WO2022041128A1 - 用于多个相机的自动标定方法及系统 - Google Patents

用于多个相机的自动标定方法及系统 Download PDF

Info

Publication number
WO2022041128A1
WO2022041128A1 PCT/CN2020/112155 CN2020112155W WO2022041128A1 WO 2022041128 A1 WO2022041128 A1 WO 2022041128A1 CN 2020112155 W CN2020112155 W CN 2020112155W WO 2022041128 A1 WO2022041128 A1 WO 2022041128A1
Authority
WO
WIPO (PCT)
Prior art keywords
calibration
cameras
multiple cameras
screen
map
Prior art date
Application number
PCT/CN2020/112155
Other languages
English (en)
French (fr)
Inventor
黄康
陈可
韩亚宁
蔚鹏飞
王立平
Original Assignee
中国科学院深圳先进技术研究院
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 中国科学院深圳先进技术研究院 filed Critical 中国科学院深圳先进技术研究院
Priority to PCT/CN2020/112155 priority Critical patent/WO2022041128A1/zh
Publication of WO2022041128A1 publication Critical patent/WO2022041128A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/80Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration

Definitions

  • the present invention relates to the technical field of machine vision, and in particular, to an automatic calibration method and system for multiple cameras.
  • an automatic calibration system consisting of two freely movable parallel plates.
  • the hardware consists of two upper and lower parallel glass plates with adjustable heights.
  • the upper glass plate can move freely.
  • Checkerboard calibration map; the lower glass plate is fixed, and the camera is placed on the lower glass plate to shoot the movable upper glass plate calibration map to complete the calibration.
  • a high-precision automatic calibration board and a calibration method are introduced in the prior art.
  • the first surface is provided with a black and white checkerboard pattern, and the white checkerboard is drawn with simple graphics or blanks inside to represent binary information or for feature recognition. , the black and white checkerboards are arranged alternately for coordinate calibration; a high-precision automatic calibration method is provided, which disperses the QR code information into the white squares of the checkerboard, and retains the high-density coordinates of the checkerboard. Independent establishment of world coordinates provides a denser checkerboard array and improves calibration accuracy.
  • the calibration usually requires at least one calibration image, each of which needs to move and rotate the calibration plate, which takes a long time, and Calibration personnel are required to participate; if one of the multiple cameras shakes, multiple cameras need to be re-calibrated; the calibration map cannot be automatically generated and automatically rotated and translated on the calibration screen; after the calibration map is transformed for multiple cameras Can not automatically shoot and can not automatically save the pictures after shooting; using the calibration board or the checkerboard printed on paper, the accuracy of the calibration board checkerboard will be damaged.
  • the present invention provides an automatic calibration method and system for multiple cameras.
  • the specific technical solutions are as follows:
  • An automatic calibration method for multiple cameras comprising the following steps:
  • S2 display the calibration map on the calibration screen, and use a plurality of the cameras to shoot the calibration map;
  • step S5 Detect whether the number of pictures reaches the preset threshold, if the detection result is no, return to step S4, and if the determination result is yes, execute step S6;
  • the step S3 of "making the calibration image on the calibration screen generate a rotation and/or translation effect, and using a plurality of the cameras to shoot the calibration image again” includes:
  • the calibration map is photographed by a plurality of the cameras to obtain a plurality of pictures
  • the pictures are stored, and a plurality of the cameras are calibrated by using the stored pictures.
  • the step S3 of "making the calibration image on the calibration screen generate a rotation and/or translation effect, and using a plurality of the cameras to shoot the calibration image again” includes:
  • the size of the calibration graph on the calibration screen remains unchanged.
  • the stored calibration map can be automatically generated on the calibration screen, and the calibration map can calibrate a plurality of the cameras through rotation and/or translation.
  • the positions of a plurality of the cameras and the calibration map are kept fixed.
  • a plurality of the cameras are arranged at preset positions, so that the plurality of cameras can take pictures of the entire display area of the calibration screen.
  • the shape of the calibration map is a rectangle, and the calibration map includes a plurality of checkerboards.
