WO2022052404A1 - 基于机器视觉内存对位接插方法与***、设备、存储介质 - Google Patents

基于机器视觉内存对位接插方法与***、设备、存储介质 Download PDF

Info

Publication number
WO2022052404A1
WO2022052404A1 PCT/CN2021/073258 CN2021073258W WO2022052404A1 WO 2022052404 A1 WO2022052404 A1 WO 2022052404A1 CN 2021073258 W CN2021073258 W CN 2021073258W WO 2022052404 A1 WO2022052404 A1 WO 2022052404A1
Authority
WO
WIPO (PCT)
Prior art keywords
memory
mechanical arm
reference position
move
mobile camera
Prior art date
Application number
PCT/CN2021/073258
Other languages
English (en)
French (fr)
Inventor
刘彬
苑森康
Original Assignee
苏州浪潮智能科技有限公司
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 苏州浪潮智能科技有限公司 filed Critical 苏州浪潮智能科技有限公司
Publication of WO2022052404A1 publication Critical patent/WO2022052404A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • 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
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection

Definitions

  • the invention relates to the technical field of image processing, in particular to a machine vision-based memory alignment method.
  • the invention also relates to a machine vision-based memory alignment and plugging system, a device and a storage medium.
  • the server is an important part of electronic equipment and is a device that provides computing services. Since the server needs to respond to the service request and process it, generally the server should have the ability to undertake and guarantee the service. According to the different types of services provided by the server, it is divided into file server, database server, application server, WEB server, etc.
  • IT equipment In the era of big data, a large number of IT equipment will be centrally placed in data centers. These data centers contain various types of servers, storage, switches, and numerous racks and other infrastructure. Each type of IT equipment is composed of various hardware boards, such as computing modules, memory modules, storage modules, and chassis.
  • the memory module is one of the important components in the computer, and it is the bridge between the memory module and the CPU. All programs in the computer run in the memory, so the performance of the memory has a great impact on the computer.
  • Memory is also an essential component in server architecture, and the number is large.
  • the memory module needs to be inserted into the memory slot to complete the memory insertion operation.
  • the insertion of the memory stick and the memory slot is generally done manually by means of a memory installation jig.
  • the manual operation efficiency is low, the plugging and unplugging operations are time-consuming and labor-intensive, the labor intensity of workers is high, and the hands are easily injured.
  • the number of memory sticks is large, it is easy to cause inaccurate alignment between the memory stick and the memory slot due to external environmental influences, such as vibration, etc. during assembly, which in turn causes the memory stick to collide with the memory socket during insertion.
  • the gold fingers of the memory module are damaged or broken, and the memory slot is damaged.
  • the purpose of the present invention is to provide a machine vision-based memory alignment and insertion method, which can improve the operation efficiency of memory insertion, ensure the accurate alignment of memory sticks and memory slots, and prevent collision accidents during the insertion operation.
  • Another object of the present invention is to provide a machine vision-based memory alignment plug-in system.
  • the present invention provides a machine vision-based memory alignment method, including:
  • the mobile camera is driven to move to the initial position by the mechanical arm, and the mobile camera obtains the initial position of the memory module through image processing technology;
  • the image acquisition of the memory slot is performed by the mobile camera, the current position of the memory slot is calculated by image processing technology, and the compensation bit is calculated according to the current position and the insertion reference position of the memory slot. state;
  • the current insertion position is calculated according to the to-be-assembled position and the compensation position, and the robotic arm is moved to the current insertion position to perform the memory module insertion operation.
  • the method before driving the mobile camera to move to the initial position synchronously by the robotic arm, the method further includes:
  • the coordinates of the photographed images of the mobile camera and the fixed camera and the position coordinates of the mechanical arm are calibrated to obtain the coordinate conversion relationship between the image of the mobile camera and the mechanical arm and the coordinate transformation of the fixed camera.
  • the coordinate transformation relationship between the image and the robotic arm is calibrated to obtain the coordinate conversion relationship between the image of the mobile camera and the mechanical arm and the coordinate transformation of the fixed camera.
  • the mobile camera obtains the initial position of the memory module through image processing technology, which specifically includes:
  • the method before acquiring the initial position state of the memory module, the method further includes:
  • the grasping reference position of the robotic arm, the assembly reference position of the memory stick, and the insertion reference position of the memory slot are acquired.
