摘要
采用图像处理技术实现对零件中的圆形特征进行非接触式测量。该系统通过CMOS镜头采集被测零件,进行灰度变换、中值滤波,采用Canny算子提取边缘,然后利用最小二乘法对被测零件中的圆形特征进行拟合,计算得到最小二乘法拟合圆,同时采用最小二乘法评定被测零件的圆度。实验结果表明,所提出的图像测量方法测量误差小,能够满足被测零件的一般精度测量要求,适用于测量精度不高的在线测量和接触式测量方法无法完成的测量任务。
Image processing technology is used to measure circular feature in the inspected part. The image of a circular parts is captured by CM OS camera lens,treated by gray-scale transformation and median filtering. The edge of the image is extracted with Canny. And the circle curve is fitting by least square. At last,the radius and roundness of curve is calculated. Experimental results indicate that the error of this measurement method is suitably small and it can suffice for the requiring of ordinarily measurement precision,and it is applicable for some on-line measurement without requiring of high precision. And it is also applicable for the measuring task in which the contact measuring methods are noneffective.
出处
《组合机床与自动化加工技术》
北大核心
2015年第7期74-76,79,共4页
Modular Machine Tool & Automatic Manufacturing Technique
关键词
图像处理
图像预处理
非接触测量
最小二乘法
digital image processing
image pre-processing
non-contact measurement
least squares