  • the shape of the checkerboard is square; the colors of two adjacent checkerboards in the calibration diagram are different.
  • an automatic calibration system for a plurality of cameras includes a camera fixing frame, a plurality of cameras and a calibration screen
  • the camera fixing frame includes at least one support structure
  • the plurality of cameras are fixed on On the support structure, the plurality of cameras face the calibration screen for photographing the calibration map on the calibration screen
  • the control module includes:
  • the initialization unit is used to initialize the multiple cameras
  • a shooting control unit used for displaying a calibration map on a calibration screen, using a plurality of the cameras to shoot the calibration map, and by displaying a control instruction, the calibration map on the calibration screen has a rotation and/or translation effect, Using a plurality of the cameras to shoot the calibration map again, and in this cycle, to obtain a larger number of pictures, until the number of detected pictures reaches a preset threshold;
  • a calibration unit configured to calibrate a plurality of the cameras according to the obtained pictures.
  • the calibration screen includes a flat panel display.
  • the calibration screen is connected to a preset computer host, so as to display different calibration diagrams in sequence under the control of the computer host.
  • the calibration map can be automatically generated and rotated and/or translated on the calibration screen.
  • the calibration personnel can control the start of the calibration in front of the computer and observe the entire calibration process without the need for the calibration personnel to participate in the rotation and/or translation of the calibration screen.
  • Multiple cameras The calibration map can be automatically and synchronously shot after the calibration map is transformed, and the captured image can be automatically saved.
  • the program can complete the shooting of the calibration map of multiple cameras at one time. The position of the camera is fixed on the support structure, and there will be no shaking. As a result, the calibration of multiple cameras is affected.
  • the use of a calibration screen with a completely flat surface can ensure that the checkerboard has no distortion and high contrast.
  • Embodiment 1 is a schematic diagram of an automatic calibration method for multiple cameras in Embodiment 1;
  • Embodiment 2 is a schematic diagram of an automatic calibration system of a plurality of cameras in Embodiment 1;
  • FIG. 3 is a schematic diagram of the automatic calibration system for multiple cameras in Embodiment 2 in another state.
  • this embodiment provides an automatic calibration method and system for multiple cameras, and the specific technical solutions are as follows:
  • An automatic calibration method for multiple cameras comprising the following steps:
  • S2 display the calibration diagram 3 on the calibration screen 2, and use a plurality of cameras 4 to photograph the calibration diagram 3;
  • step S5 Detect whether the number of pictures reaches the preset threshold, if the detection result is no, return to step S4, and if the determination result is yes, execute step S6;
  • S6 Calibrate the plurality of cameras 4 through the obtained pictures.
  • the entire calibration process can be completed within 30 seconds. Compared with manually moving the calibration plate, it takes at least 10 minutes, and this embodiment can shorten the calibration time; compared with moving the calibration plate by a mechanical arm or a robot or mechanical structure, this embodiment is more stable and should not be shaken.
  • the plurality of cameras 4 are kept fixed to the positions of the calibration FIG. 3 .
  • the calibration screen 2 is connected to a preset computer host, so as to display different calibration diagrams 3 in sequence under the control of the computer host.
  • the multiple cameras 4 Preferably, to initialize the multiple cameras 4, use Python combined with OpenCV to write a script to control the multiple cameras 4 to open, and the multiple cameras 4 are automatically initialized and configured internally. The script needs to wait for a few seconds until the internal initialization of the multiple cameras 4 is completed. .
  • Multiple cameras 4 are set to 640*480 resolution, and the frame rate is 30FPS.
  • the shape of the calibration diagram 3 is a rectangle, and the calibration diagram 3 includes a plurality of checkerboards.
  • the shape of the checkerboard is square, and the colors of the two adjacent checkerboards in Figure 3 are different.
  • S3 to make the calibration map 3 on the calibration screen 2 produce a rotation and/or translation effect, and use a plurality of cameras 4 to shoot the calibration map 3 again” includes:
