非接触检测中螺纹图像牙型边界修正研究

Study on Revision of Thread Tooth Form Based on Non-Contact Measurement

摘要:

利用图像的非接触检测螺纹是一种高效的检测方法；检测中,精准的获取到螺纹图像的牙型边界是检测的关键,而现有的图像处理技术并不能消除由于螺纹旋线对螺纹图像的影响；针对螺纹图像中牙型边界被遮挡的问题,从圆锥螺纹的几何空间关系分析图像产生误差的原因,建立圆锥螺纹数学模型,得到误差的方程；最后再设计了图像检测系统,根据误差方程修正螺纹图像、提取参数,经验证,此方法可以提高螺纹的检测精度,且通用在圆锥螺纹与圆柱螺纹中。

Abstract:

Non-contact detection based on image is an efficient detection method, during the detected, to obtain the image of the tooth profile of the taper thread accurately is the key of the detection. The existing image processing technique can not eliminating the influences of helix line to the thread image. Based on the problem of the tooth profile are shaded, analyzed the reason of the errors from the geometry relation of the thread, then get the differential equation. At last, established a system for detecting the thread, according to the differential equation, revised the taper thread tooth, extracting parameters. After the validations, the new method can improve the precise of measurement based on image, be used in both taper thread and parallel screw thread