摘要
针对压缩机活塞圆度误差检测的要求,首先对采集图像进行灰度调整,然后进行中值平滑处理,使之有利于图像的边界提取操作.为解决圆拟合算法复杂、运算速度慢的问题,利用圆形的几何对称性质,提出圆拟合改进方法,通过运算处理,对提取的边界进行拟合,求出边界函数,进行拟合误差比较,从而判定其误差大小和位置.
Aming at the non -contact piston roundness error digital image detection demand, the paper first made gray image collection adjustments, then implemented midian smoothing for the image boundary extraction operation. For the complexity of the current circle fitting algorithm and the problem of slow processing speed, using circular geometric symmetry, the paper proposed circle fitting improved methods, found the boundary function by programming operations to deal with the extraction border, and compared fitting error to determine the size and location of the error.
出处
《河南理工大学学报（自然科学版）》
CAS
2010年第3期370-374,共5页
Journal of Henan Polytechnic University（Natural Science）
关键词
活塞式空压机
数字图像
最小二乘法
误差检测
piston compressor
digital Image
least square approach
error source judging