  • the calibration diagram 3 is photographed by a plurality of cameras 4 to obtain a plurality of pictures;
  • the pictures are stored, and the plurality of cameras 4 are calibrated through the stored pictures.
  • the checkerboard is photographed with multiple cameras 4 at the same time and stored.
  • the stored calibration map 3 can be automatically generated on the calibration screen 2, and the calibration map 3 can be used to calibrate multiple cameras 4 through rotation and/or translation.
  • the camera synchronous shooting module is written in Python combined with OpenCV, but if control software or language is used, it also belongs to the protection scope of the present invention.
  • this embodiment provides an automatic calibration system for multiple cameras 4, including a camera fixing frame, multiple cameras 4 and a calibration screen 2,
  • the camera fixing frame includes at least one supporting structure 1, a plurality of The camera 4 is fixed on the support structure 1, and a plurality of cameras 4 face the calibration screen 2 for shooting the calibration diagram 3 on the calibration screen, and also includes a control module, and the control module includes:
  • an initialization unit used for initializing the multiple cameras 4;
  • the shooting control unit is used to display the calibration map 3 on the calibration screen 2, use a plurality of cameras 4 to shoot the calibration map 3, and by displaying the control instructions, make the calibration map 3 on the calibration screen 2 produce a rotation and/or translation effect, using multiple cameras.
  • Each camera 4 shoots the calibration image again, and this cycle is used to obtain a larger number of images, until the number of detected images reaches a preset threshold;
  • the calibration unit is used for calibrating the plurality of cameras 4 through the obtained pictures.
  • the multiple cameras 4 are set at preset positions, so that the multiple cameras 4 can capture the entire display area of the calibration screen 2 .
  • the support structure 1 of the camera fixing frame is 1m*1m*1m in size and made of aluminum alloy, which is used for fixing multiple cameras 4, and has a stable structure and is convenient for camera installation.
  • Calibration screen 2 is an advertising machine with a flat back, RGB color screen, size 32 inches, length 0.73m, width 0.43m, height 0.05m, can directly display the computer screen, with HDMI interface, the checkerboard is directly displayed on the calibration screen 2 .
  • the calibration screen 2 includes a flat panel display, and the calibration screen 2 and its installation and fixation use a flat-panel advertising machine, but if other types of flat-panel display devices are used, it also falls within the protection scope of the present invention.
  • the calibration screen with a completely flat surface is used with high precision, which can ensure that the checkerboard is completely free of distortion and has high contrast.
  • the installation and fixation of the calibration screen 2 is to place the calibration screen 2 on the ground to ensure that the calibration diagram 3 displayed on the calibration screen 2 can be completely photographed by multiple cameras 4 at the same time.
  • Multiple cameras 4 frame rate supports 60FPS, resolution supports 1920*1080, and the interface is USB3.0.
  • the plurality of cameras 4 are inclined downward, toward the calibration screen 2 .
  • the multiple cameras 4 and their installation and fixation specifically adopt a four-camera solution, and the four cameras shoot synchronously, which is more efficient, and the calibration map shooting of the four cameras can be completed once the program is run.
  • the four cameras shoot synchronously, which is more efficient, and the calibration map shooting of the four cameras can be completed once the program is run.
  • other numbers of cameras it also falls within the protection scope of the present invention.
  • this embodiment provides an automatic calibration method and system for multiple cameras. Compared with Embodiment 1, the main differences of this embodiment are:
  • S3 "makes the calibration map 3 on the calibration screen 2 produce a rotation and/or translation effect, and uses multiple cameras 4 to shoot the calibration map 3 again” includes:
  • the size of the calibration map 3 on the calibration screen 2 remains unchanged.
  • modules in the device in the implementation scenario may be distributed in the device in the implementation scenario according to the description of the implementation scenario, or may be located in one or more devices different from the implementation scenario with corresponding changes.
  • the modules of the above implementation scenarios may be combined into one module, or may be further split into multiple sub-modules.

