技术领域technical field
本发明涉及一种大型零件的单目视觉测量方法，属于测量技术领域。The invention relates to a monocular vision measurement method for large parts, belonging to the technical field of measurement.
背景技术Background technique
近年来大尺寸的曲面在汽车工业、船舶和航天器外形等方面得到广泛的应用，随着现代加工制造和生产作业的发展需要，大型零件三维几何尺寸的测量已成为现代逆向工程和产品数字化设计及制造的基础支撑技术，而且越来越多的装配、质量控制、在线检测、工业产品的工装定位等领域也迫切需要解决大型零件的三维测量。In recent years, large-sized curved surfaces have been widely used in the automotive industry, ship and spacecraft shapes, etc. With the development of modern processing and manufacturing and production operations, the measurement of three-dimensional geometric dimensions of large parts has become an important part of modern reverse engineering and product digital design. And the basic support technology of manufacturing, and more and more fields such as assembly, quality control, online inspection, tooling positioning of industrial products, etc. also urgently need to solve the three-dimensional measurement of large parts.
随着计算机技术、电子学、光学技术的日趋完善以及图像处理、模式识别等技术的不断进步，计算机视觉测量技术得到快速发展，已逐渐成为大型零件表面三维信息最主要的测量手段。目前基于单目视觉的大型零件测量技术主要有：几何相似法测量、几何形状约束法测量、结构光法测量、几何光学法测量和辅助靶标测量，其中只有辅助靶标测量可以实现空间不可见点的测量。With the improvement of computer technology, electronics, and optical technology, as well as the continuous advancement of image processing, pattern recognition and other technologies, computer vision measurement technology has developed rapidly and has gradually become the most important means of measuring three-dimensional information on the surface of large parts. At present, the measurement technologies for large parts based on monocular vision mainly include: geometric similarity measurement, geometric shape constraint measurement, structured light measurement, geometric optics measurement, and auxiliary target measurement. Among them, only auxiliary target measurement can realize spatially invisible points. Measurement.
靶标上一般设计有具有鲜明特征的标记来产生标志点，按照标志点发光与否，靶标可分为无光源靶标和有光源靶标两类，其中无光源靶标是利用靶标上的特制图案产生标志点，通常为获得理想标志点图像，还需使用特定光源照射靶标，此种靶标受环境影响较大；有光源靶标利用发光体，例如LED产生标志点，传统有光源靶标通过二值法确定光点区域，利用重心法或椭圆拟合法提取标志点中心，由于图像的二值化处理以及标志点不同角度成像，通过重心法或椭圆拟合法所提出的光点中心并非对应于空间中同一点，使得测量精度降低。Targets are generally designed with distinctive marks to generate mark points. According to whether the mark points emit light or not, targets can be divided into two types: non-light source targets and light source targets. Among them, the non-light source targets use special patterns on the target to generate mark points. , usually in order to obtain an ideal mark point image, it is necessary to use a specific light source to irradiate the target, which is greatly affected by the environment; a light source target uses a luminous body, such as an LED to generate a mark point, and a traditional light source target determines the light point through a binary method region, using the center of gravity method or ellipse fitting method to extract the center of the mark point, due to the binarization of the image and the imaging of the mark point at different angles, the center of the light point proposed by the center of gravity method or the ellipse fitting method does not correspond to the same point in space, making Measurement accuracy is reduced.
专利号为CN200910058832，提供了一种基于相位标靶的光学三坐标测量方法，该测量方法的测量原理是摄像机获取相位标靶中特征图像屏上的特征图像，通过相移条纹分析法或傅立叶条纹分析法计算出相位分布，建立标靶电子显示屏上各点与摄相机像素点之间的对应关系，进而确定标靶测头触点的三维空间坐标，通过移动相位标靶对被测物体表面进行多点测量，以计算出物三维面形，这种相位标靶用于光学三维测量时，与有3个以上标记点的辅助标靶相比较，由于特征点数量的大量增多，以及基于相位计算的特征点精确提取，使其测量结果更为精确和可靠，但是由于摄像机的一次测量视场有限，决定了对于某些大型工件进行测量时，摄像机和被测零件保持一个相对位置不可能将所有零件上待测点测量完毕，这大大限制了测量的范围和此系统对于大型零件的应用。The patent number is CN200910058832, which provides an optical three-coordinate measurement method based on a phase target. The measurement principle of the measurement method is that the camera acquires the characteristic image on the characteristic image screen of the phase target, and uses the phase shift fringe analysis method or Fourier fringe The analysis method calculates the phase distribution, establishes the corresponding relationship between each point on the electronic display screen of the target and the pixel point of the camera, and then determines the three-dimensional space coordinates of the touch point of the target probe. Perform multi-point measurement to calculate the three-dimensional surface shape of the object. When this phase target is used for optical three-dimensional measurement, compared with the auxiliary target with more than three marking points, due to the large increase in the number of feature points and the phase-based The calculated feature points are accurately extracted to make the measurement results more accurate and reliable. However, due to the limited field of view of the camera for one measurement, it is impossible to maintain a relative position between the camera and the measured part when measuring some large workpieces. The points to be measured on all parts are measured, which greatly limits the range of measurement and the application of this system to large parts.
发明内容Contents of the invention
本发明的目的是提供一种大型零件的单目视觉测量方法，以解决目前单目视觉测量过程中测量范围有限导致无法将大型零件中所有待测点测量完毕的问题。The purpose of the present invention is to provide a monocular vision measurement method for large parts, so as to solve the problem that all the points to be measured in the large parts cannot be measured due to the limited measurement range in the current monocular vision measurement process.
