CN109540042B - Automobile morphology active vision detection system and method based on unconstrained light plane - Google Patents

Automobile morphology active vision detection system and method based on unconstrained light plane Download PDF

Info

Publication number
CN109540042B
CN109540042B CN201910038238.6A CN201910038238A CN109540042B CN 109540042 B CN109540042 B CN 109540042B CN 201910038238 A CN201910038238 A CN 201910038238A CN 109540042 B CN109540042 B CN 109540042B
Authority
CN
China
Prior art keywords
camera
laser
coordinate system
dimensional
cylindrical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201910038238.6A
Other languages
Chinese (zh)
Other versions
CN109540042A (en
Inventor
徐观
陈芳
林慧英
苏建
单红梅
张立斌
刘玉梅
潘洪达
陈熔
戴建国
李晓韬
陈俊伊
沈慧
朱尧平
胡常青
袁静
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN201910038238.6A priority Critical patent/CN109540042B/en
Publication of CN109540042A publication Critical patent/CN109540042A/en
Application granted granted Critical
Publication of CN109540042B publication Critical patent/CN109540042B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

本发明公开了一种基于无约束光平面的汽车形貌主动视觉检测系统与方法，旨在解决基于无约束光平面的汽车形貌主动视觉检测问题。基于无约束光平面的汽车形貌主动视觉检测系统主要由柱面靶标(1)、摄像机(2)、摄像机支架(3)、二维靶标板(4)、激光器连接件(5)与激光器(6)组成。基于无约束光平面的汽车形貌主动视觉检测方法由图像采集、在二维靶标板(4)坐标系下激光面的重建、车辆表面特征点的重建三个步骤组成，提供了一种结构简单、性能可靠的基于无约束光平面的汽车形貌主动视觉检测系统与方法。

Figure 201910038238

The invention discloses an active vision detection system and method for vehicle topography based on an unconstrained light plane, aiming at solving the problem of active vision detection of automobile topography based on the unconstrained light plane. The vehicle topography active vision detection system based on the unconstrained light plane is mainly composed of a cylindrical target (1), a camera (2), a camera bracket (3), a two-dimensional target plate (4), a laser connector (5) and a laser ( 6) Composition. The vehicle topography active vision detection method based on the unconstrained light plane consists of three steps: image acquisition, reconstruction of the laser surface in the two-dimensional target plate (4) coordinate system, and reconstruction of the feature points on the vehicle surface, providing a simple structure. , A reliable and reliable vehicle topography active vision detection system and method based on an unconstrained light plane.

Figure 201910038238

Description

基于无约束光平面的汽车形貌主动视觉检测系统与方法Active Vision Detection System and Method for Vehicle Topography Based on Unconstrained Light Plane

技术领域technical field

本发明涉及一种汽车形貌检测领域的测量设备与测量方法，更具体的说， 它是一种基于无约束光平面的汽车形貌主动视觉检测系统与方法。The invention relates to a measuring device and a measuring method in the field of automobile topography detection, more specifically, it is an automobile topography active vision detection system and method based on an unconstrained light plane.

背景技术Background technique

近年来随着检测技术的不断发展,机器视觉逐渐成为提高汽车检测效率和 保证汽车检测精度的关键技术。基于机器视觉的车辆形貌检测可为车辆超限超 载的自动辨识、车型分类、整车尺寸参数检测、车辆信息的获取和重建等提供 重要的研究依据。传统的汽车形貌测量方法主要有接触式和固定式，然而接触 式测量方法测量速度慢，不适合大型物体的测量，与被测物体接触还会导致其 表面磨损，同时也会对测量头本身造成损伤；固定式检测设备固定式的不能测 量一些有遮挡的位置，需要制作龙门架，因此成本较高。为解决以上问题设计 一种非接触式、测量速度快、适合测量大型物体的无严格约束的汽车形貌检测 方法十分重要，设计了一种基于无约束光平面的汽车形貌主动视觉检测系统与 方法，系统的激光平面和二维靶标板之间的位置关系在后续重建过程中不需要 再重复测量，标定后即可得到，即激光平面与二维靶标板之间的位置无严格约 束，大大降低了设备制造时对激光与二维靶标板之间的装配精度要求，降低了 设备的生产成本，扩展了设备的应用范围。In recent years, with the continuous development of detection technology, machine vision has gradually become a key technology to improve the efficiency of vehicle detection and ensure the accuracy of vehicle detection. Vehicle topography detection based on machine vision can provide an important research basis for automatic identification of vehicle overloading, vehicle classification, vehicle size parameter detection, vehicle information acquisition and reconstruction, etc. The traditional automobile topography measurement methods mainly include contact type and fixed type. However, the contact type measurement method has a slow measurement speed and is not suitable for the measurement of large objects. Cause damage; fixed detection equipment cannot measure some blocked positions, and needs to make a gantry, so the cost is high. In order to solve the above problems, it is very important to design a non-contact, fast measurement speed, and unconstrained vehicle shape detection method suitable for measuring large objects. method, the positional relationship between the laser plane and the two-dimensional target plate of the system does not need to be repeatedly measured in the subsequent reconstruction process, and can be obtained after calibration, that is, the position between the laser plane and the two-dimensional target plate is not strictly constrained, greatly The requirements for assembly precision between the laser and the two-dimensional target plate during equipment manufacturing are reduced, the production cost of the equipment is reduced, and the application range of the equipment is expanded.

发明内容SUMMARY OF THE INVENTION

本发明针对解决在获取汽车形貌过程中，接触式检测设备价格昂贵、测量 速度慢，固定式检测设备成本较高等问题，提出了一种性能可靠、结构简单、 操作简便的方法与系统，实现对汽车形貌特征点的三维自由重建，为完善汽车 形貌检测技术奠定了研究基础。该方法主要由一个柱面靶标，一部摄像机，一 个摄像机支架，一个二维靶标板，一个激光器连接件，一个激光器构成。通过 求解二维靶标坐标系的光平面坐标，实现了采用无约束的激光平面对汽车形貌 的重建。In order to solve the problems of high price of contact detection equipment, slow measurement speed, and high cost of fixed detection equipment in the process of obtaining the vehicle topography, the present invention proposes a method and system with reliable performance, simple structure and simple operation, which realizes The 3D free reconstruction of the feature points of the vehicle topography has laid a research foundation for improving the vehicle topography detection technology. The method mainly consists of a cylindrical target, a camera, a camera bracket, a two-dimensional target plate, a laser connector, and a laser. By solving the light plane coordinates of the two-dimensional target coordinate system, the reconstruction of the vehicle topography using an unconstrained laser plane is realized.