Abstract

本发明提供了一种用于多个相机的自动标定方法及系统,自动标定方法包括多个相机进行初始化设置;在标定屏上生成标定图,使用多个相机拍摄标定图;标定屏上的标定图进行旋转和/或平移,使用多个相机拍摄标定图;判断拍摄图片的数量是否足够;通过获得的拍摄图片来对多个相机进行标定。用于多个相机的自动标定系统包括相机固定框、多个相机和标定屏,相机固定框架包括至少一个支撑结构,多个相机朝向标定屏用于拍摄标定屏上的标定图。标定图能够自动生成并在标定屏上自动进行旋转、平移,标定人员能够在计算机前控制标定的开始,观察整个标定过程,多个相机能够在标定图变换后自动拍摄,并将拍摄后的图片自动保存。

Description

用于多个相机的自动标定方法及系统 技术领域
本发明涉及机器视觉技术领域,具体而言,涉及用于多个相机的自动标定方法及系统。
背景技术
在现有技术中介绍了一种两块可自由移动的平行板组成的自动标定系统,硬件上有两块可调节高度的上下平行的玻璃板组成,上层玻璃板可自由移动,中间位置有一张棋盘格标定图;下层玻璃板固定不动,相机放置在下层玻璃板上拍摄可移动的上层玻璃板标定图,以此完成标定。
在现有技术中介绍了一种高精度自动化标定板和标定方法,第一面设置有黑白相间的棋盘格图案,白色棋盘格通过于内部绘制简单的图形或者空白表示二进制信息或用于特征识别,黑白棋盘格相间排列用于坐标标定;提供了一种高精度自动化的标定方法,将二维码信息分散到棋盘格的白色方格中,保留棋盘格高密度的坐标,独立子棋盘格可以独立建立世界坐标,提供了更为密集的棋盘格阵列,提升标定精度。
在现有技术中采用打印或工厂制作的标定板,由于制作工艺的问题,棋盘格的大小不同或标定板表面弯曲导致标定精度受损;采用纸质打印的棋盘格或PET材质的标定板,表面本身会有弯曲,且对比度太低、表面反光,导致标定板棋盘格精度受损。
上述标定方法虽各有特点,但是没有专门应用于多个相机的自动化快速标定的系统,标定通常需拍摄至少一张标定图,每张都需要移动、旋转标定板,需要花费很长时间,且需要标定人员参与其中;如果多个相机中的其中一个相机发生晃动,则多个相机需重新标定;标定图不能自动生成 并在标定屏上自动进行旋转、平移;多个相机在标定图变换后不能自动拍摄且不能将拍摄后的图片自动保存;使用标定板或纸质打印的棋盘格,标定板棋盘格的精度受损。
发明内容
为了克服现有技术的不足,本发明提供了一种用于多个相机的自动标定方法及系统,具体技术方案如下所示:
一种用于多个相机的自动标定方法,包括以下步骤:
S1:对多个相机进行初始化设置;
S2:在标定屏上显示标定图,使用多个所述相机拍摄所述标定图;
S3:通过显示控制指令,使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图;
S4:重复步骤S3,以得到更多数量的图片;
S5:检测图片的数量是否达到预设阀值,若检测结果为否,返回步骤S4,若判断结果为是,则执行步骤S6;
S6:通过获得的所述图片来对多个所述相机进行标定。
在一个具体的实施例中,S3所述“使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图”包括:
通过多个所述相机对所述标定图进行拍摄,得到多张图片;
将所述图片进行存储,并通过存储的所述图片对多个所述相机进行标定。
在一个具体的实施例中,S3所述“使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图”包括:
所述标定图进行旋转和/或平移前,和进行旋转和/或平移后,所述标定图在所述标定屏的尺寸保持不变。
在一个具体的实施例中,所述存储后的标定图能在所述标定屏上自动生成,所述标定图能通过旋转和/或平移对多个所述相机进行标定。
在一个具体的实施例中,多个所述相机与所述标定图的位置保持固定。
在一个具体的实施例中,多个所述相机设置在预设位置,以使得多个所述相机均可拍摄到所述标定屏的全部显示区域。
