WO2021179400A1 - 一种基于计算机视觉的装配过程几何参数自适应测量系统及方法 - Google Patents

一种基于计算机视觉的装配过程几何参数自适应测量系统及方法 Download PDF

Info

Publication number
WO2021179400A1
WO2021179400A1 PCT/CN2020/085248 CN2020085248W WO2021179400A1 WO 2021179400 A1 WO2021179400 A1 WO 2021179400A1 CN 2020085248 W CN2020085248 W CN 2020085248W WO 2021179400 A1 WO2021179400 A1 WO 2021179400A1
Authority
WO
WIPO (PCT)
Prior art keywords
measurement
image
module
server
camera
Prior art date
Application number
PCT/CN2020/085248
Other languages
English (en)
French (fr)
Inventor
晏玉祥
邵立
白晓亮
常壮
Original Assignee
南京翱翔信息物理融合创新研究院有限公司
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 南京翱翔信息物理融合创新研究院有限公司 filed Critical 南京翱翔信息物理融合创新研究院有限公司
Publication of WO2021179400A1 publication Critical patent/WO2021179400A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • G06F3/013—Eye tracking input arrangements
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/30—Noise filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2203/00—Indexing scheme relating to G06F3/00 - G06F3/048
    • G06F2203/01—Indexing scheme relating to G06F3/01
    • G06F2203/012—Walk-in-place systems for allowing a user to walk in a virtual environment while constraining him to a given position in the physical environment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection

Definitions

  • the invention relates to an image processing and recognition technology for self-adaptive measurement of geometric parameters in an assembly process, in particular to a computer vision-based system and method for self-adaptive measurement of geometric parameters in an assembly process.
  • the purpose of the present invention is to address the above-mentioned shortcomings of the prior art, and provide a computer vision-based assembly process geometric parameter adaptive measurement system and method.
  • the camera in the present invention can adaptively adjust the shooting position according to the measurement instructions transmitted by the server. Narrow the measurement range, reduce unnecessary interference, make the measurement more accurate, and improve the measurement efficiency and quality.
  • a computer vision-based adaptive measurement system for geometric parameters of the assembly process includes a process design client unit, an image acquisition system, an image processing system, a tracking system, a visualization system, and a server.
  • the process design client unit connects Process information input server, the server inputs measurement instructions into the image acquisition system via the network, the image acquisition system inputs the image of the part to be measured into the image processing system via the network, and the image processing system inputs the processed measurement parameter information into the server
  • the server inputs the measured data into the visualization system via the network
  • the tracking system inputs the measurement instructions into the image acquisition system via the network
  • the image acquisition system inputs the image of the part to be measured into the image processing system via the network.
  • the processing system inputs the processed measurement parameter information into the server, and the server inputs the measured data into the visualization system through the network.
  • the image acquisition system is divided into a software system and a hardware system.
  • the software system includes a global image acquisition module, a measurement instruction module, and a local image acquisition module.
  • the global image acquisition module is connected to the measurement instruction module through a line, and the measurement instruction The module is connected with the local image acquisition module through a line.
  • the hardware system includes a wide-angle camera, a telephoto camera and a rotating pan/tilt. The telephoto camera is installed on the rotating pan/tilt.
  • the image processing system includes an image enhancement noise processing module, an image segmentation feature extraction module, an object recognition parameter estimation module, and a measurement parameter calculation module.
  • the image enhancement noise processing module is connected to the image segmentation feature extraction module through a data pipeline.
  • the segmentation feature extraction module is connected to the object recognition parameter estimation module through the data pipeline, and the object recognition parameter estimation module is connected to the measurement parameter calculation module through the data pipeline.
  • the partial image acquisition module is connected to the image enhancement noise processing module through a data pipeline.
  • the tracking system includes an eye tracking module and a gesture tracking module.
  • the present invention also provides an adaptive measurement method for geometric parameters of the assembly process based on computer vision, the steps are as follows:
  • Camera installation and debugging installation of wide-angle camera and telephoto camera, and binocular camera (camera pair composed of two wide-angle cameras placed parallel to the optical axis, or two telephoto cameras placed parallel to the optical axis
  • the calibration of the camera pair is used to determine the relationship between the image pixels and the actual length.
  • the height of the wide-angle camera is fixed so that the wide-angle camera is directly facing the assembly scene below, so that the entire assembly process can be clearly and comprehensively observed.
  • Server measurement instruction input the server inputs the measurement location instruction to the image acquisition system to determine the location that the camera needs to measure;
  • the tracking system inputs the measurement position instruction designated by human interaction to the image acquisition system to determine the position that the camera needs to measure;
  • Image processing and parameter calculation The image processing system completes the calculation of the parameters of the measurement area through image enhancement noise processing, image segmentation feature extraction and object recognition parameter estimation;
  • the server receives the measurement parameter information output by the image processing system, compares it with the process information in the database, and distributes the comparison results to the visualization system for visualization and other processing.
  • step (2) The specific steps of step (2) are as follows: (a) The process design client unit sends the assembly scene and the parameter requirement data to be measured to the server, where the process design client unit includes a resource packaging module, a scene data formal processing module and a process Data formal processing module; (b) The server integrates the assembly resource data package and sends the required data of the parameters to be measured to the image acquisition system of the measurement client; (c) The image acquisition system determines the key parameters and measurements to be measured based on the data to be measured Location.