结合说明书附图，本发明采用如下技术方案予以实现:In conjunction with the accompanying drawings of the description, the present invention adopts the following technical solutions to be realized:

基于无约束光平面的汽车形貌主动视觉检测系统包括有柱面靶标、摄像机、 摄像机支架、二维靶标板、激光器连接件与激光器；The vehicle topography active vision detection system based on the unconstrained light plane includes a cylindrical target, a camera, a camera bracket, a two-dimensional target plate, a laser connector and a laser;

柱面靶标放置在地面上，摄像机支架放置在地面上，摄像机通过底部的螺 纹孔与摄像机支架的螺栓螺纹固定连接，二维靶标板置入激光器连接件钢板的 细长凹槽中与激光器连接件焊接连接，激光器插入激光器连接件钢管的内孔中 与激光器连接件过盈配合连接。The cylindrical target is placed on the ground, the camera bracket is placed on the ground, the camera is fixedly connected with the bolt of the camera bracket through the threaded hole at the bottom, and the two-dimensional target plate is placed in the slender groove of the laser connector steel plate to connect with the laser connector. For welding connection, the laser is inserted into the inner hole of the steel pipe of the laser connector and connected with the interference fit of the laser connector.

技术方案中所述的柱面靶标为标准圆管沿轴向切开而制成的半圆柱状零 件，且内表面贴有规则的几何图案。The cylindrical target described in the technical solution is a semi-cylindrical part made by cutting a standard circular tube in the axial direction, and the inner surface is affixed with a regular geometric pattern.

技术方案中所述的摄像机为装有窄带滤光片的广角工业相机。The camera described in the technical solution is a wide-angle industrial camera equipped with a narrow-band filter.

技术方案中所述的摄像机支架为通用摄像机支架。The camera bracket described in the technical solution is a universal camera bracket.

技术方案中所述的二维靶标板为一块矩形钢板制成的零件，且表面贴有规 则的LED发光点阵，二维靶标板的LED发光点阵的波长与摄像机的窄带滤光片 的带通波长一致。The two-dimensional target plate described in the technical solution is a part made of a rectangular steel plate, and the surface is affixed with a regular LED light-emitting lattice. The wavelength of the LED light-emitting lattice of the two-dimensional target plate is the same as that of the narrow-band filter of the camera. The pass wavelength is the same.

技术方案中所述的激光器连接件由一块加工有细长凹槽的钢板与一段钢管 焊接而成。The laser connector described in the technical solution is formed by welding a steel plate with elongated grooves and a section of steel pipe.

技术方案中所述的激光器为可发射激光平面的圆柱形零件，激光器发出的 激光波长与摄像机的窄带滤光片的带通波长一致。The laser described in the technical solution is a cylindrical part that can emit a laser plane, and the laser wavelength emitted by the laser is consistent with the bandpass wavelength of the narrow-band filter of the camera.

基于无约束光平面的汽车形貌主动视觉检测方法的具体步骤如下：The specific steps of the vehicle topography active vision detection method based on the unconstrained light plane are as follows:

第一步：基于无约束光平面的汽车形貌主动视觉检测方法的图像采集：Step 1: Image acquisition based on an unconstrained light plane-based active vision detection method for vehicle topography:

将摄像机支架放置在地面上，摄像机固定在摄像机支架上，柱面靶标放置 在地面上,激光器固定在激光器连接件上，打开激光器，激光器发出的激光平面 与柱面靶标交于一条激光曲线，摄像机采集一幅图像，图像包含柱面靶标、二 维靶标板以及激光器与柱面靶标相交的投影激光曲线；Place the camera bracket on the ground, the camera is fixed on the camera bracket, the cylindrical target is placed on the ground, the laser is fixed on the laser connector, the laser is turned on, the laser plane emitted by the laser and the cylindrical target intersect a laser curve, the camera Collect an image that includes the cylindrical target, the two-dimensional target plate, and the projected laser curve where the laser intersects the cylindrical target;

第二步：基于无约束光平面的汽车形貌主动视觉检测方法的在二维靶标板 坐标系下激光面的重建：Step 2: Reconstruction of the laser surface in the two-dimensional target plate coordinate system based on the active vision detection method of vehicle topography based on the unconstrained light plane:

首先，选取摄像机坐标系作为固定的全局坐标系，根据摄像机采集的柱面 靶标特征点的图像，柱面靶标上特征点世界坐标

Figure BDA00019446186100000211
到图像坐标
Figure BDA00019446186100000212
的投影关系为First, the camera coordinate system is selected as a fixed global coordinate system, and according to the image of the feature points of the cylindrical target collected by the camera, the world coordinates of the feature points on the cylindrical target
Figure BDA00019446186100000211
to image coordinates
Figure BDA00019446186100000212
The projection relation is

Figure BDA0001944618610000021
Figure BDA0001944618610000021

其中，s1为比例因子，采用DLT方法解算从柱面靶标坐标系到摄像机图像坐标系的投影矩阵PQI，对PQI进行RQ分解得到从柱面靶标坐标系到摄像机坐标系的旋 转矩阵RRC和平移向量tRC；Among them, s 1 is the scale factor, the DLT method is used to solve the projection matrix P QI from the cylindrical target coordinate system to the camera image coordinate system, and the RQ decomposition of P QI is used to obtain the rotation matrix from the cylindrical target coordinate system to the camera coordinate system. R RC and translation vector t RC ;

然后，根据摄像机采集的柱面靶标上的激光线的图片，采用Steger算法换 提取图片中投影激光曲线上的点的二维图像坐标

Figure BDA0001944618610000022
Then, according to the picture of the laser line on the cylindrical target collected by the camera, the Steger algorithm is used to extract the two-dimensional image coordinates of the point on the projected laser curve in the picture.
Figure BDA0001944618610000022

从柱面靶标坐标系到摄像机坐标系的转换关系为The conversion relationship from the cylindrical target coordinate system to the camera coordinate system is:

Figure BDA0001944618610000023
Figure BDA0001944618610000023

其中，s2为比例因子；Among them, s 2 is the scale factor;

柱面坐标系下

Figure BDA0001944618610000024
满足In cylindrical coordinate system
Figure BDA0001944618610000024
Satisfy

Figure BDA0001944618610000025
Figure BDA0001944618610000025

其中，r为柱面靶标的圆柱半径，

Figure BDA0001944618610000026
为向量
Figure BDA0001944618610000027
的第一个和第二个分量，由以上两式可得激光曲线上的点在柱面靶标坐标系下的坐标
Figure BDA0001944618610000028
where r is the cylindrical radius of the cylindrical target,
Figure BDA0001944618610000026
as a vector
Figure BDA0001944618610000027
The first and second components of , the coordinates of the point on the laser curve in the cylindrical target coordinate system can be obtained from the above two equations
Figure BDA0001944618610000028

