振动条件下的尼龙拉链厚度尺寸视觉在线测量算法

Visual On-line Measurement Algorithm for Thickness Dimension of Nylon Zipper under Vibration Condition

针对尼龙拉链厚度尺寸的视觉在线测量过程中,振动对测量结果影响较大的问题,通过对拉链振动的机理分析,建立并求解出被测拉链横向振动模型的运动方程,再结合LK(Lucas Kanade)光流法测量原理,求取出某一时刻的拉链全局振动信息。在此基础上,提出了一种振动条件下的尼龙拉链厚度尺寸的视觉在线测量算法。首先提取出待补偿图像中被测拉链的边缘点坐标,其次根据运动补偿算法修正边缘点坐标,接着利用最小二乘法将边缘点坐标拟合成平行直线并求出其间距,最后根据像素当量转换为拉链厚度的实际尺寸。实验结果表明:该算法测量结果的不确定度为0.00051 mm,远优于无运动补偿的视觉测量方法测量结果,能有效地抑制拉链振动对测量结果的影响。

Aiming at the vibration has a great influence on the measurement results during visual on-line measurement of nylon zipper s thickness dimension,through analyzed of the zipper vibration mechanism,the motion equation of the transverse vibration model of zipper under vibration condition was established and its definite solution was obtained.Then combined with the principle of LK optical flow method,the global vibration information of zipper at a certain moment was obtained.On this basis,an visual on-line measurement algorithm for thickness dimension of nylon zipper under vibration condition was proposed.Firstly,the coordinates of the measured zipper edge point in the image to be compensated were extracted.Secondly,the coordinates of edge points were modified according to the motion compensation algorithm.Then the coordinates of edge points were fitted into parallel lines by least square method and the spacing of the parallel lines was calculated.Finally,the spacing was converted to the actual thickness dimension of the zipper according to the pixel equivalent.The experimental results show that the uncertainty of the measurement results of this algorithm is 0.00051 mm,which is far better than the measurement results of the vision measurement method without motion compensation,which can effectively suppress the impact of zipper vibration on the measurement results.