  • step (3) The specific steps of step (3) are: (a) After the image acquisition system receives the data to be measured, it determines the measurement and measurement position of the key parameters, and then fixes the telephoto camera so that the telephoto camera faces the measurement position and can Observe the measurement area clearly; (b) In the field of view of the wide-angle camera, add a two-dimensional code on the surface of the part, identify and locate and track the part by identifying the mark point of the two-dimensional code. When it is detected that the part is close to a certain neighborhood of the measurement position When the camera is switched automatically, the telephoto camera facing the measurement position is turned on for measurement detection, and the measurement area is further reduced by frame selection and accurate measurement.
  • step (4) The specific steps of step (4) are: (a) eye tracking to determine the measurement location and area; (b) gesture tracking to determine the measurement location and area.
  • step (5) The specific steps of step (5) are: (a) After the image in the measurement area is obtained, reduce the noise of the image through image gray-scale processing or Gaussian filter processing; (b) extraction of linear features; (c) elliptical features (D) Object recognition; (e) Measurement and calculation of parameters; (f) Convert the measured pixel parameter information into actual distance parameter information and output it to the server.
  • the camera can adaptively adjust the measurement position, accurately measure the area and reduce the measurement range, realize the accuracy of the measurement, reduce the calculation of the computer, improve the measurement efficiency and realize real-time detection and measurement;
  • Figure 1 is a schematic diagram of the computer vision system
  • Figure 2 is a schematic diagram of the composition of an adaptive measurement system
  • Figure 3 is a schematic diagram of the formalized representation of data.
  • the distance between the gauge blocks installed in the engine casing is 0.5mm ⁇ 1mm.
  • the geometric parameter adaptive measurement system of the assembly process of computer vision is adopted to guide the workers to install, so that the installation of the gauge blocks in the engine casing meets the requirements of the assembly process. Then follow the steps below to run:
  • Camera installation and debugging installation of wide-angle camera and telephoto camera, and calibration of binocular camera, used to determine the relationship between image pixels and actual length, fix the height of the wide-angle camera so that the wide-angle camera is directly facing the assembly scene below , Can observe the whole assembly process clearly and comprehensively, fix the telephoto camera on the rotating pan-tilt, adjust the position of the rotating pan-tilt so that the telephoto camera is facing the key position of the assembly process that needs to be measured, and the measurement area can be clearly observed ;
  • Server measurement instruction input the server inputs the measurement location instruction to the image acquisition system to determine the location that the camera needs to measure;
  • the server integrates the assembly resource data package and sends the required data of the parameters to be measured to the image acquisition system of the measurement client;
  • the image acquisition system determines the key parameters to be measured and the measurement location based on the data to be measured;
  • the image acquisition system After the image acquisition system receives the data that needs to be measured, it determines the measurement and measurement location of the key parameters, and then fixes the telephoto camera so that the telephoto camera faces the measurement location and can clearly observe the measurement area;
  • the tracking system inputs the measurement position instruction designated by human interaction to the image acquisition system to determine the position that the camera needs to measure;
  • Gesture tracking determines the measurement location and area, and uses LeapMotion for gesture tracking
  • Image processing and parameter calculation The image processing system completes the calculation of the parameters of the measurement area through image enhancement noise processing, image segmentation feature extraction and object recognition parameter estimation;
  • Object recognition for example: Recognizing part A, part B and part C in the measurement area by recognizing the two-dimensional code, and adaptively measure the parameters of part A and part B according to the measurement needs, or measure the part B and Parameters of part C;
  • the server receives the measurement parameter information output by the image processing system, compares it with the process information in the database, and distributes the comparison results to the visualization system for visualization and other processing.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Geometry (AREA)
  • Human Computer Interaction (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Image Analysis (AREA)