  • acquiring the grasping reference position of the robotic arm specifically includes:
  • the memory stick is grabbed by the mechanical arm and vertically raised by a preset distance to a grabbing reference position.
  • obtaining the assembly reference position of the memory module specifically includes:
  • the memory stick is kept grasped by the mechanical arm and moved to the photographing position of the fixed camera;
  • the image of the memory stick is acquired by the fixed camera, and the assembly reference position of the memory stick is calculated.
  • acquiring the insertion reference position of the memory slot specifically includes:
  • the mobile camera is driven by the mechanical arm to move to an insertion reference position at a preset distance above the memory slot.
  • the present invention also provides a machine vision-based memory alignment plug-in system, comprising:
  • the initial acquisition module is used to drive the mobile camera to move to the initial position through the mechanical arm, and make the mobile camera obtain the initial position of the memory stick through image processing technology;
  • the memory grab module is used to calculate the offset of the mechanical arm according to the deviation between the initial position state of the memory stick and the grab reference position, and move the mechanical arm to grab according to the offset position to grab the memory stick;
  • the attitude adjustment module is used to move the mechanical arm to the photographing position of the fixed camera, and calculate the post-grab position of the memory stick through image processing technology, and then calculate the post-grab position and the assembly reference position according to the deviation value of the post-grab position moving the robotic arm to the position to be assembled;
  • the slot positioning module is used to obtain an image of the memory slot through the mobile camera and calculate the current position of the memory slot through image processing technology, and then according to the current position and the insertion of the memory slot. Install the reference position to calculate the compensation position;
  • the alignment insertion module is used to calculate the current insertion position according to the to-be-assembled position and the compensation position, and move the mechanical arm to the current insertion position to perform the memory module insertion operation.
  • the present invention also provides an electronic device, comprising:
  • the processor is configured to implement the steps of the machine vision-based memory alignment method according to any one of the above when executing the computer program.
  • the present invention also provides a storage medium, where a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the method for aligning and inserting a memory based on machine vision according to any one of the above are implemented.
  • the method for aligning and inserting memory based on machine vision mainly includes five steps. Among them, in the first step, the mobile camera is first driven to move to the initial position by the movement of the mechanical arm, and then the mobile camera is used to obtain the initial position state of the memory module (the position state includes position and attitude) through image processing technology. In the second step, calculate the deviation value according to the calculated initial position of the memory stick and the pre-set grab reference position to obtain the offset of the robotic arm, and then move the robotic arm to Grab the location, and then grab the memory stick.
  • the first step the mobile camera is first driven to move to the initial position by the movement of the mechanical arm, and then the mobile camera is used to obtain the initial position state of the memory module (the position state includes position and attitude) through image processing technology.
  • the second step calculate the deviation value according to the calculated initial position of the memory stick and the pre-set grab reference position to obtain the offset of the robotic arm, and then move the robotic arm to Grab the location, and then grab the memory stick.
  • the third step after grabbing the memory stick, it needs to be moved to the memory slot, and the memory slot is located in the area where the fixed camera is located, so move the robotic arm to the photographing position of the fixed camera, and then use the image processing technology through the fixed camera Calculate the post-grab position of the memory stick, and then calculate the deviation value according to the post-grab position of the memory stick and the pre-set assembly reference position, and then move the robotic arm to the position to be assembled according to the deviation value.
  • the image of the memory slot is first acquired by moving the camera, and then the current position of the memory slot is calculated by image processing technology, and then the current position and the pre- Compensation position is calculated from the deviation value of the set insertion reference position.
  • the final insertion position can be calculated according to the position to be assembled and the compensation position, and then the robotic arm can be moved to the insertion position to perform the memory module insertion operation.
  • the machine vision-based memory alignment and plugging method acquires images of the robotic arm, the memory stick and the memory slot by using a moving camera and a fixed camera, and uses image processing technology to calculate the robotic arm, the memory stick and the memory socket.