根据从柱面靶标坐标系到摄像机坐标系的旋转矩阵RQC和平移向量tQC，以及 激光线上的点在柱面靶标坐标系下的坐标

Figure BDA0001944618610000029
由下式可求得激光线上的点在摄 像机坐标系下的坐标为According to the rotation matrix R QC and translation vector t QC from the cylindrical target coordinate system to the camera coordinate system, and the coordinates of the point on the laser line in the cylindrical target coordinate system
Figure BDA0001944618610000029
The coordinates of the point on the laser line in the camera coordinate system can be obtained from the following formula:

Figure BDA00019446186100000210
Figure BDA00019446186100000210

激光平面上的点

Figure BDA0001944618610000031
满足point on the laser plane
Figure BDA0001944618610000031
Satisfy

Figure BDA0001944618610000032
Figure BDA0001944618610000032

由SVD分解法求得激光平面在摄像机坐标系下的坐标πC；Obtain the coordinate π C of the laser plane in the camera coordinate system by the SVD decomposition method;

由张正友二维标定算法可以解算从二维靶标板坐标系到摄像机坐标系的旋 转矩阵RRC和平移向量tRC，则从二维靶标板坐标系到摄像机坐标系的单应矩阵为The two-dimensional calibration algorithm of Zhang Zhengyou can solve the rotation matrix R RC and translation vector t RC from the two-dimensional target plate coordinate system to the camera coordinate system, then the homography matrix from the two-dimensional target plate coordinate system to the camera coordinate system is

Figure BDA0001944618610000033
Figure BDA0001944618610000033

根据激光平面在摄像机坐标系下的坐标πC和二维靶标板坐标系到摄像机2 坐标系的单应矩阵HRC，在二维靶标板坐标系下激光平面坐标为According to the coordinate π C of the laser plane in the camera coordinate system and the homography matrix H RC from the two-dimensional target plate coordinate system to the camera 2 coordinate system, the coordinates of the laser plane in the two-dimensional target plate coordinate system are:

Figure BDA0001944618610000034
Figure BDA0001944618610000034

第三步：基于无约束光平面的汽车形貌主动视觉检测方法的车辆表面特征 点的重建：Step 3: Reconstruction of vehicle surface feature points based on unconstrained light plane active vision detection method for vehicle topography:

将摄像机支架放置在地面上，摄像机固定在摄像机支架上，汽车驶入摄像 机的视场内，激光器固定在激光器连接件上，打开激光器，调整二维靶标板的 位置及角度使激光器发出的激光平面与车辆表面交于一条激光线，使用摄像机 采集一幅图像，图像包含二维靶标板以及激光器与车辆表面相交的投影激光直 线，然后将二维靶标板放置在第q位置上，摄像机继续采集第q位置的图像，q ＝1,2，…，n；Place the camera bracket on the ground, fix the camera on the camera bracket, drive the car into the camera's field of view, fix the laser on the laser connector, turn on the laser, and adjust the position and angle of the two-dimensional target plate to make the laser emitted by the laser flat. It intersects with the surface of the vehicle on a laser line, and the camera is used to collect an image. The image includes the two-dimensional target plate and the projected laser line where the laser intersects the surface of the vehicle. Then, the two-dimensional target plate is placed at the qth position, and the camera continues to collect the first. image at position q, q = 1, 2, ..., n;

根据张正友二维标定算法可以得到从二维靶标板坐标系到摄像机坐标由可 求得从二维靶标板坐标系到摄像机坐标系的旋转矩阵RRC,q和平移向量tRC,q以及 摄像机内参数K，则从二维靶标板坐标系到摄像机坐标系的单应矩阵为According to Zhang Zhengyou's two-dimensional calibration algorithm, the rotation matrix R RC, q and translation vector t RC, q from the two-dimensional target plate coordinate system to the camera coordinate system can be obtained from the two-dimensional target plate coordinate system to the camera coordinate system. parameter K, the homography matrix from the two-dimensional target board coordinate system to the camera coordinate system is

Figure BDA0001944618610000035
Figure BDA0001944618610000035

由第二步求出的二维靶标板坐标系下激光平面坐标πR以及二维靶标板坐标系到摄像机坐标系的单应矩阵HRC,q，摄像机坐标系下激光平面坐标为The laser plane coordinates π R in the two-dimensional target plate coordinate system obtained in the second step and the homography matrix H RC,q from the two-dimensional target plate coordinate system to the camera coordinate system, the laser plane coordinates in the camera coordinate system are

Figure BDA0001944618610000036
Figure BDA0001944618610000036

其中，

Figure BDA0001944618610000037
in,
Figure BDA0001944618610000037

在摄像机坐标系下激光线上的点的三维坐标

Figure BDA0001944618610000038
满足3D coordinates of points on the laser line in the camera coordinate system
Figure BDA0001944618610000038
Satisfy

Figure BDA0001944618610000041
Figure BDA0001944618610000041

由摄像机获取的图像可得二维靶标板以及激光器与车辆表面相交的投影激 光线上点的图像坐标

Figure BDA0001944618610000042
图像坐标
Figure BDA0001944618610000043
和在摄像机坐标系下激光线上的点的三 维坐标
Figure BDA0001944618610000044
转换关系为The image acquired by the camera can obtain the two-dimensional target plate and the image coordinates of the point on the projected laser line where the laser intersects the vehicle surface
Figure BDA0001944618610000042
image coordinates
Figure BDA0001944618610000043
and the 3D coordinates of the point on the laser line in the camera coordinate system
Figure BDA0001944618610000044
The conversion relationship is

Figure BDA0001944618610000045
Figure BDA0001944618610000045

其中，s3为比例因子，K为摄像机内参数；Among them, s 3 is the scale factor, and K is the parameter in the camera;

根据以上三式及SVD分解法解算在摄像机坐标系下激光平面与车身交点的 三维坐标

Figure BDA0001944618610000046
实现车身三维特征点的重建。According to the above three formulas and the SVD decomposition method, the three-dimensional coordinates of the intersection of the laser plane and the vehicle body in the camera coordinate system are calculated
Figure BDA0001944618610000046
Realize the reconstruction of three-dimensional feature points of the car body.

本发明的有益效果是：The beneficial effects of the present invention are:

(1)本发明的系统测量范围广、性能可靠、结构简单、操作简便、适用范 围广，实现了对汽车形貌的非接触、快速测量，解决了固定接触式测量系统测 量效率低、便捷性差等问题。(1) The system of the present invention has a wide measurement range, reliable performance, simple structure, simple operation, and wide application range, realizes non-contact and rapid measurement of the appearance of automobiles, and solves the problem of low measurement efficiency and poor convenience of the fixed contact measurement system. And other issues.

