针对视觉测量中硬盘圆孔孔径容易受到圆孔周围局部强反射、外部噪声的影响和测量精度不高的问题, 提出一种基于小波变换、数学形态学和机器视觉相结合的圆形零件孔径测量方法。对摄像头采集圆孔图片, 通过小波变换分解出高低频图像, 利用小波变换对高频部分进行边缘提取, 数学形态学对低频部分进行边缘提取, 然后进行小波边缘融合, 获取有效的圆孔边缘。利用最小二乘法对边缘进行孔径尺寸计算, 经实验验证, 该算法与其他算法相比, 在有效地保留圆孔周围边缘信息的情况下, 对周围噪声进行了有效的抑制, 检测精度为001 mm以内, 实验结果表明此方法简单易行, 且具有较高的精度。

In order to solve the problem that hole diameter of hard disk is easy to be affected by local strong reflection and external noise,and low accuracy of measurements,a method for measuring the diameter of circular parts based on wavelet transform,mathematical morphology and machine vision is proposed.Hole image is collected by camera,high/low-frequency images are resolved through wavelet transform,and wavelet transform is used to extract edge of high frequency part and mathematical morphology is used to extract low frequency part,then wavelet edge fusion is used to obtain effective hole edge.Finally, the least square method is utilized to calculate pore size. It is verified experimently that,compared with other algorithms,in the case of effectively preserving edge information of round hole and suppressing surrounding noise effectively,detection accuracy is controlled within 0.01 mm.Experimental results show that this method is simple and feasible,and has a high accuracy.