Abstract

一种基于计算机视觉的装配过程几何参数自适应测量系统，包括有工艺设计客户端单元、图像采集系统、图像处理系统、追踪系统、可视化系统和服务器。通过服务器传入的测量指令，相机可以自适应地调整测量位置，精确测量区域并缩小测量范围，实现测量的精确化，减少计算机的计算量，提高测量效率并实现实时检测测量。

Description

一种基于计算机视觉的装配过程几何参数自适应测量系统及方法 技术领域
本发明涉及图像处理及识别技术自适应测量装配过程中的几何参数，具体为一种基于计算机视觉的装配过程几何参数自适应测量系统及方法。
背景技术
在工业生产中，传统的形状尺寸检测手段有卡尺、量规、万能工具显微镜、轮廓仪和X射线等，这些检测手段在工业生产中发挥着巨大的作用，但是随着现代工业的发展和进步，特别是在一些高精产业，传统的检测手段已经不能满足生产的需要，卡尺和量规等检测手段虽然简便、快捷，但测量数据较少，精度不高。万能工具显微镜及轮廓仪等检测手段虽然有较高的精度，但要求在特定的设备、特定的环境下进行检测，不但劳动强度大，而且效率低，传统的检测测量技术需要众多的检测工人，不仅影响生产效率，而且带来不可靠因素。而计算机视觉检测测量技术克服了传统检测技术的缺点，它以检测的安全性、可靠性及自动化程度高等优点而得到广泛的应用，成为当今检测技术的研究热点之一。
发明内容
本发明的目的就是针对上述现有技术的不足，提供一种基于计算机视觉的装配过程几何参数自适应测量系统及方法，该发明中的相机可以根据服务器传入的测量指令自适应调整拍摄位置，缩小测量范围，减小不必要的因素干扰，使测量更加精确化，提升测量效率和质量。
本发明采用的技术方案如下：
一种基于计算机视觉的装配过程几何参数自适应测量系统，它包括有工艺设计客户端单元、图像采集系统、图像处理系统、追踪系统、可视化系统和服务器，所述工艺设计客户端单元通过网络将工艺信息输入服务器，所述服务器通过网络将测量指令输入图像采集系统，所述图像采集系统通过网络将需测量部分的图像输入图像处理系统，所述图像处理系统将处理后的测量参数信息输入服务器，所述服务器通过网络将测量后的数据输入可视化系 统，所述追踪系统通过网络将测量指令输入图像采集系统，所述图像采集系统通过网络将需测量部分的图像输入图像处理系统，所述图像处理系统将处理后的测量参数信息输入服务器，所述服务器通过网络将测量后的数据输入可视化系统。
所述图像采集系统分为软件系统和硬件系统，所述软件系统包括全局图像采集模块、测量指令模块和局部图像采集模块，所述全局图像采集模块通过线路与测量指令模块相连，所述测量指令模块通过线路与局部图像采集模块相连，所述硬件系统包括广角相机、长焦相机和转动云台，所述长焦相机安装在转动云台上。
所述图像处理系统包括图像增强噪声处理模块、图像分割特征提取模块、物体识别参数估计模块和测量参数计算模块，所述图像增强噪声处理模块通过数据管道与图像分割特征提取模块相连，所述图像分割特征提取模块通过数据管道与物体识别参数估计模块相连，所述物体识别参数估计模块通过数据管道与测量参数计算模块相连。
所述局部图像采集模块通过数据管道与图像增强噪声处理模块相连。
所述追踪系统包括眼动追踪模块和手势追踪模块。
本发明还提供一种基于计算机视觉的装配过程几何参数自适应测量方法，步骤如下：
