基于机器视觉的螺纹参数测量系统

发布时间：2018-08-16 08:20

【摘要】：螺纹件在机械装备中有着广泛的应用,随着精密机械的发展,不仅要求螺纹联接可靠,使用时间长,而且要求螺纹联接的部件之间要有较高的配合精度,例如在空间技术和武器装备中,因此实现螺纹的高精度测量有重要的实用价值。近些年,基于机器视觉的技术得到飞速发展,在螺纹测量领域得到广泛应用。在国内外机器视觉测量螺纹中,大部分存在测量范围较小,不能够获得螺纹整体的轮廓坐标,螺纹遮挡失真等问题。针对螺纹参数测量问题,本课题以CCD相机为核心器件,通过相机采集螺纹轮廓数据,坐标计算的方法,建立了基于机器视觉的螺纹参数测量系统,能够实现大范围,高精度,快速的测量。本课题主要进行了如下研究:(1)根据投影法的测量原理,建立了螺纹参数测量平台,采用螺纹工件旋转、相机跟随拍摄的方式采集螺纹在平行背光下的轮廓数据,扩大了系统的测量范围,能够对多种尺寸的螺纹进行测量。(2)相机采集螺纹牙型轮廓图像,中值滤波实现图像的平滑,最大类间方差法分割图像,Canny算子提取牙型的整数精度轮廓,二次曲线拟合的方法实现牙型边缘的亚像素定位,得到轮廓的精确定位信息。(3)推导了坐标系转换公式,根据螺纹图像的几何特征,提取螺纹的特征点,设计了螺纹大径、小径、中径、螺距和牙型角的测量算法,实现螺纹的各个参数测量。通过转台旋转获得螺纹轮廓数据,经过三维坐标转换实现三维重构。(4)根据空间几何原理分析了螺纹投影图像的遮挡失真现象,推导了螺纹真实牙型边界方程和遮挡曲线方程,建立了失真模型。分析了牙型失真对各个螺纹参数测量的影响,实现了失真造成中径测量误差的补偿。根据螺纹测量原理及过程,开发了系统软硬件。实验结果证明,系统能够完成螺纹的各项参数测量。
[Abstract]:Thread parts are widely used in machinery and equipment. With the development of precision machinery, it is required not only to have reliable thread connection and long service time, but also to have high matching precision between the parts of thread connection. For example, in space technology and weapon equipment, it has important practical value to realize high precision measurement of thread. In recent years, machine vision based technology has been developed rapidly and widely used in thread measurement. In the domestic and foreign machine vision measurement thread, most of them have some problems, such as small measuring range, unable to obtain the overall contour coordinates of the thread, thread occlusion distortion and so on. In order to solve the problem of thread parameter measurement, a thread parameter measurement system based on machine vision is established by using CCD camera as the core device and the method of collecting thread contour data and coordinate calculation by the camera, which can achieve large range and high precision. A quick measurement. The main contents of this paper are as follows: (1) according to the measuring principle of projection method, the thread parameter measurement platform is established. The thread workpiece is rotated and the camera follows the camera to capture the contour data of the thread in parallel backlight. The measuring range of the system is enlarged, and the thread of various sizes can be measured. (2) the contour image of thread tooth is collected by camera, the median filter is used to smooth the image, and the Canny operator is used to extract the contour of tooth shape by Canny operator. The conic curve fitting method realizes the sub-pixel location of the tooth shape edge, and obtains the accurate location information of the contour. (3) the coordinate system transformation formula is derived. According to the geometric features of the thread image, the feature points of the thread are extracted, and the thread diameter and diameter are designed. The measurement algorithm of center diameter, pitch and tooth angle is used to measure the parameters of thread. The thread contour data are obtained by rotating the rotary table, and the 3D reconstruction is realized through the 3D coordinate transformation. (4) according to the spatial geometry principle, the occlusion distortion phenomenon of the thread projection image is analyzed, and the boundary equation and occlusion curve equation of the true thread tooth are derived. The distortion model is established. The influence of tooth shape distortion on the measurement of thread parameters is analyzed, and the compensation of center diameter measurement error caused by distortion is realized. According to the principle and process of thread measurement, the software and hardware of the system are developed. The experimental results show that the system can measure the parameters of thread.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TG85;TP391.41

【相似文献】