本发明为解决上述技术问题而提供一种大型零件的单目视觉测量方法，该测量方法通过辅助靶标上的转站球实现摄像机的转战测量，所使用的辅助靶标包括一测杆，测杆上方转动装配有转站球，该转站球表面上均匀布设有标记孔，该测杆上还固定安装有一能随测杆转动的测量棒，该测量棒的一个面上设置有标志点，所述转站球下方固定设置有刻有角度值的转站台；In order to solve the above-mentioned technical problems, the present invention provides a monocular vision measurement method for large parts. The measurement method realizes the measurement of the camera through the transfer station ball on the auxiliary target. The auxiliary target used includes a measuring rod. The rotating assembly is equipped with a transfer station ball, and the surface of the transfer station ball is uniformly provided with marking holes, and a measuring rod that can rotate with the measuring rod is fixedly installed on the measuring rod, and a marking point is set on one surface of the measuring rod. There is a fixed transfer station engraved with the angle value under the transfer station ball;
所述测量方法测量时，使辅助靶标上设置有标志点的一面正对摄像机，通过摄像机获取靶标标志点的特征图像信息，当摄像机在一个有效视场内测量完毕后，移动摄像机到另一个位置，保证转站球位姿不发生变化，同时旋转测杆使贴有标志点的一面正对摄像机，并记录测杆的旋转角度，同时通过转站球和转站球上每一个标记孔，得到转站后和转站前的相互关系，通过它们之间的相互关系将得到的若干组坐标值进行空间拼接以实现对大型零件形貌的重构。When the measurement method measures, make the side of the auxiliary target that is provided with the marker point face the camera, obtain the characteristic image information of the target marker point through the camera, and move the camera to another position after the camera is measured in an effective field of view , to ensure that the pose of the transfer ball does not change, and at the same time rotate the measuring rod so that the side with the marked point is facing the camera, and record the rotation angle of the measuring rod. The relationship between the post-transfer station and the front-transfer station, through the interrelationship between them, several sets of coordinate values obtained are spatially spliced to realize the reconstruction of the shape of large parts.
所述的测量方法包括以下步骤：Described measuring method comprises the following steps:
1）对摄像机视觉测量系统进行标定，确定相机内部参数和系统结构参数，对辅助靶标进行标定，确定在辅助靶标坐标系下转站球各孔中心、测棒上各标志点中心和测头坐标；1) Calibrate the camera visual measurement system, determine the internal parameters of the camera and system structure parameters, calibrate the auxiliary target, and determine the center of each hole of the transfer station ball, the center of each mark point on the measuring rod and the coordinates of the probe in the auxiliary target coordinate system ;
2）将辅助靶标测棒贴有标志点的一面正对摄像机，通过摄像机获取靶标标志点的特征图像信息；2) Face the side of the auxiliary target measuring stick with the marker points facing the camera, and obtain the characteristic image information of the target marker points through the camera;
3）对采集到的靶标标志点的特征图像信息进行图像处理得到各标志点中心像素坐标；3) Perform image processing on the collected characteristic image information of target marker points to obtain the center pixel coordinates of each marker point;
4）根据针孔成像原理建立系统测量模型，将标志点中心的像素坐标进行坐标转换得到测量标志点中心在世界坐标系中的坐标值，计算出辅助靶标坐标系到世界坐标系的旋转和平移矩阵，根据所计算出的旋转和平移矩阵以及测头中心在辅助靶标坐标系下的坐标，计算出测头中心的世界坐标值；4) Establish a system measurement model based on the principle of pinhole imaging, convert the pixel coordinates of the center of the marker point into coordinates to obtain the coordinate value of the center of the marker point in the world coordinate system, and calculate the rotation and translation from the auxiliary target coordinate system to the world coordinate system Matrix, according to the calculated rotation and translation matrix and the coordinates of the probe center in the auxiliary target coordinate system, calculate the world coordinate value of the probe center;
5）当摄像机在一个有效视场内测量完毕后，移动摄像机到另一个位置，开始另一位置的测量，直到将大型零件的所有曲面测量完毕；5) After the camera is measured in an effective field of view, move the camera to another position and start the measurement of another position until all the surfaces of the large parts are measured;
6）根据转站球上每一个唯一标识的孔在辅助靶标坐标系中的坐标值和像面坐标系中位置的相互关系，将得到的各组三维坐标值进行空间拼接，最终实现对大型零件物体三维形貌的重构。6) According to the relationship between the coordinate values of each uniquely identified hole on the transfer station ball in the auxiliary target coordinate system and the position in the image plane coordinate system, the obtained three-dimensional coordinate values are spatially spliced, and finally realize the large-scale parts Reconstruction of the three-dimensional shape of the object.
所述步骤2）在测量时，测棒轴线位于转站台0刻度处，且辅助靶标测棒贴有标志点一面正对摄像机，当辅助靶标上的刚体测量头与曲面待测点垂直接触时，测头上的开关同步控制摄像机来获取靶标标志点的特征图像信息，图像采集卡对图像进行采集。Step 2) During the measurement, the axis of the measuring rod is located at the 0 scale of the transfer platform, and the side of the auxiliary target measuring rod with a mark point is facing the camera. When the rigid body measuring head on the auxiliary target is in vertical contact with the point to be measured on the curved surface, The switch on the probe synchronously controls the camera to obtain the characteristic image information of the target mark point, and the image acquisition card collects the image.
所述步骤4）中辅助靶标坐标系到世界坐标系的变换公式为：The transformation formula from the auxiliary target coordinate system to the world coordinate system in step 4) is:
其中，(xf,yf,zf,1)T和(xw,yw,zw,1)T分别是P在辅助靶标坐标系和世界坐标系中的坐标；R′是由辅助靶标坐标系到世界坐标系的旋转矩阵，t′是由辅助靶标坐标系到世界坐标系的平移向量。Among them, (xf ,yf ,zf ,1)T and (xw ,yw ,zw ,1)T are the coordinates of P in the auxiliary target coordinate system and the world coordinate system respectively; The rotation matrix from the target coordinate system to the world coordinate system, and t' is the translation vector from the auxiliary target coordinate system to the world coordinate system.
所述步骤5）当摄像机在一个有效视场内测量完毕后，移动摄像机到另一个位置，保证转站球位置不发生变化同时旋转辅助靶标的测棒使贴有标志点的一面正对摄像机，开始另一位置的测量，实现转站测量。Step 5) After the camera is measured in an effective field of view, move the camera to another position to ensure that the position of the transfer ball does not change, and at the same time rotate the measuring rod of the auxiliary target so that the side with the mark point is facing the camera. Start the measurement of another location to realize the transfer station measurement.