(1)相机安装和调试：广角相机和长焦相机的安装以及双目相机(两个广角相机光轴平行地放置在一起组成的相机对，或者两个长焦相机光轴平行地放置在一起组成的相机对)的标定，用于确定图像像素和实际长度之间的关系，固定广角相机的高度，使广角相机正对下方装配场景，能清晰全面地观测到整个装配过程，在转动云台上固定长焦相机，调整转动云台的位置，使长焦相机正对装配过程关键处需要测量的位置，并能清晰观测到测量区域；
(2)服务器测量指令的输入：服务器将测量位置指令输入到图像采集系统，用于确定相机需要测量的位置；
(3)测量区域的确定：图像采集系统接收需要测量的数据后，相机自适应地调整测量区域，用于确定测量位置的区域和大小；
(4)追踪系统测量指令的输入：追踪系统中将人的交互指定的测量位置指令输入到图像采集系统，用于确定相机需要测量的位置；
(5)图像处理与参数的计算：图像处理系统通过图像增强噪声处理、图像分割特征提取及物体识别参数估计完成测量区域的参数的计算；
(6)服务器测量参数的输出：服务器接收图像处理系统输出的测量参数信息，并与数据库中工艺信息进行比较，将比较结果分发到可视系统，用于可视化和其他处理。
步骤(2)的具体步骤为：(a)工艺设计客户端单元将装配场景和需测量的参数要求数据发送到服务器，其中工艺设计客户端单元包括资源封装模块，场景数据形式化处理模块和工艺数据形式化处理模块；(b)服务器整合装配资源数据包将需测量的参数要求数据发送到测量客户端的图像采集系统；(c)图像采集系统根据需测量的数据确定需要测量的关键参数及测量位置。
步骤(3)的具体步骤为：(a)图像采集系统接收到需要测量的数据后，确定了关键参数的测量及测量位置，接着将长焦相机固定，使长焦相机正对测量位置并能清晰观测到测量区域；(b)在广角相机的视野范围内，在零件表面添加二维码，通过识别二维码标志点的方式识别并定位追踪零件，当检测到零件接近测量位置一定邻域内时，相机自适应切换，打开正对测量位置的长焦相机进行测量检测，并通过框选的方式进一步减小测量区域并精确测量。
步骤(4)的具体步骤为：(a)眼动追踪确定测量位置和区域；(b)手势追踪确定测量位置和区域。
步骤(5)的具体步骤为：(a)在得到测量区域内的图像后，通过图像灰度化处理或高斯滤波处理减小图像的噪声；(b)直线特征的提取；(c)椭圆特征的提取；(d)物体识别；(e)参数的测量和计算；(f)将测量得到的像素参数信息转化为实际距离参数信息输出到服务器。
本发明的有益效果有：
(1)通过在测量场景配置一套基于计算机视觉的感知系统，实现测量场景的全局图像采集，使得操作人员能看到整个测量的场景；
(2)通过服务器传入的测量指令，相机可以自适应地调整测量位置，精确测量区域并缩小测量范围，实现测量的精确化，减少计算机的计算量，提高测量效率并实现实时检测测量；
(3)通过在测量场景配置一套追踪系统，实现人的状态实时追踪，通过捕捉人的注视点和手势确定测量的位置区域，相机自适应地调整拍摄位置，精确测量区域并缩小测量范围，使得操作人员可以根据测量需要自由的选择测量位置区域；
(4)通过在测量场景配置一套图像处理系统，实现测量区域的物体识别和参数计算，并将测量参数信息输出到服务器，服务器再将测量参数分发到其他系统，使得测量参数信息能进行再加工处理。
附图说明
图1为计算机视觉系统组成示意图；
图2为自适应测量系统组成示意图；
图3为数据形式化表示示意图。
具体实施方式
下面结合附图对本发明作进一步地说明：
实施例：
在发动机机匣内安装的量块之间的距离为0.5mm～1mm，采用计算机视觉的装配过程几何参数自适应测量系统，指导工人进行安装，使发动机机匣内量块安装符合装配工艺要求，接着按以下步骤运行：
