基于图像的模切机纸张边缘检测定位技术研究  

Research on the technology of die-cutting machine's paper edge detection and position based on image

作　　者：吴建成[1] 曹冬梅[1] 吴绍锋[1] 李东波[1] 

机构地区：[1]南京理工大学机械工程学院,江苏南京210094

出　　处：《机械设计与制造工程》2016年第12期33-36,共4页Machine Design and Manufacturing Engineering

基　　金：江苏省前瞻性联合研究项目(BY2015004-05)

摘　　要：为进一步提高模切纸张精度,在模切机叼纸机构前加装一套纸张定位装置。该装置基于机器视觉原理,在模切机的前规、侧规处安装3部微型相机,运用图像处理的方法粗略定出纸张边缘点位置,然后采用一维灰度矩法对边缘点进行亚像素级边缘检测,再通过最小二乘法直线拟合出纸张边缘所在直线的方程,计算出3幅图像中各纸张偏离零位位置的距离值。根据3个偏移值指导叼纸机构进行适当微调,可使模切精度达到0.02mm。In order to further improve the die- cutting precision of die- cutting machine,a paper positioning system is installed in front of the paper gripper mechanism. The system includes three micro cameras at the front lay and side lay,adopting image processing to position paper edge points roughly based on the principle of machine vision. And then,the edge points are detected at sub- pixel level with the one dimensional gray moment method. Furthermore,the straight line equation of paper edge is fitted by the method of least square method,which can be used to calculate the distance from the zero position of the paper in the three images. Guide the paper gripper mechanism with appropriate fine- tuning according to the three offset values,bringing the die- cutting precision at 0. 02 millimeter.

关 键 词：模切精度 图像处理 一维灰度矩法 亚像素边缘检测 最小二乘法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]