所述步骤1）对摄像机视觉测量系统进行标定时采用辅助靶标完成摄像机的标定，将摄像机与辅助靶标相距一定距离固定好，打开CCD摄像机电源；在摄像机视场范围内，选取量块的顶点作为标定点，每移动一个位置拍摄一幅图像，将获得的标志点的二维图像信息通过网络数据线传送并保存到计算机中；利用提取的所有位置的标志点图像坐标及其对应的已知世界坐标，带入小孔成像测量系统模型中，进而完成对摄像机内外参数的求解，并保存到系统参数文件中，以备测量阶段调用。Step 1) When calibrating the camera visual measurement system, the auxiliary target is used to complete the calibration of the camera, the camera and the auxiliary target are fixed at a certain distance, and the power of the CCD camera is turned on; within the field of view of the camera, the apex of the gauge block is selected as Calibrate the point, take an image every time a position is moved, and transmit the obtained two-dimensional image information of the mark point to the computer through the network data line; use the extracted image coordinates of all the mark points and their corresponding known world The coordinates are brought into the model of the small hole imaging measurement system, and then the internal and external parameters of the camera are solved, and saved in the system parameter file for calling in the measurement stage.
本发明的有益效果是:本发明通过采用摄像机转站的方法对待测器件的最优视点进行规划分析，克服了摄像机一次测量的有限性，扩大了测量的范围，当摄像机在一个有效视场内测量完毕后，移动摄像机到另一个位置，保证转站球位姿不发生变化，同时旋转测杆使贴有标志点的一面正对摄像机，并记录测杆的旋转角度，同时通过转站球和转站球上每一个标记孔，可以得到转站后和转站前的相互关系，通过它们之间的相互关系可以将若干组坐标值进行空间拼接，实现对大型零件三维形貌的重构。The beneficial effects of the present invention are: the present invention plans and analyzes the optimal view point of the device to be tested by adopting the method of camera transfer station, overcomes the limitation of one-time measurement of the camera, expands the scope of measurement, when the camera is in an effective field of view After the measurement is completed, move the camera to another position to ensure that the pose of the transfer ball does not change. At the same time, rotate the measuring rod so that the side with the marked point is facing the camera, and record the rotation angle of the measuring rod. For each marked hole on the transfer ball, the relationship between the post-transfer and the front of the transfer can be obtained. Through the relationship between them, several sets of coordinate values can be spliced in space to realize the reconstruction of the three-dimensional shape of large parts.
附图说明Description of drawings
图1是本发明所使用的单目视觉测量系统的结构图；Fig. 1 is the structural diagram of the monocular vision measuring system used in the present invention;
图2是本发明所使用的辅助测量靶标的结构示意图；Fig. 2 is a schematic structural view of the auxiliary measurement target used in the present invention;
图3是本发明的单目视觉测量方法的坐标转换示意图；Fig. 3 is a schematic diagram of coordinate conversion of the monocular vision measurement method of the present invention;
图4是摄像机坐标系和像素坐标系的关系示意图；Fig. 4 is a schematic diagram of the relationship between the camera coordinate system and the pixel coordinate system;
图5是辅助坐标系示意图；Fig. 5 is a schematic diagram of the auxiliary coordinate system;
图6是本发明所使用的辅助测量靶标系统标定模型示意图。Fig. 6 is a schematic diagram of the calibration model of the auxiliary measurement target system used in the present invention.
具体实施方式detailed description
下面结合附图对本发明的具体实施方式作进一步的说明。The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings.
本发明的一种大型零件的单目视觉测量方法的实施例Embodiment of a monocular vision measurement method for large parts of the present invention
本发明的大型零件的单目视觉测量方法通过合理设置标志点精确提取出标志点中心坐标，通过辅助靶标上的转站球实现摄像机的转战测量，扩大了测量的范围，实现了对大型零件的快速、低成本的测量要求。本发明所使用的测量系统如图1所示，包括摄像机6、计算机和辅助测量靶标，其中辅助测量靶标如图2所示，包括测杆1，测杆1上转动装配有转站球2，转站球2下方固定安装有转站台3，测杆1上还固定安装有能随测杆1旋转的测棒4，测棒3的一个面上设置有标志点，测杆1下方设置有测头5。The monocular vision measurement method for large parts of the present invention accurately extracts the central coordinates of the mark points by reasonably setting the mark points, and realizes the measurement of the camera through the transfer station ball on the auxiliary target, expands the scope of measurement, and realizes the measurement of large parts. Fast, low-cost measurement requirements. The measuring system used in the present invention is as shown in Figure 1, comprises video camera 6, computer and auxiliary measuring target, and wherein auxiliary measuring target is shown in Figure 2, comprises measuring rod 1, and rotating station ball 2 is equipped with on measuring rod 1, A transfer platform 3 is fixedly installed under the transfer station ball 2, and a measuring rod 4 that can rotate with the measuring rod 1 is also fixedly installed on the measuring rod 1. A mark point is set on one surface of the measuring rod 3, and a measuring point is arranged under the measuring rod 1. head 5.
转站球2为一球体，转站球2为一球体，其外表面上按照一定角度均匀的在经线和纬线的相交位置上布设有标记孔，标记孔为低于球体表面的圆柱孔G1,G2,…,Gn，各标记孔的轴线与球心相交，各标记孔上分别粘贴各种颜色不同的彩色回光反射圆形薄贴，各标记孔在辅助靶标中的位置由颜色标识唯一确定，转转站台3上刻有360度均匀编码的角度，用于显示转战测量时测杆的旋转的角度，测棒3为一纯黑色长方体，其中一个表面上设置有尺寸信息已知的16个圆形回光反射辅助标志点N1,N2,…,N16和4个测量用标志点P1,P2,P3,P4，每4个辅助标志点呈长方形分布，分别处于长方形的4个顶点，16个辅助标志点分成4个长方形分布区域，4个测量用标志点分别处于上述4个长方形区域对角线的交点上，每个标志点设置有一种高折射率的玻璃粒子，反射强度高，与黑色测棒形成鲜明对比，易于与背景光源相分离，以形成清晰而突出的二值图像，测棒有标志点的一面与转站台相切处为起始位置，测量时，手持测棒两侧使贴有标志点的一面正对摄像机6。The transfer station ball 2 is a sphere, and the transfer station ball 2 is a sphere, and its outer surface is uniformly arranged with a marking hole at the intersection of the longitude and latitude according to a certain angle. The marking hole is a cylindrical hole G1 lower than the surface of the sphere ,G2 ,…,Gn , the axis of each marked hole intersects with the center of the sphere, each marked hole is pasted with various color retro-reflective circular thin stickers, and the position of each marked hole in the auxiliary target is determined by the color The logo is uniquely determined. The 360-degree uniform coded angle is engraved on the turntable platform 3, which is used to display the rotation angle of the measuring rod during the turnaround measurement. The measuring rod 3 is a pure black cuboid, and the size information is set on one of the surfaces. 16 circular retroreflective auxiliary marker points N1 , N2 ,...,N16 and 4 measurement marker points P1 , P2 , P3 , P4 , each of which is distributed in a rectangular shape, They are respectively located at the four vertices of the rectangle, and the 16 auxiliary mark points are divided into four rectangular distribution areas, and the four measurement mark points are respectively located at the intersection of the diagonal lines of the above four rectangular areas, and each mark point is set with a high refractive index Glass particles with high reflection intensity form a sharp contrast with the black measuring rod and are easy to separate from the background light source to form a clear and prominent binary image. The side of the measuring rod that is tangent to the transfer platform is the starting position , when measuring, hold both sides of the measuring rod so that the side with the marked point is facing the camera 6.
