摘要
为了改善图像分辨率在测量精度上的限制,从图像处理的角度实现高精度的视觉测量。利用开源计算机视觉库(OpenCV)为平台,采用USB摄像头作为视觉传感器,通过畸变矫正、图像预处理及亚像素检测等技术,构建了以单目相机为测量窗口的实时在线高精度视觉测量系统。测量三个已知标准尺寸的物体,其测量结果的最大绝对误差为0.082mm,最大相对误差为0.430%,由结果可得,该系统可以满足大多数机械加工行业实时在线检测及测量仪器要求。
In order to improve the limitation of visual measurement accuracy in resolution, high-precision visual measurement is realized from the perspective of image processing.The open source computer vision library(OpenCV) is used as a platform, USB camera is used as visual sensor.Through distortion correction, image preprocessing, sub-pixel detection and other technologies, a real-time online high-precision visual measurement system is constructed with monocular camera as the measurement window.And three objects of known standard dimensions are measured, the maximum absolute error of the result is 0.082 mm, and the maximum relative error is 0.430%,which meets the actual needs of later measurement.
作者
陈红
王代强
Chen Hong;Wang Daiqiang(School of Mechatronics Engineering,Guizhou Minzu University,Guiyang 550025,China)
出处
《工具技术》
北大核心
2022年第7期128-132,共5页
Tool Engineering
基金
贵州省科技计划项目(黔科合成果[2020]2Y074,黔科中引地[2020]4009)
贵州民族大学科研基金(GZMU[2019]YB21,GZMUZK[2021]YB06)
贵州省教育厅研究项目(黔科合KY字[2017]035)。
关键词
视觉测量
OPENCV
畸变矫正
图像预处理
亚像素检测
vision measurement
OpenCV
distortion correction
image preprocessing
sub-pixel detection