在一个具体的实施例中,所述标定图的形状为长方形,所述标定图包括多个棋盘格。
在一个具体的实施例中,所述棋盘格的形状为正方形;所述标定图中相邻的两所述棋盘格的颜色不一样。
在一个具体的实施例中,一种用于多个相机的自动标定系统,包括相机固定框、多个相机和标定屏,所述相机固定框架包括至少一个支撑结构,所述多个相机固定在所述支撑结构上,所述多个相机朝向所述标定屏用于拍摄所述标定屏上的标定图,还包括控制模块,所述控制模块包括:
初始化单元,用于对多个相机进行初始化设置;
拍摄控制单元,用于在标定屏上显示标定图,使用多个所述相机拍摄所述标定图,通过显示控制指令,使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图,以此循环,以得到更多数量的图片,直至检测图片的数量达到预设阀值;
标定单元,用于通过获得的图片来对多个所述相机进行标定。
在一个具体的实施例中,所述标定屏包括平板显示器。
在一个具体的实施例中,所述标定屏连接预设的电脑主机,以在所述电脑主机的控制下依次显示不同的所述标定图。
本发明至少具有以下有益效果:
标定图能够自动生成并在标定屏上进行旋转和/或平移,标定人员能够在计算机前控制标定的开始,观察整个标定过程,不需要标定人员参与标定屏的旋转和/或平移,多个相机能够在标定图变换后自动、同步地拍摄标定图,并将拍摄后的图片自动保存,程序运行一次即可完成多个相机的标定图拍摄,相机的位置固定在支撑结构上,不会产生晃动从而影响多个相机的标定,采用表面完全平整的标定屏,可保证棋盘格无畸变,对比度高。
为使本发明的上述目的、特征和优点能更明显易懂,下文特举较佳实施例,并配合所附附图,作详细说明如下。
附图说明
为了更清楚地说明本发明的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,应当理解,以下附图仅示出了本发明的某些实施例,因此不应被看作是对范围的限定,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他相关的附图。
图1是实施例1中多个相机的自动标定方法示意图;
图2是实施例1中多个相机的自动标定系统示意图;
图3是实施例2中多个相机的自动标定系统另一状态下的示意图。
主要元件符号说明:
1-支撑结构;2-标定屏;3-标定图;4-相机。
具体实施方式
实施例1
如图1-2所示,本实施例提供了一种用于多个相机的自动标定方法及系统,具体技术方案如下所示:
一种用于多个相机的自动标定方法,包括以下步骤:
S1:对多个相机4进行初始化设置;
S2:在标定屏2上显示标定图3,使用多个相机4拍摄所述标定图3;
S3:通过显示控制指令,使标定屏2上的标定图3产生旋转和/或平移效果,使用多个相机4再次拍摄标定图3;
S4:重复步骤S3,以得到更多数量的图片;
S5:检测图片的数量是否达到预设阀值,若检测结果为否,返回步骤S4,若判断结果为是,则执行步骤S6;
S6:通过获得的图片来对多个相机4进行标定。
首先将多个相机4和标定屏2位置固定,标定屏2连接预设的电脑主机,进而控制标定屏2生成一张标定图3,再控制多个相机4同时拍摄此标定图3,拍摄完成之后将标定屏2上的标定图3进行旋转和/或平移,控制多个相机4拍摄下一组标定图3,如此往复直至拍摄完成。
在30秒内即可完成整个标定流程。相对于手动移动标定板,需要至少10分钟,本实施例可缩短标定时间;相对于机械臂或机器人或机械结构方式移动标定板,本实施例更加稳定,不宜晃动。
优选的,多个相机4与标定图3的位置保持固定。
优选的,标定屏2连接预设的电脑主机,以在电脑主机的控制下依次显示不同的标定图3。
优选的,对多个相机4初始化,用Python结合OpenCV编写脚本,控制多个相机4打开,多个相机4内部自动进行初始化配置,脚本需要等待几秒,待多个相机4内部初始化完成即可。多个相机4设置为640*480分辨率,帧率为30FPS。