(2)本发明系统中的二维靶标板4和激光器6所组成的组合体可以自由扫 描，重建的视野更宽，可扫描的位置更广，对于汽车车身内部等不易直接观测 到的区域重建具有显著的重建效果。(2) The combination composed of the two-dimensional target plate 4 and the laser 6 in the system of the present invention can be scanned freely, the reconstructed field of view is wider, and the scannable position is wider. Has a significant reconstruction effect.

(3)本发明的系统采用的摄像机2为装有窄带滤光片的广角工业相机，可 获取的视野范围更广，所采用的激光器6发出的激光波长以及二维靶标板4的 LED波长与摄像机2的窄带滤光片的带通波长一致，能够有效降低环境光源对测 量过程的干扰，大大提高了测量精度。(3) The camera 2 adopted by the system of the present invention is a wide-angle industrial camera equipped with a narrow-band filter, and the available field of view is wider. The laser wavelength emitted by the adopted laser 6 and the LED wavelength of the two-dimensional target plate 4 are the same The bandpass wavelengths of the narrow-band filters of the camera 2 are consistent, which can effectively reduce the interference of the ambient light source on the measurement process and greatly improve the measurement accuracy.

(4)激光平面和二维靶标板4之间的位置关系在后续重建过程中不需要再 重复测量，标定后即可得到，即激光平面与二维靶标板4之间的位置无严格约 束，大大降低了设备制造时对激光器6与二维靶标板4之间的装配精度要求， 降低了设备的生产成本，扩展了设备的应用范围。本发明通过柱面靶标1到摄 像机2图像的投影关系，得到从柱面靶标1坐标系到摄像机2坐标系的投影矩 阵，再根据投影矩阵、激光线上的点的图像坐标以及柱面公式，可得到激光线 上的点的世界坐标，进而得到其在摄像机2坐标系下的坐标；根据点在激光平 面上及点在摄像机2坐标系下的坐标，可得到激光平面在摄像机2坐标系下的 坐标；由从二维靶标板4坐标系到摄像机2坐标系的单应矩阵及激光平面的在 摄像机2坐标系下的坐标可得到在二维靶标板4坐标系下激光平面的坐标。(4) The positional relationship between the laser plane and the two-dimensional target plate 4 does not need to be repeatedly measured in the subsequent reconstruction process, and can be obtained after calibration, that is, the position between the laser plane and the two-dimensional target plate 4 is not strictly constrained. This greatly reduces the assembly precision requirement between the laser 6 and the two-dimensional target plate 4 during equipment manufacturing, reduces the production cost of the equipment, and expands the application range of the equipment. The present invention obtains the projection matrix from the coordinate system of the cylindrical target 1 to the coordinate system of the camera 2 through the projection relationship between the cylindrical target 1 and the image of the camera 2, and then according to the projection matrix, the image coordinates of the points on the laser line and the cylindrical formula, The world coordinates of the point on the laser line can be obtained, and then its coordinates in the camera 2 coordinate system can be obtained; according to the coordinates of the point on the laser plane and the point in the camera 2 coordinate system, the laser plane can be obtained in the camera 2 coordinate system. The coordinates of the laser plane in the two-dimensional target plate 4 coordinate system can be obtained from the homography matrix from the two-dimensional target plate 4 coordinate system to the camera 2 coordinate system and the coordinates of the laser plane in the camera 2 coordinate system.

附图说明Description of drawings

图1是基于无约束光平面的汽车形貌主动视觉检测系统标定的轴测图；Fig. 1 is the axonometric view of the calibration of the vehicle topography active vision detection system based on the unconstrained light plane;

图2是基于无约束光平面的汽车形貌主动视觉检测系统重建的轴测图；Figure 2 is an axonometric view of the reconstruction of an active vision detection system for automotive topography based on an unconstrained light plane;

图3是基于无约束光平面的汽车形貌主动视觉检测系统中柱面靶标1的示 意图；Fig. 3 is the schematic diagram of cylindrical target 1 in the vehicle topography active vision detection system based on unconstrained light plane;

图4是基于无约束光平面的汽车形貌主动视觉检测系统中摄像机2的轴测 图；Fig. 4 is the axonometric view of the camera 2 in the vehicle topography active vision detection system based on the unconstrained light plane;

图5是基于无约束光平面的汽车形貌主动视觉检测系统中摄像机支架3的 轴测图；Fig. 5 is the axonometric view of the camera bracket 3 in the vehicle topography active vision detection system based on the unconstrained light plane;

图6是基于无约束光平面的汽车形貌主动视觉检测系统中二维靶标板4、激 光器连接件5、激光器6装配体的轴测图；Fig. 6 is the axonometric view of the two-dimensional target plate 4, the laser connector 5, the laser 6 assembly in the vehicle topography active vision detection system based on the unconstrained light plane;

图7是基于无约束光平面的汽车形貌主动视觉检测系统标定的原理图；Figure 7 is a schematic diagram of the calibration of an active vision detection system for automotive topography based on an unconstrained light plane;

图8是基于无约束光平面的汽车形貌主动视觉检测系统车辆表面特征点重 建的原理图；Figure 8 is a schematic diagram of the reconstruction of vehicle surface feature points based on the unconstrained light plane active vision detection system for vehicle topography;

图9是基于无约束光平面的汽车形貌主动视觉检测系统中求解激光平面在 二维靶标板4坐标系下坐标的流程图；Fig. 9 is the flow chart of solving the coordinates of the laser plane under the two-dimensional target plate 4 coordinate system in the vehicle topography active vision detection system based on the unconstrained light plane;

图10是基于无约束光平面的汽车形貌主动视觉检测系统中在摄像机2坐标 系下车辆表面特征点重建的流程图；Fig. 10 is the flow chart of the reconstruction of vehicle surface feature points in the camera 2 coordinate system in the vehicle topography active vision detection system based on the unconstrained light plane;

图中：1.柱面靶标，2.摄像机，3.摄像机支架，4.二维靶标板，5.激光器 连接件，6.激光器。In the picture: 1. Cylindrical target, 2. Camera, 3. Camera bracket, 4. Two-dimensional target plate, 5. Laser connector, 6. Laser.

具体实施方式Detailed ways

下面结合附图对本发明作进一步的详细描述:Below in conjunction with accompanying drawing, the present invention is described in further detail:

参阅图1至图6，基于无约束光平面的汽车形貌主动视觉检测系统包括柱面 靶标1、摄像机2、摄像机支架3、二维靶标板4、激光器连接件5与激光器6。Referring to Figures 1 to 6, the vehicle topography active vision detection system based on the unconstrained light plane includes a cylindrical target 1, a camera 2, a camera bracket 3, a two-dimensional target plate 4, a laser connector 5 and a laser 6.

