现场获取端面销孔图像,提取图像感兴趣区域(ROI),采用双边滤波进行降噪预处理、二值化和边缘检测;基于改进方向模板法提取2个十字线结构光的中心线.依据测量系统的设计原理分别测出2个销孔的孔径及其位置度.结果表明:对于孔径,机器视觉在线测量装置与电子塞规的测量结果的偏差平均值为0.001 mm;对于孔组位置度,机器视觉在线测量装置与三坐标测量仪的测量结果的最大偏差为0.02 mm.这表明机器视觉在线测量装置能满足加工精度要求和在线实时测量需求.
A two-cameras vision measurement system was designed using cross-hair structured light;an image acquisition device was constructed, which consisted of two high-precision industrial cameras, two cross lasers, and a ring light source. The image of the end pinhole at the transmission box processing site was obtained;the image region of interest(ROI) was extracted;bilateral filtering was used for noise reduction preprocessing, binarization and edge detection. The center lines of two cross structured light were extracted based on improved direction template method. According to the design principle of the measuring system, the diameter and position degree of two pin holes were measured, respectively. Results show that, for diameter of pin hole, the average deviation of the measurement results between the machine vision online measureement device and the electronic plug gauge is 0.001 mm;for hole set position degree, the maximum deviation of the measurement results between the machine vision online measuring device and the coordinate measuring device is 0.02 mm. The proposed machine vision online measuring device can meet the requirements of processing accuracy and online real-time measurement.
作者
李瑛
成芳
赵志林
LI Ying;CHENG Fang;ZHAO Zhi-lin(College of Biosystems Engineering and Food Science,Zhejiang University,Hangzhou 310058,China)
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2020年第3期557-565,613,共10页
Journal of Zhejiang University：Engineering Science
基金
国家重点研发计划资助项目(2017YFD0700205).
关键词
改进方向模板法
结构光
大跨度
加工精度
在线测量
机器视觉
图像处理
improved direction template method
structured light
large span
machining precision
on-line measurement
computer vision
image processing