  • the coordinate position of the slot and then use the deviation value between the real-time position and the preset reference position to correct and guide the robot arm to move during the insertion operation of the memory module, and finally introduce the position of the memory slot. state compensation, so that the memory module remains aligned with the memory slot before being inserted. Therefore, the present invention can improve the operation efficiency of memory insertion, ensure the accurate alignment of the memory stick and the memory slot, and prevent collision accidents during the insertion operation.
  • FIG. 1 is a flow chart of a method according to a specific embodiment provided by the present invention.
  • FIG. 2 is a system structure diagram of a specific implementation manner provided by the present invention.
  • Initial acquisition module-1 memory grab module-2, attitude adjustment module-3, slot positioning module-4, alignment insertion module-5.
  • FIG. 1 is a schematic diagram of the overall structure of a specific embodiment provided by the present invention.
  • the method for aligning and inserting memory based on machine vision mainly includes five steps, which are:
  • step S1 since the mobile camera is set on the mechanical arm, the mobile camera can be driven to move to the initial position by the movement of the mechanical arm, and then the mobile camera can obtain the initial position (position state) of the memory module through image processing technology. including position and attitude).
  • step S2 the deviation value is calculated according to the calculated initial position of the memory stick and the preset grasping reference position to obtain the offset of the robotic arm, and then the robotic arm is moved to the grasping position according to the offset. Take the location, and then grab the memory stick.
  • step S3 after grabbing the memory stick, it needs to be moved to the memory slot, and the memory slot is located in the area where the fixed camera (fixed) is located, so move the mechanical arm to the photographing position of the fixed camera, and then pass the fixed camera Image processing technology is used to calculate the post-grab position of the memory module, and then the deviation value is calculated according to the post-grab position of the memory module and the pre-set assembly reference position, and then the robot arm is moved to the position to be assembled according to the deviation value.
  • step S4 in order to clarify the position of the current memory slot, firstly, the image of the memory slot is acquired by moving the camera, and then the current position of the memory slot is calculated by using image processing technology, and then the current position and the preset Compensation position is calculated based on the deviation value of the specified insertion reference position.
  • step S5 the final insertion position can be calculated according to the position to be assembled and the compensation position, and then the robotic arm can be moved to the insertion position to perform the memory module insertion operation.
  • images of the robotic arm, memory sticks and memory slots are acquired by moving cameras and fixed cameras, and image processing technology is used to calculate the robotic arms, memory sticks and memory slots.
  • the coordinate position of the slot and then use the deviation value between the real-time position and the preset reference position to correct and guide the robot arm to move during the insertion of the memory module, and finally introduce the memory slot.
  • a calibration process is first performed before the memory module is inserted.
  • the calibration process is mainly used to determine the transformation relationship between the coordinates of the robotic arm and the coordinates of the images captured by the mobile camera and the fixed camera according to the actual equipment and environment.
  • the "Eye-in-Hand” hand-eye calibration method can be used for the mobile camera in the calibration process, while the fixed camera is fixed outside the robot arm and does not move with the movement of the robot arm, so the eye-in-hand method is adopted.
  • the "Eye-on-Hand” hand-eye calibration method the only difference between the two is the installation method of the camera.
  • the calibration principle in this article is also the nine-point calibration method used.
  • the principle of nine-point calibration is to first move the manipulator so that there is a Mark point in the image field of the moving camera and the fixed camera, then use template matching or connected domain analysis to find the position of the Mark point, and then move the manipulator nine times, so that each Mark point Mark points are located at different positions in the image and can be found in the image field of view, and the robot arm coordinates and image coordinates are saved nine times respectively.
  • the rotation matrix T is:
  • the translation matrix M is:
  • step S1 the mobile camera is used to obtain the initial position of the memory module through image processing technology, which specifically includes:
  • the mobile camera on the robotic arm can be used to first judge the presence or absence of the memory stick, and then when confirming the existence of the memory stick, use the method of connected domain analysis to find the outline of the memory stick in a specific ROI area, and then use the ROI area according to the memory stick.
  • the contour is divided into multiple sub-regions, and then the values of x (length size) and y (width size) of these sub-regions are obtained respectively, and then the position of each sub-region in the ROI area is judged according to the value of x; then select the leftmost Or the rightmost position is the position of the currently required memory module, and find the minimum enclosing rectangle of the corresponding sub-region, and then move the center point of the smallest enclosing rectangle to the right to divide it into multiple small sub-rectangles, and place the minimum enclosing rectangle in each small sub-rectangle.