(1)相机安装和调试：广角相机和长焦相机的安装以及双目相机的标定，用于确定图像像素和实际长度之间的关系，固定广角相机的高度，使广角相机正对下方装配场景，能清晰全面地观测到整个装配过程，在转动云台上固定长焦相机，调整转动云台的位置，使长焦相机正对装配过程关键处需要测量的位置，并能清晰观测到测量区域；
(2)服务器测量指令的输入：服务器将测量位置指令输入到图像采集系统，用于确定相机需要测量的位置；
具体实现步骤如下：
(a)工艺设计客户端单元将装配场景和需测量的参数要求数据发送到服务器；
(b)服务器整合装配资源数据包将需测量的参数要求数据发送到测量客户端的图像采集系统；
(c)图像采集系统根据需测量的数据确定需要测量的关键参数及测量位置；
(3)测量区域的确定：图像采集系统接收需要测量的数据后，相机自适应地调整测量区域，用于确定测量位置的区域和大小；
具体实现步骤如下：
(a)图像采集系统接受到需要测量的数据后，确定了关键参数的测量及测量位置，接着将长焦相机固定，使长焦相机正对测量位置并能清晰观测到测量区域；
(b)在广角相机的视野范围内，在零件表面添加二维码，通过识别二维码标志点的方式识别并定位追踪零件，当检测到零件接近测量位置一定邻域内时，相机自适应切换，打开正对测量位置的长焦相机进行测量检测，并通过框选的方式进一步减小测量区域并精确测量；
(4)追踪系统测量指令的输入：追踪系统中将人的交互指定的测量位置指令输入到图像采集系统，用于确定相机需要测量的位置；
具体实现步骤如下：
(a)眼动追踪确定测量位置和区域，通过在HTC vive头盔显示器上加上aGlass-DK2来进行眼动追踪；
(b)手势追踪确定测量位置和区域，通过采用LeapMotion进行手势追踪；
(5)图像处理与参数的计算：图像处理系统通过图像增强噪声处理、图像分割特征提取及物体识别参数估计完成测量区域的参数的计算；
具体实现步骤如下：
(a)将得到测量区域内的图像后，通过图像灰度化处理或高斯滤波处理减小图像的噪声；
(b)直线特征的提取，例如：在获得灰度化图像后通过LSD直线检测所有直线，再根据直线的长度剔除大量无效的短小的直线筛选出所需长度的直线，为消除零件的阴影对直线检测的影响，通过筛选出来的直线端点坐标拟合一条中心线，并求每一条直线的端点到中心线距离，通过距离长短筛选出最靠外的两条直线；
(c)椭圆特征的提取，例如：在获得高斯滤波后的图像后，通过Canny边缘检测、查找轮廓和椭圆拟合等方式找到图像中的椭圆特征，并根据椭圆面积和椭圆周长等方式筛选出所需的椭圆，并输出椭圆的中心坐标；
(d)物体识别，例如：通过识别二维码的方式识别出测量区域内的零件A，零件B和零件C，并根据测量需要自适应测量零件A与零件B的参数，或者测量零件B与零件C的参数；
(e)参数的测量和计算，根据测量的需要计算所需的测量参数，例如：测量零件之间的缝隙，可通过直线检测筛选出来的两条直线计算一条直线两端点的中点到另一条直线的距离，并计算两条直线之间的角度或求出两直线的平行度；再例如：测量有孔零件装配时的位姿，可根据筛选出的椭圆中心点坐标以及零件本身的三维模型信息求出其他关键点坐标并进行位姿的计算；
(f)将测量得到的像素参数信息转化为实际距离参数信息输出到服务器；
(6)服务器测量参数的输出：服务器接收图像处理系统输出的测量参数信息，并与数据库中工艺信息进行比较，将比较结果分发到可视系统，用于可视化和其他处理。
本发明涉及的其它未说明部分与现有技术相同。