优选的,标定图3的形状为长方形,标定图3包括多个棋盘格。棋盘格的形状为正方形,标定图3中相邻的两棋盘格的颜色不一样。
用Python结合OpenCV编写脚本标定屏2棋盘格显示,控制在标定屏2上显示合适大小的标定图3。在多个相机4拍摄之后,进行棋盘格的旋转和 /或平移,棋盘格格点数和具体尺寸由实际大小决定,在此使用格点数为7*4、格子大小为28mm*28mm的棋盘格。
优选的,S3“使标定屏2上的标定图3产生旋转和/或平移效果,使用多个相机4再次拍摄标定图3”包括:
通过多个相机4对标定图3进行拍摄,得到多张图片;
将图片进行存储,并通过存储的图片对多个相机4进行标定。用Python结合OpenCV编写脚本,控制多个相机4同时拍摄标定屏2上的棋盘格。在标定屏2棋盘格每次变化之后,用多个相机4同时拍摄棋盘格并存储。存储后的标定图3能在标定屏2上自动生成,标定图3能通过旋转和/或平移对多个相机4进行标定。
优选的,相机同步拍摄模块采用Python结合OpenCV的编写形式,但如果采用控制软件或语言,也属于本发明的保护范围。
如图2所示,本实施例提供了一种用于多个相机4的自动标定系统,包括相机固定框、多个相机4和标定屏2,相机固定框架包括至少一个支撑结构1,多个相机4固定在支撑结构1上,多个相机4朝向标定屏2用于拍摄标定屏上的标定图3,还包括控制模块,控制模块包括:
初始化单元,用于对多个相机4进行初始化设置;
拍摄控制单元,用于在标定屏2上显示标定图3,使用多个相机4拍摄标定图3,通过显示控制指令,使标定屏2上的标定图3产生旋转和/或平移效果,使用多个相机4再次拍摄标定图,以此循环,以得到更多数量的图片,直至检测图片的数量达到预设阀值;
标定单元,用于通过获得的图片来对多个相机4进行标定。
多个相机4设置在预设位置,以使得多个相机4均可拍摄到标定屏2的全部显示区域。
其中,相机固定框的支撑结构1大小为1m*1m*1m,材质为铝合金,用于固定多个相机4,结构稳定,便于摄像头安装。
标定屏2是一个背部平整的广告机,RGB彩色屏,大小32寸,长0.73m,宽0.43m,高0.05m,可以直接显示电脑屏幕,具有HDMI接口,棋盘格在标定屏2上直接显示。
优选的,标定屏2包括平板显示器,标定屏2及其安装固定具体采用平板广告机,但如果采用其他型号的类平板显示设备,也属于本发明的保护范围。
采用表面完全平整的标定屏,精度高,可保证棋盘格完全无畸变,对比度高。
标定屏2安装及固定是将标定屏2放置在地面位置,保证标定屏2上放映的标定图3可以被多个相机4同时完整拍摄。
多个相机4帧率支持60FPS,分辨率支持1920*1080,接口为USB3.0。多个相机4斜向下方,朝向标定屏2。
优选的,多个相机4及其安装固定具体采用4个相机方案,四个相机同步拍摄,效率更高,程序运行一次即可完成四个相机的标定图拍摄。但如果采用其他相机数量,也属于本发明的保护范围。
实施例2
如图3所示,本实施例提供了一种用于多个相机的自动标定方法及系统,与实施例1相比,本实施例的主要区别在于:
特别的,S3“使标定屏2上的标定图3产生旋转和/或平移效果,使用多个相机4再次拍摄标定图3”包括:
标定图3进行旋转和/或平移前,和进行旋转和/或平移后,标定图3在标定屏2的尺寸保持不变。
本实施例其它特征与实施例1相同,不再赘述。
本领域技术人员可以理解附图只是一个优选实施场景的示意图,附图中的模块或流程并不一定是实施本发明所必须的。
本领域技术人员可以理解实施场景中的装置中的模块可以按照实施场景描述进行分布于实施场景的装置中,也可以进行相应变化位于不同于本实施场景的一个或多个装置中。上述实施场景的模块可以合并为一个模块,也可以进一步拆分成多个子模块。
上述本发明序号仅仅为了描述,不代表实施场景的优劣。
以上公开的仅为本发明的几个具体实施场景,但是,本发明并非局限于此,任何本领域的技术人员能思之的变化都应落入本发明的保护范围。

