摘要
针对目前对于大尺寸工件二维形貌测量效率低、精度低的问题,设计了一种基于机器视觉的大尺寸工件二维形貌检测系统,主要包括硬件系统设计和软件系统设计。针对工件尺寸大的特点,利用光电编码器精确定位相机位置,将大尺寸工件分解为多个部分进行采集,建立坐标转换数学模型,实现大尺寸工件的尺寸测量;针对通用的直线/圆弧检测算法精度低的问题,设计了一种基于改进随机抽样一致性(RANSAC)的亚像素直线/圆弧检测算法,可以得到精确的直线/圆弧方程。实验结果表明:设计的系统算法精度高,稳定性好。
Aiming at problems of low efficiency and low precision of two-dimensional( 2 D) morphology measurement of large size workpiece,a 2 D mophology measurement system of large size workpiece is designed based on machine vision. This system mainly includes hardware system design and software system design. The camera position is precisely located by optical-electricity encoder. The large size workpiece is divided into several parts to be captured,and coordinate transformation math model is constructed,size measurement of large size workpiece is realized. A subpixel straight line/arc detection algorithm based on improved random sample consensus( RANSAC) is designed,aiming at problem of low precision of normal linear/arc detection algorithm,which can get exact equation of the straight line/arc. Experimental results show that the system has high precision algorithm and good stability.
作者
冯西
吴静静
安伟
FENG Xi;WU Jing-Jing;AN Wei(School of Mechanical Engineering,Jiangnan University,Wuxi 214122,China;Key Laboratory of Advanced Food Manufacturing Equipment and Technology of Jiangsu Province,Wuxi 214122,China)
出处
《传感器与微系统》
CSCD
2019年第4期104-107,共4页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(61305016)
关键词
机器视觉
大尺寸
二维形貌测量
亚像素边缘检测
坐标转换
machine vision
large size
two-dimensional(2D)mophology measurement
subpixel edge detection
coordinate transformation