柱面靶标1为标准圆管沿轴向切开而制成的半圆柱状零件，且内表面贴有 规则的几何图案，柱面靶标1放置在地面上，摄像机2为装有窄带滤光片的广 角工业相机，摄像机支架3为通用摄像机支架，摄像机支架3放置在地面上， 摄像机2通过底部的螺纹孔与摄像机支架3的螺栓螺纹固定连接，二维靶标板4 为一块矩形钢板制成的零件，且表面贴有规则的LED发光点阵，激光器连接件5 由一块加工有细长凹槽的钢板与一段钢管焊接而成，二维靶标板4置入激光器 连接件5钢板的细长凹槽中与激光器连接件5焊接连接，激光器6为可发射激 光平面的圆柱形零件，激光器6发出的激光波长、二维靶标板4的LED发光点 阵的波长与摄像机2的窄带滤光片的带通波长一致，激光器6插入激光器连接 件5钢管的内孔中与激光器连接件5过盈配合连接。The cylindrical target 1 is a semi-cylindrical part made by cutting a standard round tube in the axial direction, and the inner surface is pasted with a regular geometric pattern. The cylindrical target 1 is placed on the ground, and the camera 2 is equipped with a narrow-band filter. Wide-angle industrial camera, the camera bracket 3 is a general camera bracket, the camera bracket 3 is placed on the ground, the camera 2 is fixedly connected with the bolts of the camera bracket 3 through the threaded holes at the bottom, and the two-dimensional target plate 4 is a rectangular steel plate. , and the surface is affixed with a regular LED light-emitting lattice, the laser connector 5 is welded by a steel plate with a slender groove and a section of steel pipe, and the two-dimensional target plate 4 is placed in the slender groove of the laser connector 5 steel plate It is connected with the laser connector 5 by welding, the laser 6 is a cylindrical part that can emit a laser plane, the wavelength of the laser light emitted by the laser 6, the wavelength of the LED light-emitting lattice of the two-dimensional target plate 4 and the band of the narrow-band filter of the camera 2 The pass wavelengths are the same, and the laser 6 is inserted into the inner hole of the steel pipe of the laser connector 5 and connected with the laser connector 5 by interference fit.

参阅图7至图10，基于无约束光平面的汽车形貌主动视觉检测方法可分为 以下三步：Referring to Figures 7 to 10, the active vision detection method for vehicle topography based on an unconstrained light plane can be divided into the following three steps:

第一步：基于无约束光平面的汽车形貌主动视觉检测方法的图像采集：Step 1: Image acquisition based on an unconstrained light plane-based active vision detection method for vehicle topography:

将摄像机支架3放置在地面上，摄像机2固定在摄像机支架3上，柱面靶 标1放置在地面上,激光器6固定在激光器连接件5上，打开激光器6，激光器 6发出的激光平面与柱面靶标1交于一条激光曲线，摄像机2采集一幅图像，图 像包含柱面靶标1、二维靶标板4以及激光器6与柱面靶标1相交的投影激光曲 线；Put the camera bracket 3 on the ground, the camera 2 is fixed on the camera bracket 3, the cylindrical target 1 is placed on the ground, the laser 6 is fixed on the laser connector 5, and the laser 6 is turned on. The target 1 intersects a laser curve, the camera 2 collects an image, and the image includes the cylindrical target 1, the two-dimensional target plate 4, and the projected laser curve of the intersection of the laser 6 and the cylindrical target 1;

第二步：基于无约束光平面的汽车形貌主动视觉检测方法的在二维靶标板4 坐标系下激光面的重建：Step 2: Reconstruction of the laser surface under the 4-coordinate system of the two-dimensional target plate based on the active vision detection method of vehicle topography based on the unconstrained light plane:

首先，选取摄像机2坐标系作为固定的全局坐标系，根据摄像机2采集的 柱面靶标1特征点的图像，柱面靶标1上特征点世界坐标

Figure BDA0001944618610000061
到图像坐标
Figure BDA0001944618610000062
的投 影关系为First, the camera 2 coordinate system is selected as a fixed global coordinate system, and according to the image of the feature points of the cylindrical target 1 collected by the camera 2, the world coordinates of the feature points on the cylindrical target 1
Figure BDA0001944618610000061
to image coordinates
Figure BDA0001944618610000062
The projection relation is

Figure BDA0001944618610000063
Figure BDA0001944618610000063

其中，s1为比例因子，采用DLT方法解算从柱面靶标1坐标系到摄像机2图像坐 标系的投影矩阵PQI，对PQI进行RQ分解得到从柱面靶标1坐标系到摄像机2坐 标系的旋转矩阵RRC和平移向量tRC；Among them, s 1 is the scale factor, the DLT method is used to solve the projection matrix P QI from the coordinate system of the cylindrical target 1 to the image coordinate system of the camera 2, and the RQ decomposition of P QI is used to obtain the coordinates from the coordinate system of the cylindrical target 1 to the camera 2. the rotation matrix R RC and the translation vector t RC of the system;

然后，根据摄像机2采集的柱面靶标1上的激光线的图片，采用Steger算 法换提取图片中投影激光曲线上的点的二维图像坐标

Figure BDA0001944618610000064
Then, according to the picture of the laser line on the cylindrical target 1 collected by the camera 2, the Steger algorithm is used to extract the two-dimensional image coordinates of the point on the projected laser curve in the picture.
Figure BDA0001944618610000064

从柱面靶标1坐标系到摄像机2坐标系的转换关系为The conversion relationship from the coordinate system of cylindrical target 1 to the coordinate system of camera 2 is:

Figure BDA0001944618610000065
Figure BDA0001944618610000065

其中，s2为比例因子；Among them, s 2 is the scale factor;

柱面坐标系下

Figure BDA0001944618610000066
满足In cylindrical coordinate system
Figure BDA0001944618610000066
Satisfy

Figure BDA0001944618610000067
Figure BDA0001944618610000067

其中，r为柱面靶标1的圆柱半径，

Figure BDA0001944618610000068
为向量
Figure BDA0001944618610000069
的第一个和第二个分量，由以上两式可得激光曲线上的点在柱面靶标1坐标系下的坐标
Figure BDA00019446186100000610
Among them, r is the cylindrical radius of the cylindrical target 1,
Figure BDA0001944618610000068
as a vector
Figure BDA0001944618610000069
The first and second components of , the coordinates of the point on the laser curve in the coordinate system of cylindrical target 1 can be obtained from the above two formulas
Figure BDA00019446186100000610

根据从柱面靶标1坐标系到摄像机2坐标系的旋转矩阵RQC和平移向量tQC， 以及激光线上的点在柱面靶标1坐标系下的坐标

Figure BDA00019446186100000611
由下式可求得激光线上的 点在摄像机2坐标系下的坐标为According to the rotation matrix R QC and translation vector t QC from the cylindrical target 1 coordinate system to the camera 2 coordinate system, and the coordinates of the point on the laser line under the cylindrical target 1 coordinate system
Figure BDA00019446186100000611
The coordinates of the point on the laser line in the camera 2 coordinate system can be obtained from the following formula:

Figure BDA00019446186100000612
Figure BDA00019446186100000612

激光平面上的点

Figure BDA00019446186100000613
满足point on the laser plane
Figure BDA00019446186100000613
Satisfy

Figure BDA00019446186100000614
Figure BDA00019446186100000614

由SVD分解法求得激光平面在摄像机2坐标系下的坐标πC；Obtain the coordinate π C of the laser plane in the camera 2 coordinate system by the SVD decomposition method;

由张正友二维标定算法可以解算从二维靶标板4坐标系到摄像机2坐标系 的旋转矩阵RRC和平移向量tRC，则从二维靶标板4坐标系到摄像机2坐标系的单 应矩阵为The two-dimensional calibration algorithm of Zhang Zhengyou can solve the rotation matrix R RC and translation vector t RC from the two-dimensional target board 4 coordinate system to the camera 2 coordinate system, then the homography from the two-dimensional target board 4 coordinate system to the camera 2 coordinate system The matrix is

Figure BDA00019446186100000615
Figure BDA00019446186100000615

根据激光平面在摄像机2坐标系下的坐标πC和二维靶标板4坐标系到摄像 机2坐标系的单应矩阵HRC，在二维靶标板4坐标系下激光平面坐标为According to the coordinate π C of the laser plane in the coordinate system of the camera 2 and the homography matrix H RC from the coordinate system of the two-dimensional target plate 4 to the coordinate system of the camera 2, the coordinates of the laser plane in the coordinate system of the two-dimensional target plate 4 are:

Figure BDA0001944618610000071
Figure BDA0001944618610000071

第三步：基于无约束光平面的汽车形貌主动视觉检测方法的车辆表面特征 点的重建：Step 3: Reconstruction of vehicle surface feature points based on unconstrained light plane active vision detection method for vehicle topography:

将摄像机支架3放置在地面上，摄像机2固定在摄像机支架3上，汽车驶 入摄像机2的视场内，激光器6固定在激光器连接件5上，打开激光器6，调整 二维靶标板4的位置及角度使激光器6发出的激光平面与车辆表面交于一条激 光线，使用摄像机2采集一幅图像，图像包含二维靶标板4以及激光器6与车 辆表面相交的投影激光直线，然后将二维靶标板4放置在第q位置上，摄像机2 继续采集第q位置的图像，q＝1,2，…，n；Place the camera bracket 3 on the ground, fix the camera 2 on the camera bracket 3, drive the car into the field of view of the camera 2, fix the laser 6 on the laser connector 5, turn on the laser 6, and adjust the position of the two-dimensional target plate 4 and the angle so that the laser plane emitted by the laser 6 intersects the surface of the vehicle with a laser line, and the camera 2 is used to collect an image, the image includes the two-dimensional target plate 4 and the projected laser line intersecting the laser 6 and the vehicle surface, and then the two-dimensional target The board 4 is placed at the qth position, and the camera 2 continues to capture the image of the qth position, q=1,2,...,n;

根据张正友二维标定算法可以得到从二维靶标板4坐标系到摄像机2坐标 由可求得从二维靶标板4坐标系到摄像机2坐标系的旋转矩阵RRC,q和平移向量 tRC,q以及摄像机2内参数K，则从二维靶标板4坐标系到摄像机2坐标系的单应 矩阵为According to Zhang Zhengyou's two-dimensional calibration algorithm, it can be obtained from the two-dimensional target board 4 coordinate system to the camera 2 coordinate system. The rotation matrix R RC, q and translation vector t RC can be obtained from the two-dimensional target board 4 coordinate system to the camera 2 coordinate system, q and the internal parameter K of the camera 2, the homography matrix from the coordinate system of the two-dimensional target plate 4 to the coordinate system of the camera 2 is

Figure BDA0001944618610000072
Figure BDA0001944618610000072

由第二步求出的二维靶标板4坐标系下激光平面坐标πR以及二维靶标板4坐标 系到摄像机2坐标系的单应矩阵HRC,q，摄像机2坐标系下激光平面坐标为The laser plane coordinates π R in the two-dimensional target plate 4 coordinate system obtained from the second step and the homography matrix H RC,q from the two-dimensional target plate 4 coordinate system to the camera 2 coordinate system, the laser plane coordinates in the camera 2 coordinate system for

Figure BDA0001944618610000073
Figure BDA0001944618610000073

其中，

Figure BDA0001944618610000074
in,
Figure BDA0001944618610000074

在摄像机2坐标系下激光线上的点的三维坐标

Figure BDA0001944618610000075
满足The three-dimensional coordinates of the point on the laser line in the camera 2 coordinate system
Figure BDA0001944618610000075
Satisfy

Figure BDA0001944618610000076
Figure BDA0001944618610000076

由摄像机2获取的图像可得二维靶标板4以及激光器6与车辆表面相交的 投影激光线上点的图像坐标

Figure BDA0001944618610000077
图像坐标
Figure BDA0001944618610000078
和在摄像机2坐标系下激光线上 的点的三维坐标
Figure BDA0001944618610000079
转换关系为The image obtained by the camera 2 can obtain the two-dimensional target plate 4 and the image coordinates of the point on the projected laser line where the laser 6 intersects the vehicle surface
Figure BDA0001944618610000077
image coordinates
Figure BDA0001944618610000078
and the three-dimensional coordinates of the point on the laser line in the camera 2 coordinate system
Figure BDA0001944618610000079
The conversion relationship is

Figure BDA0001944618610000081
Figure BDA0001944618610000081

其中，s3为比例因子，K为摄像机2内参数；Among them, s 3 is the scale factor, and K is the internal parameter of camera 2;

根据以上三式及SVD分解法解算在摄像机2坐标系下激光平面与车身交点 的三维坐标

Figure BDA0001944618610000082
实现车身三维特征点的重建。According to the above three equations and the SVD decomposition method, the three-dimensional coordinates of the intersection of the laser plane and the vehicle body in the camera 2 coordinate system are calculated
Figure BDA0001944618610000082
Realize the reconstruction of three-dimensional feature points of the car body.