测头4具有良好的刚性和球度，能够灵活探测到各种内外表面上的点，测头4上安装有触发开关，以方便控制被测点的采样，一般选用球度高有高硬度陶瓷材料制成的的工业用红宝石，测量时测头紧紧靠住被测点，保证靶标在图像采集中的稳定性，测杆5的长度可以根据被测零件进行调节，为满足测量刚性要求，测杆长度越短越好，增加测杆长度会降低测量精度，但是对于有些难测的位置或者盲点，测杆长度可根据实际测量情况进行调节，特别适合飞机机翼或机身、汽车底盘或者车身、车间平台等物体的测量，综合考虑测杆的长度应该选择50-110mm。The probe 4 has good rigidity and sphericity, and can flexibly detect various points on the inner and outer surfaces. The trigger switch is installed on the probe 4 to facilitate the sampling of the measured points. Generally, ceramics with high sphericity and high hardness are selected. The industrial ruby made of material, when measuring, the measuring head is tightly close to the measured point to ensure the stability of the target in image acquisition. The length of the measuring rod 5 can be adjusted according to the measured part. In order to meet the measurement rigidity requirements, The shorter the length of the measuring rod, the better, increasing the length of the measuring rod will reduce the measurement accuracy, but for some difficult to measure positions or blind spots, the length of the measuring rod can be adjusted according to the actual measurement situation, especially suitable for aircraft wings or fuselage, automobile chassis or For the measurement of objects such as car bodies and workshop platforms, the length of the measuring rod should be selected to be 50-110mm.
本发明的使用上述辅助测量靶标的单目视觉测量方法的测量过程如下：The measurement process of the monocular vision measurement method using the above-mentioned auxiliary measurement target of the present invention is as follows:
1.首先建立坐标系，如图3所示，以摄像机的焦点为原点建立摄像机坐标系OcXcYcZc，将摄像机的光轴作为Zc轴，f是摄像机的焦距，它固定在摄像机上，遵循右手法则，将摄像机坐标当作世界坐标系，将摄像机光轴与像平面的焦点作为坐标原点Ｏ建立像平面坐标系ＯＸＹ，像素坐标系ＯＵＶ，如图4所示，其坐标原点在摄像机与像平面坐标系图像的左上角，以像素为单位，以辅助靶标转战球的球心为原点；如图5所示，建立辅助坐标系OfXfYfZf，转站球表面上孔的中心位置在辅助坐标系下的坐标为xfG,yfG,zfG，测量标志点中心在辅助坐标系下的坐标为（j=1,2,3,4），测头中心在辅助坐标系下的坐标为（xfN,yfN,zfN）；Ｐ为空间中的任意一点，ｐ为Ｐ在像平面上的成像点。从空间中任一点Ｐ投影到摄像机平面上ｐ，结合刚体变换知识，其变换过程如下：1. First establish a coordinate system, as shown in Figure 3, establish a camera coordinate system Oc Xc Yc Zc with the focal point of the camera as the origin, take the optical axis of the camera as the Zc axis, f is the focal length of the camera, and it is fixed On the camera, following the right-hand rule, the camera coordinates are regarded as the world coordinate system, and the focal point of the camera optical axis and the image plane is taken as the coordinate origin O to establish the image plane coordinate system OXY and the pixel coordinate system OUV, as shown in Figure 4, its coordinates The origin is at the upper left corner of the image in the camera and image plane coordinate system, with pixels as the unit, and the center of the ball where the auxiliary target is transferred to the ball as the origin; as shown in Figure 5, establish the auxiliary coordinate system Of Xf Yf Zf , and transfer The coordinates of the center position of the hole on the surface of the sphere in the auxiliary coordinate system are xfG , yfG , zfG , and the coordinates of the center of the measurement mark point in the auxiliary coordinate system are (j=1,2,3,4), the coordinates of the probe center in the auxiliary coordinate system are (xfN , yfN , zfN ); P is any point in space, p is the position of P on the image plane imaging point. Projecting from any point P in space to the camera plane p, combined with the knowledge of rigid body transformation, the transformation process is as follows:
世界坐标系相对于摄像机坐标系的变换公式：The transformation formula of the world coordinate system relative to the camera coordinate system:
(xc,yc,zc,1)T和(xw,yw,zw,1)T分别是P在摄像机坐标系和世界坐标系中的坐标；Ｒ是由世界坐标系到摄像机坐标系的旋转矩阵，t是由世界坐标系到摄像机坐标系的平移向量。摄像机坐标系到摄像机像平面坐标系的变换公式：(xc ,yc ,zc ,1)T and (xw ,yw ,zw ,1)T are the coordinates of P in the camera coordinate system and the world coordinate system respectively; R is from the world coordinate system to the camera The rotation matrix of the coordinate system, t is the translation vector from the world coordinate system to the camera coordinate system. The transformation formula from the camera coordinate system to the camera image plane coordinate system:
其中，(x,y,1)T是p在摄像机像平面坐标系的坐标。Among them, (x, y, 1)T is the coordinate of p in the camera image plane coordinate system.
