基于结构光图像的轴径检测方法研究

发布时间：2019-03-29 15:22

【摘要】：结构光视觉测量技术具有非接触、易于计算机集成化的优点,在机械工业中受到了高度的关注,应用越来越广泛。提高结构光测量机械零件的精度是发展与应用结构光测量技术的关键。本文以阶梯轴校直中轴颈径向尺寸在线测量为目标,研究线结构光光条图像测量点检测优化新方法,提高激光线结构光测量机械零件尺寸的精度。根据结构光三角测量原理,本文将同时位于光平面和轴颈表面上的点的成像位置作为要检测的光条图像测量点。首先,在包含激光光源和阶梯轴的摄像机坐标系里,建立光平面与轴颈相交的交线方程;接下来根据小孔成像关系,研究将光平面与轴颈的交线方程变换到理论图像平面的方法;然后利用变换后的交线方程,沿光条图像宽度方向,通过非线性方程迭代求解检测光条图像测量点;最后实验定性检验轴颈表面粗糙度和转速等因素对新方法检测精度的影响,并与数显卡尺的测量结果对比,检验新方法测量轴径的精度。本文重点研究了一种基于结构光视觉技术在线测量轴径的方法。根据被测对象轴表面与光平面在空间上的交线为椭圆的一部分,建立了描述椭圆交线的世界坐标系,并根据已经标定好的摄像机参数、轴线方程和结构光光平面方程,推导了世界坐标系与摄像机坐标系的转换关系。把椭圆交线方程和摄像机成像模型结合起来,推导出光平面与轴表面的椭圆交线在图像平面的方程。应用几何形状约束法,筛选原始光条中心位置坐标数据,测量得到轴径尺寸初始值,然后在图像平面坐标系对椭圆中心位置和半轴长度做搜索优化,当径向偏差平方和最小时,解算出测量模型中的轴半径作为最终测量值。创新点:利用小孔成像关系变换到理论图像平面的交线方程,检测光条图像测量点。
[Abstract]:Structured light vision measurement technology has the advantages of non-contact and easy computer integration. It has been paid more and more attention in the mechanical industry and has been used more and more widely. Improving the precision of structural light measuring machinery is the key to the development and application of structured light measuring technology. In this paper, aiming at on-line measurement of radial dimension of axis neck in step alignment, a new method of detecting and optimizing the measuring point of line structured light strip image is studied in order to improve the precision of measuring the size of mechanical parts by laser line structured light. According to the principle of structured light triangulation, the imaging position of the point on both the light plane and the journal surface is regarded as the measuring point of the strip image. Firstly, in the camera coordinate system including laser source and step axis, the intersecting line equation between light plane and axis is established. Then, according to the imaging relationship of small holes, the method of transforming the intersection equation between light plane and journal into the theoretical image plane is studied. Then, using the transformed intersecting line equation, along the width direction of the light strip image, the measurement point of the detection strip image is solved iteratively by the nonlinear equation. Finally, the effects of the surface roughness and rotational speed of the journal on the accuracy of the new method are tested qualitatively, and compared with the measured results of the digital caliper, the accuracy of the new method for measuring the diameter of the shaft is tested. In this paper, an on-line measurement method of axis diameter based on structured light vision technology is studied. According to the intersection between the axis surface and the light plane of the measured object is part of the ellipse, the world coordinate system describing the intersection line of the ellipse is established. According to the camera parameters, the axis equation and the structured light plane equation, the camera parameters have been calibrated, the axis equation and the structured light plane equation have been set up. The transformation relationship between the world coordinate system and the camera coordinate system is deduced. Combining the elliptic intersection equation with the camera imaging model, the equation of the ellipse intersecting line between the light plane and the axis surface in the image plane is derived. The geometric shape constraint method is used to select the coordinate data of the center position of the original light strip, measure the initial value of the axis diameter, and then search and optimize the center position and half axis length of the ellipse in the image plane coordinate system, when the sum of square of the radial deviation is the smallest. The axis radius in the measurement model is calculated as the final measurement value. Innovation: using hole imaging relationship to transform to the intersection equation of theoretical image plane, detect the measuring point of light strip image.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【相似文献】