Claims (10)

  1. 一种基于计算机视觉的装配过程几何参数自适应测量系统，其特征是它包括有工艺设计客户端单元、图像采集系统、图像处理系统、追踪系统、可视化系统和服务器，所述工艺设计客户端单元通过网络将工艺信息输入服务器，所述服务器通过网络将测量指令输入图像采集系统，所述图像采集系统通过网络将需测量部分的图像输入图像处理系统，所述图像处理系统将处理后的测量参数信息输入服务器，所述服务器通过网络将测量后的数据输入可视化系统，所述追踪系统通过网络将测量指令输入图像采集系统，所述图像采集系统通过网络将需测量部分的图像输入图像处理系统，所述图像处理系统将处理后的测量参数信息输入服务器，所述服务器通过网络将测量后的数据输入可视化系统。
  2. 根据权利要求1所述的基于计算机视觉的装配过程几何参数自适应测量系统，其特征是所述图像采集系统分为软件系统和硬件系统，所述软件系统包括全局图像采集模块、测量指令模块和局部图像采集模块，所述全局图像采集模块通过线路与测量指令模块相连，所述测量指令模块通过线路与局部图像采集模块相连，所述硬件系统包括广角相机、长焦相机和转动云台，所述长焦相机安装在转动云台上。
  3. 根据权利要求1所述的基于计算机视觉的装配过程几何参数自适应测量系统，其特征是所述图像处理系统包括图像增强噪声处理模块、图像分割特征提取模块、物体识别参数估计模块和测量参数计算模块，所述图像增强噪声处理模块通过数据管道与图像分割特征提取模块相连，所述图像分割特征提取模块通过数据管道与物体识别参数估计模块相连，所述物体识别参数估计模块通过数据管道与测量参数计算模块相连。
  4. 根据权利要求2所述的基于计算机视觉的装配过程几何参数自适应测量系统，其特征是所述局部图像采集模块通过数据管道与图像增强噪声处理模块相连。
  5. 根据权利要求1所述的基于计算机视觉的装配过程几何参数自适应测量系统，其特征是所述追踪系统包括眼动追踪模块和手势追踪模块。
  6. 一种基于计算机视觉的装配过程几何参数自适应测量方法，其特征是，步骤如下：
    (1)相机安装和调试：广角相机和长焦相机的安装以及双目相机的标定，用于确定图像像素和实际长度之间的关系，固定广角相机的高度，使广角相机正对下方装配场景，能清晰全面地观测到整个装配过程，在转动云台上固定长焦相机，调整转动云台的位置，使长焦相机正对装配过程关键处需要测量的位置，并能清晰观测到测量区域；
    (2)服务器测量指令的输入：服务器将测量位置指令输入到图像采集系统，用于确定相机需要测量的位置；
    (3)测量区域的确定：图像采集系统接收需要测量的数据后，相机自适应地调整测量区域，用于确定测量位置的区域和大小；
    (4)追踪系统测量指令的输入：追踪系统中将人的交互指定的测量位置指令输入到图像采集系统，用于确定相机需要测量的位置；
    (5)图像处理与参数的计算：图像处理系统通过图像增强噪声处理、图像分割特征提取及物体识别参数估计完成测量区域的参数的计算；
    (6)服务器测量参数的输出：服务器接收图像处理系统输出的测量参数信息，并与数据库中工艺信息进行比较，将比较结果分发到可视系统，用于可视化和其他处理。
  7. 根据权利要求1所述的基于计算机视觉的装配过程几何参数自适应测量方法，其特征是，步骤(2)的具体步骤为：(a)工艺设计客户端单元将装配场景和需测量的参数要求数据发送到服务器；(b)服务器整合装配资源数据包将需测量的参数要求数据发送到测量客户端的图像采集系统；(c)图像采集系统根据需测量的数据确定需要测量的关键参数及测量位置。
  8. 根据权利要求1所述的基于计算机视觉的装配过程几何参数自适应测量方法，其特征是，步骤(3)的具体步骤为：(a)图像采集系统接收到需要测量的数据后，确定了关键参数的测量及测量位置，接着将长焦相机固定，使长焦相机正对测量位置并能清晰观测到测量区域；(b)在广角相机的视野范围内，在零件表面添加二维码，通过识别二维码标志点的方式识别 并定位追踪零件，当检测到零件接近测量位置一定邻域内时，相机自适应切换，打开正对测量位置的长焦相机进行测量检测，并通过框选的方式进一步减小测量区域并精确测量。
  9. 根据权利要求1所述的基于计算机视觉的装配过程几何参数自适应测量方法，其特征是，步骤(4)的具体步骤为：(a)眼动追踪确定测量位置和区域；(b)手势追踪确定测量位置和区域。
  10. 根据权利要求1所述的基于计算机视觉的装配过程几何参数自适应测量方法，其特征是，步骤(5)的具体步骤为：(a)在得到测量区域内的图像后，通过图像灰度化处理或高斯滤波处理减小图像的噪声；(b)直线特征的提取；(c)椭圆特征的提取；(d)物体识别；(e)参数的测量和计算；(f)将测量得到的像素参数信息转化为实际距离参数信息输出到服务器。
PCT/CN2020/085248 2019-10-29 2020-04-17 一种基于计算机视觉的装配过程几何参数自适应测量系统及方法 WO2021179400A1 (zh)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CN201911040106 2019-10-29
CN202010166853.8 2020-03-11
CN202010166853.8A CN111475016A (zh) 2019-10-29 2020-03-11 一种基于计算机视觉的装配过程几何参数自适应测量系统及方法