摄像机平面坐标系到像素坐标系的变换公式为：The transformation formula from the camera plane coordinate system to the pixel coordinate system is:
其中，(u,v,1)T是p在像素坐标系下的坐标；dx、dy分别是像素坐标系中每个像素在Ｘ轴和Ｙ轴方向上的物理尺寸，(u0,v0)是物理坐标系的原点Ｏ在像素坐标系中的坐标。综上所述，空间一点Ｐ在世界坐标系中的坐标(xw,yw,zw)与其在像平面上的投影点ｐ在像素坐标系中的坐标(u,v)的关系如下：Among them, (u,v,1)T is the coordinate of p in the pixel coordinate system; dx , dy are the physical dimensions of each pixel in the pixel coordinate system in the X-axis and Y-axis direction respectively, (u0 , v0 ) is the coordinate of the origin O of the physical coordinate system in the pixel coordinate system. To sum up, the relationship between the coordinates (xw , yw , zw ) of a point P in the world coordinate system and the coordinates (u, v) of the projected point p on the image plane in the pixel coordinate system is as follows:
其中，矩阵M1由fx、fy、u0、v0决定，fx、fy分别表示焦距ｆ在图像Ｘ轴和Ｙ轴方向上的投影距离，这些参数和相机的内部参数相关，所以将Ｍ1称为摄像机的内参矩阵，Ｍ2反映了世界坐标系到摄像机坐标系的变换过程，与摄像机的内参数无关，所以Ｍ2称为摄像机的外参数矩阵。Among them, the matrix M1 is determined by fx , fy , u0 , and v0 , and fx , fy represent the projection distance of the focal length f on the X-axis and Y-axis of the image respectively. These parameters are related to the internal parameters of the camera. Therefore, M1 is called the internal parameter matrix of the camera, and M2 reflects the transformation process from the world coordinate system to the camera coordinate system, which has nothing to do with the internal parameters of the camera, so M2 is called the external parameter matrix of the camera.
辅助靶标坐标系到世界坐标系的变换公式：Transformation formula from auxiliary target coordinate system to world coordinate system:
其中，(xf,yf,zf,1)T和(xw,yw,zw,1)T分别是P在辅助靶标坐标系和世界坐标系中的坐标；R′是由辅助靶标坐标系到世界坐标系的旋转矩阵，t′是由辅助靶标坐标系到世界坐标系的平移向量。Among them, (xf ,yf ,zf ,1)T and (xw ,yw ,zw ,1)T are the coordinates of P in the auxiliary target coordinate system and the world coordinate system respectively; The rotation matrix from the target coordinate system to the world coordinate system, and t' is the translation vector from the auxiliary target coordinate system to the world coordinate system.
2.系统标定2. System Calibration
摄像机标定，在单目视觉中，摄像机标定参数分为内部参数和外部参数，内部参数确定了摄像机内部的几何和光学特征，不随摄像机位置的变化而改变，包括摄像机有效焦距和姿态，摄像机位置发生变化时需要重新标定。本发明采用的标定方法是利用辅助靶标来完成摄像机的标定，标定时将摄像机与辅助靶标相距一定距离固定好，打开CCD摄像机电源，在摄像机视场范围内，如图6所示，选取量块的顶点作为标定点，每移动一个位置拍摄一副图像，将获得的标志点的二维图像信息通过网络数据线传送并保存到计算机中，利用提取的的所有位置的标志点图像坐标及其对应的已知世界坐标，带入小孔成像测量系统模型中，进而完成对摄像机内外参数的求解，并保存到系统参数文件中，以备测量阶段调用，具体的计算工程如下：Camera calibration. In monocular vision, camera calibration parameters are divided into internal parameters and external parameters. The internal parameters determine the geometric and optical characteristics of the camera, which do not change with the change of the camera position, including the effective focal length and attitude of the camera, and the position of the camera. Changes require recalibration. The calibration method that the present invention adopts is to utilize the auxiliary target to complete the calibration of the camera. During calibration, the camera is fixed at a certain distance from the auxiliary target, the power of the CCD camera is turned on, and within the field of view of the camera, as shown in Figure 6, a measuring block is selected The apex of the vertex is used as a calibration point, and an image is taken every time a position is moved, and the two-dimensional image information of the obtained marker points is transmitted and saved to the computer through the network data line, and the extracted image coordinates of the marker points and their corresponding The known world coordinates are brought into the pinhole imaging measurement system model, and then the internal and external parameters of the camera are solved, and saved to the system parameter file for calling in the measurement phase. The specific calculation project is as follows:
本发明采用线性模型来计算，建立到顶点pii＝(1,2,3,4,5，6，7)的世界坐标系坐标(xwi,ywi,zwi)与其成像点ｐ在图像像素坐标系中坐标(u,v)的关系。The present invention uses a linear model to calculate, and establishes the world coordinate system coordinates (xwi , ywi , zwi ) to the vertex pi i=(1,2,3,4,5,6,7) and its imaging point p at The relationship between coordinates (u, v) in the image pixel coordinate system.
其中mij为变换矩阵M的第i行j列元素，上式可以写成以下方程组：Where mij is the i-th row j column element of the transformation matrix M, the above formula can be written as the following equations:
将上式中进行化简，消去zci，可以推出两个未知数是mij的线性方程：By simplifying the above formula and eliminating zci , two linear equations whose unknowns are mij can be deduced:
将靶标上n个标定点的世界坐标代入上式，则可以得到2n个线性方程，用矩阵表示如下：Substituting the world coordinates of n calibration points on the target into the above formula, 2n linear equations can be obtained, expressed as follows in matrix:
假定m34为1,公式上式可以写成：Assuming thatm34 is 1, the formula above can be written as:
Km＝UKm=U
其中,K是的2n×11矩阵，m是需要求出来的11维向量；U为2n维向量。Among them, K is a 2n×11 matrix, m is an 11-dimensional vector to be obtained; U is a 2n-dimensional vector.