Claims (11)

  1. 一种用于多个相机的自动标定方法,其特征在于,包括以下步骤:
    S1:对多个相机进行初始化设置;
    S2:在标定屏上显示标定图,使用多个所述相机拍摄所述标定图;
    S3:通过显示控制指令,使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图;
    S4:重复步骤S3,以得到更多数量的图片;
    S5:检测图片的数量是否达到预设阀值,若检测结果为否,返回步骤S4,若判断结果为是,则执行步骤S6;
    S6:通过获得的所述图片来对多个所述相机进行标定。
  2. 按照权利要求1所述的一种用于多个相机的自动标定方法,其特征在于,S3所述“使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图”包括:
    通过多个所述相机对所述标定图进行拍摄,得到多张图片;
    将所述图片进行存储,并通过存储的所述图片对多个所述相机进行标定。
  3. 按照权利要求1所述的一种用于多个相机的自动标定方法,其特征在于,S3所述“使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图”包括:
    所述标定图进行旋转和/或平移前,和进行旋转和/或平移后,所述标定图在所述标定屏的尺寸保持不变。
  4. 按照权利要求2所述的一种用于多个相机的自动标定方法,其特征在于,所述存储后的标定图能在所述标定屏上自动生成,所述标定图能通 过旋转和/或平移对多个所述相机进行标定。
  5. 按照权利要求1所述的一种用于多个相机的自动标定方法,其特征在于,多个所述相机与所述标定图的位置保持固定。
  6. 按照权利要求1所述的一种用于多个相机的自动标定方法,其特征在于,多个所述相机设置在预设位置,以使得多个所述相机均可拍摄到所述标定屏的全部显示区域。
  7. 按照权利要求1所述的一种用于多个相机的自动标定方法,其特征在于,所述标定图的形状为长方形,所述标定图包括多个棋盘格。
  8. 按照权利要求1所述的一种用于多个相机的自动标定方法,其特征在于,所述棋盘格的形状为正方形;所述标定图中相邻的两所述棋盘格的颜色不一样。
  9. 一种用于多个相机的自动标定系统,其特征在于,包括相机固定框、多个相机和标定屏,所述相机固定框架包括至少一个支撑结构,所述多个相机固定在所述支撑结构上,所述多个相机朝向所述标定屏用于拍摄所述标定屏上的标定图,还包括控制模块,所述控制模块包括:
    初始化单元,用于对多个相机进行初始化设置;
    拍摄控制单元,用于在标定屏上显示标定图,使用多个所述相机拍摄所述标定图,通过显示控制指令,使所述标定屏上的所述标定图产生旋转和/或平移效果,使用多个所述相机再次拍摄所述标定图,以此循环,以得到更多数量的图片,直至检测图片的数量达到预设阀值;
    标定单元,用于通过获得的图片来对多个所述相机进行标定。
  10. 按照权利要求9所述的用于多个相机的自动标定系统,其特征在于,所述标定屏包括平板显示器。
  11. 按照权利要求9所述的用于多个相机的自动标定系统,其特征在于,所述标定屏连接预设的电脑主机,以在所述电脑主机的控制下依次显示不同的所述标定图。
PCT/CN2020/112155 2020-08-28 2020-08-28 用于多个相机的自动标定方法及系统 WO2022041128A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/112155 WO2022041128A1 (zh) 2020-08-28 2020-08-28 用于多个相机的自动标定方法及系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/112155 WO2022041128A1 (zh) 2020-08-28 2020-08-28 用于多个相机的自动标定方法及系统

Publications (1)

Publication Number Publication Date
WO2022041128A1 true WO2022041128A1 (zh) 2022-03-03

Family

ID=80352502

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/112155 WO2022041128A1 (zh) 2020-08-28 2020-08-28 用于多个相机的自动标定方法及系统

Country Status (1)

Country Link
WO (1) WO2022041128A1 (zh)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103366356A (zh) * 2012-04-06 2013-10-23 北京三星通信技术研究有限公司 通用标定系统和方法及其集成成像重构系统和方法
CN104376560A (zh) * 2014-11-17 2015-02-25 国家电网公司 基于光学投影仪的相机多功能标定方法及装置
CN106352806A (zh) * 2016-08-10 2017-01-25 中国科学技术大学 一种立体视觉三维数字图像相关测量的高精度标定方法
CN106510626A (zh) * 2016-11-08 2017-03-22 重庆理工大学 基于双目立体视觉的人体浅层静脉三维重建装置及方法
US20170188022A1 (en) * 2015-12-28 2017-06-29 Automotive Research & Test Center Calibration method for merging object coordinates and calibration board device using the same
CN108171757A (zh) * 2017-12-28 2018-06-15 华勤通讯技术有限公司 摄像头标定系统及方法
CN108230404A (zh) * 2018-03-20 2018-06-29 珊口(上海)智能科技有限公司 标定系统、标定方法及所适用的设备
CN110232716A (zh) * 2019-05-31 2019-09-13 深圳市道通智能航空技术有限公司 一种相机标定方法、装置和电子设备
CN111402344A (zh) * 2020-04-23 2020-07-10 Oppo广东移动通信有限公司 标定方法、标定装置和非易失性计算机可读存储介质