Publications (1)

Publication Number Publication Date
WO2021179400A1 true WO2021179400A1 (zh) 2021-09-16

Family

ID=71747330

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/085248 WO2021179400A1 (zh) 2019-10-29 2020-04-17 一种基于计算机视觉的装配过程几何参数自适应测量系统及方法

Country Status (2)

Country Link
CN (1) CN111475016A (zh)
WO (1) WO2021179400A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115096905A (zh) * 2022-06-23 2022-09-23 国网(西安)环保技术中心有限公司 一种均压电极结垢程度原位光学检测装置及方法
CN115661726A (zh) * 2022-12-26 2023-01-31 江苏中车数字科技有限公司 一种轨道列车工件装配的自主视频采集与分析方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102927937A (zh) * 2012-10-10 2013-02-13 东莞新吉凯氏测量技术有限公司 一种基于云技术的测量系统
JP2013161281A (ja) * 2012-02-06 2013-08-19 Sumitomo Electric Ind Ltd 交通画像取得装置、交通画像取得方法および交通画像取得プログラム
KR20150081732A (ko) * 2014-01-06 2015-07-15 한화테크윈 주식회사 이동-물체 추적 시스템
CN108200339A (zh) * 2017-12-29 2018-06-22 北京臻迪科技股份有限公司 目标追踪定位方法、系统及摄像装置
CN108229615A (zh) * 2017-12-30 2018-06-29 清华大学 一种航空发动机零部件的数字化装配支持系统及方法
CN110362193A (zh) * 2018-04-10 2019-10-22 脸谱科技有限责任公司 用手或眼睛跟踪辅助的目标跟踪方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013161281A (ja) * 2012-02-06 2013-08-19 Sumitomo Electric Ind Ltd 交通画像取得装置、交通画像取得方法および交通画像取得プログラム
CN102927937A (zh) * 2012-10-10 2013-02-13 东莞新吉凯氏测量技术有限公司 一种基于云技术的测量系统
KR20150081732A (ko) * 2014-01-06 2015-07-15 한화테크윈 주식회사 이동-물체 추적 시스템
CN108200339A (zh) * 2017-12-29 2018-06-22 北京臻迪科技股份有限公司 目标追踪定位方法、系统及摄像装置
CN108229615A (zh) * 2017-12-30 2018-06-29 清华大学 一种航空发动机零部件的数字化装配支持系统及方法
CN110362193A (zh) * 2018-04-10 2019-10-22 脸谱科技有限责任公司 用手或眼睛跟踪辅助的目标跟踪方法及系统

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115096905A (zh) * 2022-06-23 2022-09-23 国网(西安)环保技术中心有限公司 一种均压电极结垢程度原位光学检测装置及方法
CN115661726A (zh) * 2022-12-26 2023-01-31 江苏中车数字科技有限公司 一种轨道列车工件装配的自主视频采集与分析方法

Also Published As

Publication number Publication date
CN111475016A (zh) 2020-07-31

Similar Documents

Publication Publication Date Title
Shan et al. A stereovision-based crack width detection approach for concrete surface assessment
CN110501342B (zh) 一种筒子纱纱杆定位视觉检测方法
Peng et al. Study of tool wear monitoring using machine vision
CN109489591B (zh) 基于机器视觉的平面划痕长度非接触测量方法
JP6899189B2 (ja) ビジョンシステムで画像内のプローブを効率的に採点するためのシステム及び方法
Molleda et al. A profile measurement system for rail quality assessment during manufacturing
Zhang et al. Research on tool wear detection based on machine vision in end milling process
WO2021179400A1 (zh) 一种基于计算机视觉的装配过程几何参数自适应测量系统及方法
CN107817044B (zh) 基于机器视觉的板材振动的测量装置及方法
He et al. A critical review for machining positioning based on computer vision
CN107388991B (zh) 一种端面多圆角轴类零件圆角半径测量方法
CN110986834A (zh) 一种自动化装配穿管监测方法
CN105184792B (zh) 一种圆锯片磨损量在线测量方法
CN104316033A (zh) 汽车零部件视觉检测系统
CN104677782A (zh) 一种电连接器壳体机器视觉在线检测系统及方法
TW201723422A (zh) 光滑物體的量測系統及其量測方法
Su et al. Measuring wear of the grinding wheel using machine vision
CN116393982B (zh) 一种基于机器视觉的螺丝锁付方法及装置
CN206410679U (zh) 一种非接触式柱状物垂直度测量装置
CN104748684A (zh) 一种曲轴轴肩清根的视觉检测方法及装置
Molleda et al. A profile measurement system for rail manufacturing using multiple laser range finders
CN116091488B (zh) 一种发动机摇摆试验的位移测试方法及位移测试系统
Zhou et al. Multipose measurement of surface defects on rotary metal parts with a combined laser-and-camera sensor
CN109632815A (zh) 表面缺陷在线检测系统及方法
Jie et al. Method for measuring rotation angle of faucet handle based on machine vision

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20924559

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20924559

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20924559

Country of ref document: EP

Kind code of ref document: A1