已知2n＞11，则可用最小二乘法求出矩阵M的解It is known that 2n>11, then the solution of matrix M can be obtained by the method of least squares
m＝(KTK)-1KTUm＝(KT K)-1 KT U
3.图像采集3. Image acquisition
将图像采集卡采集的图像传输给计算机，经过图像处理得到每个辅助标志点轮廓像素坐标，先采用最小二乘椭圆拟合得到16个辅助标志点中心的像素坐标值，每4个辅助标志点中心可构造四边形，利用四边形对角线交点求出4个测量标志点中心像素坐标值(ui,vi)（i=1,2,3,4）。The image collected by the image acquisition card is transmitted to the computer, and the pixel coordinates of the outline of each auxiliary marker point are obtained through image processing. A quadrilateral can be constructed in the center, and the center pixel coordinates (u i , v i ) of the four measurement marker points (ui , vi ) (i=1,2,3,4) can be obtained by using the intersection points of the diagonals of the quadrilateral.
4.待测点坐标解算，摄像机的成像模型是基于针孔成像原理，C1、C2、C3、C4是标志点在图像平面上对应的像点，Lij分别表示的是目标物上标志点Pi和Pj的距离，lij是像点的距离，di是Oc到像点的距离，γij是OPi和OPj的夹角，A是四个标志点的交点，B是四个标志点像点的交点，α1、α2分别是AP4和P1P4、P3P4所成的夹角，β1、β2分别是OcB和OcC1、OcC3所成的夹角，θ是AP3、AP4所成的夹角，以上这些数据均可以通过图像处理和测量求出。像素坐标(u,v)通过图像处理算法可以求得，经过转换求得像坐标(xi,yi)和交点B的坐标(xb,yb)。D1、D2、D3、D4分别是Oc到Pi的距离，经过合适的算法可得：4. Calculate the coordinates of the points to be measured. The imaging model of the camera is based on the principle of pinhole imaging. C1 , C2 , C3 , and C4 are the corresponding image points of the marker points on the image plane, and Lij respectively represent the target The distance between the mark points Pi and Pj on the object, lij is the distance of the image point, di is the distance from Oc to the image point, γij is the angle between OPi and OPj , A is the distance between the four mark points intersection point, B is the intersection point of four landmarks, α1 , α2 are the angles formed by AP4 and P1 P4 , P3 P4 respectively, β1 , β2 are Oc B and O The angle formed byc C1 and Oc C3 , θ is the angle formed by AP3 and AP4 , the above data can be obtained through image processing and measurement. The pixel coordinates (u, v) can be obtained through image processing algorithms, and the image coordinates (xi , yi ) and the coordinates of the intersection point B (xb , yb ) can be obtained through conversion. D1 , D2 , D3 , and D4 are the distances from Oc to Pi respectively, and can be obtained through a suitable algorithm:
其中同理亦可以求出D2,D4。in Similarly, D2 and D4 can also be obtained.
确定四个测量标志点中心在摄像机坐标系中的坐标：Determine the coordinates of the centers of the four measurement marker points in the camera coordinate system:
经上述坐标转换可得到四个测量标志点中心在世界坐标系中的坐标值根据测量标志点中心在辅助坐标系下的坐标将测量标志点中心在世界坐标系下的坐标值和靶标坐标系下的坐标值带入公式计算辅助靶标坐标系到世界坐标系的旋转和平移矩阵，根据测头中心在辅助靶标坐标系下的坐标，计算测头中心的世界坐标值。通过移动辅助靶标对被测物体表面进行多点测量。After the above coordinate conversion, the coordinate values of the centers of the four measurement marker points in the world coordinate system can be obtained According to the coordinates of the center of the measurement mark point in the auxiliary coordinate system Put the coordinate value of the center of the measurement mark point in the world coordinate system and the coordinate value in the target coordinate system into the formula to calculate the rotation and translation matrix from the auxiliary target coordinate system to the world coordinate system, according to the coordinate value of the probe center in the auxiliary target coordinate system Coordinate, calculate the world coordinate value of the probe center. Multi-point measurement is performed on the surface of the measured object by moving the auxiliary target.
5.转站测量和数据空间拼接，由于摄像机的一次测量视场有限，决定了对于某些大型工件进行测量时摄像机和被测零件保持一个相对位置不可能将所有零件上待测点测量完毕，所以要进行转站测量，调整摄像机和被测物体的位置，使辅助靶标测棒上贴有标志点的一面正对摄像机，且侧棒的轴线中心位于转站台0刻度线处，并记录此时摄像机的位置1，在位置1上进行有效视场内部分点的测量，其中第一次拍摄时将转站球上可见得标识孔G保存到集合φ中，当位置1的可测目标点测量完毕后，将摄像机调整到位置2，调整位置时，转站球姿态不发生变化，同时旋转辅助靶标的测棒使贴有标志点的一面正对摄像机，且要保证集合φ中至少有3个标识孔在相机中可见，根据每个标记孔在辅助靶标坐标系中的坐标值（xfG,yfG,zfG），计算得到标记孔在位置1所在的世界坐标系统的坐标值为，在位置2所在的世界坐标系中的坐标值为则两次位置的转换关系为：5. Transfer station measurement and data space splicing, due to the limited field of view of the camera for one measurement, it is determined that when measuring some large workpieces, the camera and the measured parts maintain a relative position and it is impossible to measure all the points to be measured on the parts. Therefore, to carry out the transfer station measurement, adjust the position of the camera and the object to be measured so that the side of the auxiliary target measuring rod with the marked point is facing the camera, and the axis center of the side rod is at the 0 scale line of the transfer station, and record at this time The position of the camera is 1, and the measurement of the points in the effective field of view is carried out at position 1. When the first shot is taken, the visible mark hole G on the transfer ball is saved in the set φ. When the measurable target point at position 1 is measured After the completion, adjust the camera to position 2. When adjusting the position, the posture of the transfer ball does not change. At the same time, rotate the measuring stick of the auxiliary target so that the side with the marked point is facing the camera, and it must be ensured that there are at least 3 in the set φ The marked hole is visible in the camera. According to the coordinate value (xfG , yfG , zfG ) of each marked hole in the auxiliary target coordinate system, the coordinate value of the world coordinate system where the marked hole is located at position 1 is calculated. The coordinate value in the world coordinate system where position 2 is located is Then the conversion relationship between the two positions is:
其中G2为位置2上的与位置1共有的标记孔，G1为φ中相对应的共有标记孔。Among them, G2 is the marked hole shared with position 1 on position2 , andG1 is the corresponding shared marked hole in φ.