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103366356A (zh) * 2012-04-06 2013-10-23 北京三星通信技术研究有限公司 通用标定系统和方法及其集成成像重构系统和方法
CN104376560A (zh) * 2014-11-17 2015-02-25 国家电网公司 基于光学投影仪的相机多功能标定方法及装置
US20170188022A1 (en) * 2015-12-28 2017-06-29 Automotive Research & Test Center Calibration method for merging object coordinates and calibration board device using the same
CN106352806A (zh) * 2016-08-10 2017-01-25 中国科学技术大学 一种立体视觉三维数字图像相关测量的高精度标定方法
CN106510626A (zh) * 2016-11-08 2017-03-22 重庆理工大学 基于双目立体视觉的人体浅层静脉三维重建装置及方法
CN108171757A (zh) * 2017-12-28 2018-06-15 华勤通讯技术有限公司 摄像头标定系统及方法
CN108230404A (zh) * 2018-03-20 2018-06-29 珊口(上海)智能科技有限公司 标定系统、标定方法及所适用的设备
CN110232716A (zh) * 2019-05-31 2019-09-13 深圳市道通智能航空技术有限公司 一种相机标定方法、装置和电子设备
CN111402344A (zh) * 2020-04-23 2020-07-10 Oppo广东移动通信有限公司 标定方法、标定装置和非易失性计算机可读存储介质

Similar Documents

Publication Publication Date Title
CN110763152B (zh) 一种水下主动旋转结构光三维视觉测量装置及测量方法
CN106887023A (zh) 用于双目摄像机标定的标定板及其标定方法和标定系统
CN106210679B (zh) 多影像输出装置校正系统及其校正方法
US8197070B2 (en) Color-based feature identification
US20040239885A1 (en) Super-resolution overlay in multi-projector displays
CN108830906B (zh) 一种基于虚拟双目视觉原理的摄像机参数自动标定方法
CN109215108A (zh) 基于激光扫描的全景三维重建系统及方法
KR102123488B1 (ko) 디스플레이 패널 검사 장치 및 디스플레이 패널 검사 방법
CN107025881B (zh) 显示单元的校正方法及装置
JP4411588B2 (ja) 欠陥検査装置
CN104574355B (zh) 立体摄像机的校准系统和立体摄像机的校准方法
CN108986721B (zh) 一种用于显示面板检测的检测图形生成方法
CN104836953B (zh) 多投影仪屏幕特征点自动拍照及去噪识别方法
WO2022041128A1 (zh) 用于多个相机的自动标定方法及系统
WO2020075768A1 (ja) 三次元走査装置、三次元モデルの生成方法、訓練用データ、及び機械学習モデル
JP3879560B2 (ja) 投写型画像表示装置
JP4230839B2 (ja) マルチカメラシステム及びその調整装置
CN109983531A (zh) 生成系统、生成方法和计算机程序
CN106441238A (zh) 一种基于红外视觉技术的机器人的定位装置及其定位导航算法
CN110784692A (zh) 投影装置、投影系统及影像校正方法
CN112991457A (zh) 手术导航中投影仪空间位置和内外参标定方法、装置
CN101980299B (zh) 基于棋盘标定的摄像机映射方法
CN115086631B (zh) 图像生成方法和信息处理装置
CN103630542A (zh) 缺陷检测装置、缺陷修正装置及缺陷检测方法
CN112102416A (zh) 用于多个相机的自动标定方法及系统

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: 20950798

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: 20950798

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20950798

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 14.08.2023)