Claims (7)

1. An automobile morphology active visual detection method based on an unconstrained light plane is characterized by comprising a cylindrical target (1), a camera (2), a camera bracket (3), a two-dimensional target plate (4), a laser connecting piece (5) and a laser (6);
the cylindrical surface target (1) is placed on the ground, the camera support (3) is placed on the ground, the camera (2) is fixedly connected with a bolt thread of the camera support (3) through a threaded hole in the bottom, the two-dimensional target plate (4) is placed in a long and thin groove of a steel plate of the laser connecting piece (5) and is connected with the laser connecting piece (5) in a welding mode, and the laser (6) is inserted into an inner hole of the steel pipe of the laser connecting piece (5) and is connected with the laser connecting piece (5) in an interference fit mode;
the method comprises the following specific steps:
the first step is as follows: the method comprises the following steps of image acquisition of an automobile morphology active visual detection method based on an unconstrained light plane:
placing a camera support (3) on the ground, fixing a camera (2) on the camera support (3), placing a cylindrical target (1) on the ground, fixing a laser (6) on a laser connecting piece (5), opening the laser (6), enabling a laser plane emitted by the laser (6) to intersect with the cylindrical target (1) on a laser curve, and acquiring an image by the camera (2), wherein the image comprises the cylindrical target (1), a two-dimensional target plate (4) and a projection laser curve formed by the intersection of the laser (6) and the cylindrical target (1);
the second step is that: the reconstruction of the laser surface under a two-dimensional target board (4) coordinate system based on the active visual detection method of the automobile morphology of the unconstrained light plane is as follows:
firstly, selecting a coordinate system of a camera (2) as a fixed global coordinate system, and according to an image of a characteristic point of a cylindrical target (1) acquired by the camera (2) and a world coordinate of the characteristic point on the cylindrical target (1)
Figure FDA0002356548120000011
To the image coordinates
Figure FDA0002356548120000012
Has a projection relation of
Figure FDA0002356548120000013
Wherein s is1For a scale factor, a projection matrix P from a cylindrical target (1) coordinate system to a camera (2) image coordinate system is solved by adopting a DLT methodQITo PQIRQ decomposition is carried out to obtain a rotation matrix R from a cylindrical target (1) coordinate system to a camera (2) coordinate systemRCAnd a translation vector tRC；
Then, according to the picture of the laser line on the cylindrical target (1) acquired by the camera (2), a Steger algorithm is adopted to extract the two-dimensional image coordinates of the points on the projection laser curve in the picture
Figure FDA0002356548120000014
The conversion relation from the coordinate system of the cylindrical target (1) to the coordinate system of the camera (2) is
Figure FDA0002356548120000015
Wherein s is2Is a scale factor;
under the cylindrical coordinate system
Figure FDA0002356548120000016
Satisfy the requirement of
Figure FDA0002356548120000017
Wherein r is the cylindrical radius of the cylindrical target (1),
Figure FDA0002356548120000018
is a vector
Figure FDA0002356548120000019
The coordinates of the points on the laser curve in the coordinate system of the cylindrical target (1) can be obtained by the above two formulas
Figure FDA00023565481200000110
According to a rotation matrix R from the cylindrical target (1) coordinate system to the camera (2) coordinate systemQCAnd a translation vector tQCAnd the coordinates of the point on the laser line in the coordinate system of the cylindrical target (1)
Figure FDA00023565481200000111
The coordinates of the point on the laser line in the coordinate system of the camera (2) can be obtained as
Figure FDA0002356548120000021
Points on the laser plane
Figure FDA0002356548120000022
Satisfy the requirement of
Figure FDA0002356548120000023
The coordinate pi of the laser plane under the coordinate system of the camera (2) is obtained by an SVD decomposition methodC；
The rotation matrix R from the coordinate system of the two-dimensional target board (4) to the coordinate system of the camera (2) can be solved by a Zhang-Zhengyou two-dimensional calibration algorithmRCAnd a translation vector tRCThe homography matrix from the two-dimensional target plate (4) coordinate system to the camera (2) coordinate system is
Figure FDA0002356548120000024
According to laser planeCoordinate pi of camera (2) coordinate systemCAnd a homography matrix H from the coordinate system of the two-dimensional target plate (4) to the coordinate system of the camera (2)RCThe laser plane coordinate under the coordinate system of the two-dimensional target plate (4) is
Figure FDA0002356548120000025
The third step: reconstructing vehicle surface characteristic points based on an automobile morphology active visual detection method of an unconstrained light plane:
placing a camera support (3) on the ground, fixing a camera (2) on the camera support (3), driving an automobile into a view field of the camera (2), fixing a laser (6) on a laser connecting piece (5), opening the laser (6), adjusting the position and the angle of a two-dimensional target plate (4) to enable a laser plane emitted by the laser (6) to intersect with the surface of the vehicle to form a laser line, acquiring an image by using the camera (2), wherein the image comprises the two-dimensional target plate (4) and a projection laser line intersecting the surface of the vehicle by the laser (6), then placing the two-dimensional target plate (4) on a q-th position, and continuously acquiring an image of the q-th position by using the camera (2), wherein q is 1,2, … and n;
according to the Zhang Zhengyou two-dimensional calibration algorithm, a rotation matrix R from the two-dimensional target board (4) coordinate system to the camera (2) coordinate system can be obtained, and then the rotation matrix R from the two-dimensional target board (4) coordinate system to the camera (2) coordinate system can be obtainedRC,qAnd a translation vector tRC,qAnd the camera (2) internal parameter K, the homography matrix from the two-dimensional target board (4) coordinate system to the camera (2) coordinate system is
Figure FDA0002356548120000026
The laser plane coordinate pi under the coordinate system of the two-dimensional target plate (4) calculated by the second stepRAnd a homography matrix H from the coordinate system of the two-dimensional target plate (4) to the coordinate system of the camera (2)RC,qThe laser plane coordinate under the coordinate system of the camera (2) is
Figure FDA0002356548120000031
Wherein,
Figure FDA0002356548120000032
three-dimensional coordinates of points on the laser line in the camera (2) coordinate system
Figure FDA0002356548120000033
Satisfy the requirement of
Figure FDA0002356548120000034
The image obtained by the camera (2) can obtain the image coordinates of the two-dimensional target board (4) and the point on the projection laser line of the laser (6) intersected with the surface of the vehicle
Figure FDA0002356548120000035
Image coordinates
Figure FDA0002356548120000036
And the three-dimensional coordinates of the points on the laser line in the camera (2) coordinate system
Figure FDA0002356548120000037
The conversion relation is
Figure FDA0002356548120000038
Wherein s is3Is a scale factor, and K is an internal parameter of the camera (2);
according to the three formulas and the SVD decomposition method, the three-dimensional coordinate of the intersection point of the laser plane and the vehicle body under the coordinate system of the camera (2) is solved
Figure FDA0002356548120000039
And realizing the reconstruction of the three-dimensional characteristic points of the vehicle body.
2. The active visual inspection method for automobile morphology based on unconstrained light plane as claimed in claim 1, characterized in that the cylindrical target (1) is a semi-cylindrical part made by cutting a standard circular tube along the axial direction, and the inner surface is pasted with a regular geometric pattern.
3. The active visual inspection method of car morphology based on unconstrained light plane according to claim 1, characterized in that said video camera (2) is a wide-angle industrial camera equipped with a narrow-band filter.
4. The active visual inspection method of car topography based on unconstrained light planes according to claim 1, characterized in that said camera mount (3) is a universal camera mount.
5. The active visual inspection method for automobile morphology based on unconstrained light plane as claimed in claim 1, characterized in that the two-dimensional target board (4) is a part made of a rectangular steel plate, and a regular LED light emitting lattice is pasted on the surface, and the wavelength of the LED light emitting lattice of the two-dimensional target board (4) is consistent with the band-pass wavelength of the narrow-band filter of the camera (2).
6. The active visual inspection method of the morphology of an automobile based on an unconstrained light plane according to claim 1, characterized in that the laser connecting piece (5) is formed by welding a steel plate with a long and thin groove and a section of steel pipe.
7. The active visual inspection method of car morphology based on unconstrained light plane according to claim 1, characterized in that said laser (6) is a cylindrical part capable of emitting laser plane, and the laser wavelength emitted by the laser (6) is consistent with the band-pass wavelength of the narrow-band filter of the camera (2).
CN201910038238.6A 2019-01-14 2019-01-14 Automobile morphology active vision detection system and method based on unconstrained light plane Expired - Fee Related CN109540042B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910038238.6A CN109540042B (en) 2019-01-14 2019-01-14 Automobile morphology active vision detection system and method based on unconstrained light plane