在位置2上的第一次拍摄时，根据与集合φ中共有的标记孔计算位置1所在世界坐标系和位置2所在世界坐标系之间的坐标转换参数R21和t21，将位置2上新增加的标记孔用坐标转换参数R21和t21进行变换，将变换后的结构添加到φ中，从而对φ进行了更新。同时继续在位置2上对目标点进行量，得到测头在位置2的世界坐标系中的坐标值为，其中每次测量结果通过坐标转换参数R21和t21进行旋转和平移变换，则在位置2测头中心的测量结果变换到位置1所在的世界坐标系下的结果为：When taking the first shot at position 2, the coordinate conversion parameters R21 and t21 between the world coordinate system of position 1 and the world coordinate system of position 2 are calculated according to the marked hole shared with the set φ, and the position 2 The newly added marked hole is transformed with the coordinate transformation parametersR21 andt21 , and the transformed structure is added to φ, thereby updating φ. At the same time, continue to measure the target point at position 2, and obtain the coordinate value of the probe in the world coordinate system at position 2, where each measurement result is rotated and translated by the coordinate conversion parameters R21 and t21 , then in Measurement result of probe center at position 2 The result of transforming to the world coordinate system where position 1 is located is:
根据上式从而将两个位置上的测量结果统一到一个坐标中，实现三位拼接测量，以此类推不断调整摄像机和被测物体之间的相对位置，直到将大型零件上所有的待测点测量完毕，并将各个位置上的测量结果统一到一个坐标系下，实现各个位置的拼接，从而最终实现对大型零件的三维重构。According to the above formula, the measurement results at the two positions are unified into one coordinate, realizing three-dimensional splicing measurement, and so on, constantly adjusting the relative position between the camera and the measured object until all the points to be measured on the large part are integrated After the measurement is completed, the measurement results at each position are unified into one coordinate system to realize the splicing of each position, so as to finally realize the three-dimensional reconstruction of large parts.
本发明采用摄像机转站的方法对待测器件最优视点进行规划分析，克服了摄像机一次测量视场的有限性，扩大了测量的范围，同时测量辅助靶标上转站球外表面上按照一定的角度均匀地在经线和纬线的交叉位置布置圆柱孔，作为标记孔，圆柱孔的轴线与转战球的球心相交，靶标转站球圆柱孔上粘贴各种颜色不相同的彩色回光反射圆形薄贴，标记孔在辅助靶标中的位置由颜色标识唯一确定，从而攻克可立体视觉及其它视觉测量技术标志点匹配难、三维空间拼接复杂的技术难关，实现对大型零件的快速、高精度、大范围和低成本的自动测量和重建。The present invention adopts the method of camera transfer station to plan and analyze the optimal viewpoint of the device under test, which overcomes the limitation of the camera's one-time measurement field of view, expands the measurement range, and simultaneously measures the auxiliary target on the outer surface of the transfer station ball according to a certain angle. Arrange cylindrical holes evenly at the crossing positions of warp and latitude as marking holes. The axis of the cylindrical holes intersects with the center of the transfer ball. Paste various colored light-reflecting circular thin films on the cylindrical holes of the target transfer ball. The position of the marking hole in the auxiliary target is uniquely determined by the color mark, so as to overcome the technical difficulties of stereo vision and other visual measurement technology mark point matching and complex three-dimensional space splicing, and realize fast, high-precision, large-scale measurement of large parts. range and low-cost automated measurement and reconstruction.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410003137.2ACN103759669B (en) | 2014-01-03 | 2014-01-03 | A kind of monocular vision measuring method of heavy parts |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410003137.2ACN103759669B (en) | 2014-01-03 | 2014-01-03 | A kind of monocular vision measuring method of heavy parts |
| Publication Number | Publication Date |
|---|---|
| CN103759669A CN103759669A (en) | 2014-04-30 |
| CN103759669Btrue CN103759669B (en) | 2016-11-23 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410003137.2AExpired - Fee RelatedCN103759669B (en) | 2014-01-03 | 2014-01-03 | A kind of monocular vision measuring method of heavy parts |
| Country | Link |
|---|---|
| CN (1) | CN103759669B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108871190A (en)* | 2018-06-27 | 2018-11-23 | 西安交通大学 | A kind of hand-held ball-type target and measurement method in binocular stereo vision measurement |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104406540B (en)* | 2014-12-02 | 2017-03-29 | 吉林大学 | Calibration system based on point to the automobile Shape measure structure light of initial point distance constraint |
| CN104517291B (en)* | 2014-12-15 | 2017-08-01 | 大连理工大学 | Pose measurement method based on target coaxial circle feature |
| CN104807476A (en)* | 2015-04-23 | 2015-07-29 | 上海大学 | Pose estimation-based quick probe calibration device and method |
| CN105571519B (en)* | 2016-02-19 | 2018-09-25 | 广西玉柴机器股份有限公司 | The auxiliary device and its joining method of the point cloud of spatial digitizer |
| CN105588484A (en)* | 2016-02-26 | 2016-05-18 | 共享装备股份有限公司 | Measuring arm position-shifting reference correcting device |
| CN106289071B (en)* | 2016-08-18 | 2018-10-19 | 温州大学 | A kind of structure three-dimensional displacement monocular photographing measurement method |
| CN106248014A (en)* | 2016-08-23 | 2016-12-21 | 中国人民解放军信息工程大学 | A kind of three-dimensional coordinate measurement method and device based on single-phase |
| CN106225720A (en)* | 2016-09-07 | 2016-12-14 | 吉林大学 | Automobile pattern based on three-dimensional rectangular coordinate is without kinematic constraint detector |
| CN106247944B (en)* | 2016-09-26 | 2019-01-11 | 西安理工大学 | Code targets and vision coordinate measurement method based on Code targets |
| CN107328387A (en)* | 2017-07-04 | 2017-11-07 | 成都康烨科技有限公司 | Angle measuring method, device and video camera |
| WO2019041349A1 (en)* | 2017-09-04 | 2019-03-07 | 大连理工大学 | Three-dimensional visual information measuring method based on rotating lens |
| CN108317958A (en)* | 2017-12-29 | 2018-07-24 | 广州超音速自动化科技股份有限公司 | A kind of image measuring method and measuring instrument |
| CN110686750A (en)* | 2018-07-05 | 2020-01-14 | 杭州海康威视数字技术股份有限公司 | Water level detection method, device and system |
| CN109357633B (en)* | 2018-09-30 | 2022-09-30 | 先临三维科技股份有限公司 | Three-dimensional scanning method, device, storage medium and processor |
| CN109470142A (en)* | 2018-11-22 | 2019-03-15 | 北京航天计量测试技术研究所 | A monocular pose measurement method based on circular landmarks |
| CN110702008B (en)* | 2019-11-11 | 2025-06-24 | 深圳市泰斯特尔系统科技有限公司 | A machine vision and scanning detection device and working method thereof |
| CN111578837B (en)* | 2020-04-30 | 2022-03-01 | 北京农业智能装备技术研究中心 | Plant shape visual tracking measurement method for agricultural robot operation |
| CN112008729A (en)* | 2020-09-01 | 2020-12-01 | 云南电网有限责任公司电力科学研究院 | Collision detection method for overhead line maintenance mechanical arm |
| CN113781571B (en)* | 2021-02-09 | 2025-03-18 | 北京沃东天骏信息技术有限公司 | Image processing method and device |
| CN116499354B (en)* | 2022-01-19 | 2025-09-16 | 广东美的制冷设备有限公司 | Scale measurement method and device based on monocular vision, electrical equipment and medium |
| CN115120256B (en)* | 2022-07-19 | 2025-07-25 | 邦世科技(南京)有限公司 | Automatic X-ray equipment and shooting method thereof |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1912532A (en)* | 2005-08-11 | 2007-02-14 | 北京航空航天大学 | Three-D visual measuring quantity transmission method and plane target mark |
| CN102062578A (en)* | 2010-12-13 | 2011-05-18 | 西安交通大学 | Handheld optical target for measuring visual coordinate and measuring method thereof |
| CN102288106A (en)* | 2010-06-18 | 2011-12-21 | 合肥工业大学 | Large-space visual tracking six-dimensional measurement system and method |
| CN202710025U (en)* | 2012-07-31 | 2013-01-30 | 河南科技大学 | Station transferring plate for transferring stations during reverse measurement |
| CN103162622A (en)* | 2013-02-28 | 2013-06-19 | 西安交通大学 | Monocular vision system, portable ball target used by monocular vision system and measuring method of monocular vision system |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5252820B2 (en)* | 2007-03-27 | 2013-07-31 | パナソニック株式会社 | 3D measurement method and 3D shape measurement apparatus using the same |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1912532A (en)* | 2005-08-11 | 2007-02-14 | 北京航空航天大学 | Three-D visual measuring quantity transmission method and plane target mark |
| CN102288106A (en)* | 2010-06-18 | 2011-12-21 | 合肥工业大学 | Large-space visual tracking six-dimensional measurement system and method |
| CN102062578A (en)* | 2010-12-13 | 2011-05-18 | 西安交通大学 | Handheld optical target for measuring visual coordinate and measuring method thereof |
| CN202710025U (en)* | 2012-07-31 | 2013-01-30 | 河南科技大学 | Station transferring plate for transferring stations during reverse measurement |
| CN103162622A (en)* | 2013-02-28 | 2013-06-19 | 西安交通大学 | Monocular vision system, portable ball target used by monocular vision system and measuring method of monocular vision system |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108871190A (en)* | 2018-06-27 | 2018-11-23 | 西安交通大学 | A kind of hand-held ball-type target and measurement method in binocular stereo vision measurement |
| Publication number | Publication date |
|---|---|
| CN103759669A (en) | 2014-04-30 |
| Publication | Publication Date | Title |
|---|---|---|
| CN103759669B (en) | A kind of monocular vision measuring method of heavy parts | |
| CN107883870B (en) | Overall calibration method based on binocular vision system and laser tracker measuring system | |
| CN111156925B (en) | Three-dimensional measurement method for large component based on line structured light and industrial robot | |
| CN103559735B (en) | A kind of three-dimensional rebuilding method and system | |
| CN103267491B (en) | The method and system of automatic acquisition complete three-dimensional data of object surface | |
| ES2399883T3 (en) | Procedure and system for displaying surface errors | |
| CN101387501B (en) | Apparatus and method for measuring circular cross-sectional shape and orientation of ultra-large workpiece | |
| CN102445164B (en) | Three-dimensional shape vision measuring method and system for large component surface | |
| CN100562707C (en) | Binocular vision rotating axis calibration method | |
| CN104165750B (en) | Measurement method for pose of wind tunnel model combining stereoscopic vision with gyroscope | |
| CN104634248B (en) | Revolving shaft calibration method under binocular vision | |
| CN104897142A (en) | Three-dimensional target for binocular or multi-view vision dimension measuring | |
| CN103759637B (en) | A kind of subsidiary target | |
| CN102506758B (en) | Object surface three-dimensional morphology multi-sensor flexible dynamic vision measurement system and method | |
| CN110966935B (en) | Deflection measurement system integrated geometric calibration method based on mark points | |
| CN104567666A (en) | Measuring method for roller bearing block spatial position | |
| CN105424058B (en) | Digital camera projection centre position method for precisely marking based on photogrammetric technology | |
| CN111091599B (en) | Multi-camera-projector system calibration method based on sphere calibration object | |
| CN105783773A (en) | Numerical value calibration method for line structured light vision sensor | |
| CN103606147B (en) | Coordinate system conversion and calibration method for multiple measurement cameras with different fields of view | |
| CN102692214A (en) | Narrow space binocular vision measuring and positioning device and method | |
| CN104990515A (en) | Three-dimensional shape measurement system and method for large-size object | |
| CN111256592B (en) | External parameter calibration device and method for structured light sensor | |
| CN102305598A (en) | Underwater Photogrammetry Method for Surface Accuracy of Semi-rigid Self-rebounding Reflector | |
| CN104482924A (en) | Revolution body object pose vision measurement method |
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20161123 Termination date:20180103 |