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910038238.6A CN109540042B (en) 2019-01-14 2019-01-14 Automobile morphology active vision detection system and method based on unconstrained light plane

Publications (2)

Publication Number Publication Date
CN109540042A CN109540042A (en) 2019-03-29
CN109540042B true CN109540042B (en) 2020-05-01

Family

ID=65835541

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910038238.6A Expired - Fee Related CN109540042B (en) 2019-01-14 2019-01-14 Automobile morphology active vision detection system and method based on unconstrained light plane

Country Status (1)

Country Link
CN (1) CN109540042B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111380481B (en) * 2019-12-31 2024-06-07 吉林大学 System and method for monocular active visual reconstruction of automobile morphology based on five-point identification
CN111412854B (en) * 2019-12-31 2024-07-30 吉林大学 System and method for active visual reconstruction of automobile morphology based on invariant of points and circles
CN111174725B (en) * 2019-12-31 2024-06-04 吉林大学 Coaxial-earth-connection monocular reconstruction system and method for polar line geometry of concentric secondary curve
CN111412857B (en) * 2019-12-31 2024-12-13 吉林大学 Automobile shape reconstruction system and method based on quadratic curves dual to imaginary points

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102980528A (en) * 2012-11-21 2013-03-20 上海交通大学 Calibration method of pose position-free constraint line laser monocular vision three-dimensional measurement sensor parameters
CN104180775A (en) * 2014-08-05 2014-12-03 吉林大学 Cylindrical surface coordinate based camera calibration target of automobile vision detection system
CN105698699A (en) * 2016-01-26 2016-06-22 大连理工大学 A binocular visual sense measurement method based on time rotating shaft constraint
US9488469B1 (en) * 2013-04-22 2016-11-08 Cognex Corporation System and method for high-accuracy measurement of object surface displacement using a laser displacement sensor
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
CN209230559U (en) * 2019-01-14 2019-08-09 吉林大学 Active vision inspection system for vehicle shape based on unconstrained light plane

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102980528A (en) * 2012-11-21 2013-03-20 上海交通大学 Calibration method of pose position-free constraint line laser monocular vision three-dimensional measurement sensor parameters
US9488469B1 (en) * 2013-04-22 2016-11-08 Cognex Corporation System and method for high-accuracy measurement of object surface displacement using a laser displacement sensor
CN104180775A (en) * 2014-08-05 2014-12-03 吉林大学 Cylindrical surface coordinate based camera calibration target of automobile vision detection system
CN105698699A (en) * 2016-01-26 2016-06-22 大连理工大学 A binocular visual sense measurement method based on time rotating shaft constraint
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
CN209230559U (en) * 2019-01-14 2019-08-09 吉林大学 Active vision inspection system for vehicle shape based on unconstrained light plane

Also Published As

Publication number Publication date
CN109540042A (en) 2019-03-29

Similar Documents

Publication Publication Date Title
CN109540042B (en) Automobile morphology active vision detection system and method based on unconstrained light plane
CN109540033B (en) Closed-loop consistent binocular active vision automobile morphology global free reconstruction system and method
CN109540040B (en) Active vision detection system and method based on unconstrained concentric beam family automobile morphology
CN209230559U (en) Active vision inspection system for vehicle shape based on unconstrained light plane
WO2022188293A1 (en) Visual inspection system with integrated lens
CN214408670U (en) Visual inspection system with integrated camera lens
CN211626384U (en) Automobile detection non-common-view-field camera global calibration system based on surface light field
CN111412854A (en) Point and circle invariant based active vision reconstruction system and method for automobile morphology
CN209399945U (en) Active visual inspection system for vehicle shape based on unconstrained concentric beam family
CN111242901A (en) Space point-based global calibration system and method for automobile detection camera without common view field
CN108253929A (en) A kind of four-wheel position finder, system and its implementation
CN208207916U (en) Color calibration device and three dimensional reconstructive CT system
CN211060854U (en) Automobile morphology visual detection system based on unchanged angle similarity
CN211042116U (en) Active Vision Reconstruction System for Vehicle Topography Based on Point and Circle Invariants
CN211060856U (en) Automobile morphology active vision reconstruction system based on straight line and circle invariants
CN211060859U (en) Automobile morphology reconstruction system based on triangular array affine invariant
CN207504970U (en) A kind of servo-actuated camera using ultrasonic ranging
CN112484663A (en) Intelligent visual 3D information acquisition equipment of many angles of rolling
CN211626404U (en) Shaft end grounding monocular reconstruction system of concentric quadratic curve polar line geometry
CN111412858A (en) Automobile morphology visual detection system and method based on unchanged angle similarity
CN111380480B (en) Automobile morphology reconstruction system and method based on affine invariant of triangular array
CN211060860U (en) Vehicle Topography Reconstruction System Based on Dual Quadratic Curve with Imaginary Circle
CN111412855B (en) Active visual reconstruction system and method of automobile shape based on point and line invariants
CN111380482B (en) Automobile morphology active vision reconstruction system and method based on straight line and circle invariants
CN211060861U (en) Automobile morphology active vision reconstruction system based on point and straight line invariant

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20200501

Termination date: 20220114

CF01 Termination of patent right due to non-payment of annual fee