  • the canny operator used in image processing is used to obtain points with large gradient changes. Finally, a straight line is fitted according to the calculated points.
  • the fitted straight line is the edge contour of the memory module.
  • the angle of the memory stick can be calculated from the point-slope equation of the straight line, and then find the center point of the minimum circumscribed rectangle of the memory stick, and spread along the center point in the direction of the angle of the memory stick on both sides.
  • the position of the center point of the most edge of the memory bank in the image can be obtained, and finally the initial position of the memory bank can be calculated according to the position, length and angle of the edge contour.
  • the Each reference position is corrected and determined before operation according to the actual equipment and environment.
  • the method for obtaining the grasping reference position of the robotic arm and the assembly reference position of the memory module specifically includes:
  • the robotic arm manually move the robotic arm to grab the memory stick and move it to the photo-taking position of the fixed camera for photo processing, and determine the coordinates p9 (x9, y9, q9) of the memory stick.
  • the base position of the groove In this way, the photographing reference position p6 of the robotic arm, the first coordinate reference position p7 of the memory stick, the grasping reference position p8 of the robotic arm, and the assembly reference position p9 of the memory stick can be determined respectively.
  • q is the attitude angle.
  • the method of obtaining the insertion reference position of the memory slot specifically includes:
  • Image processing and identification are performed on the location of the memory slot, the coordinates of the memory slot are calculated, and the coordinates p2 (x2, y2, q2) of the robotic arm and the coordinates p3 (x3, y3, q3) of the memory slot are saved respectively.
  • the point p1 is the insertion reference position when the memory module is inserted
  • the point p2 is the identification reference position for identifying the memory slot
  • p3 is the second coordinate reference position when the memory slot is identified.
  • step S4 the current position of the memory slot is first calculated by moving the camera to be p4 (x4, y4, q4), and then the difference value is calculated with the aforementioned second coordinate reference position p3, and then calculated according to the difference value.
  • Compensation position p5 (x5, y5, q5) compared to the reference:
  • step S2 when calculating the offset (xp, yp, qp) of the robotic arm, the initial position of the memory stick can be followed by the photographing reference position, the first coordinate reference position and the grasping reference. The difference between the positions calculates the total offset, and then corrects the grasping position of the robotic arm.
  • the coordinates of the grasping position are p10 (x10, y10, q10):
  • step S3 the post-grab position of the memory stick is calculated by fixing the camera to be p11 (x11, y11, q11), and then the difference between it and the assembly reference position p9 (x9, y9, q9) is calculated, that is, The coordinates of the to-be-assembled position of the memory stick can be obtained as p12 (x12, y12, q12):
  • FIG. 2 is a system structure diagram of a specific implementation manner provided by the present invention.
  • This embodiment also provides a machine vision-based memory alignment and insertion system, which mainly includes an initial acquisition module 1 , a memory grabbing module 2 , an attitude adjustment module 3 , a slot positioning module 4 and an alignment insertion module 5 .
  • the initial acquisition module 1 is mainly used to drive the mobile camera to move to the initial position through the mechanical arm, and enable the mobile camera to obtain the initial position of the memory module through image processing technology.
  • the memory grabbing module 2 is mainly used to calculate the offset of the robotic arm according to the deviation between the initial position of the memory stick and the grabbing reference state, and move the robotic arm to the grabbing position to grab the memory stick according to the offset.
  • Attitude adjustment module 3 is mainly used to move the robotic arm to the photographing position of the fixed camera, and calculate the post-grab position of the memory module through image processing technology, and then move the manipulator to the position according to the deviation between the post-grab position and the assembly reference position. position to be assembled.
  • the slot positioning module 4 is mainly used to obtain the image of the memory slot by moving the camera, calculate the current position of the memory slot through image processing technology, and then calculate the compensation position according to the current position and the insertion reference position of the memory slot. state.
  • the alignment plug-in module 5 is mainly used to calculate the current plug-in position according to the position to be assembled and the compensation position, and move the mechanical arm to the current plug-in position to perform the memory module plug-in operation.
  • This embodiment also provides a device, which mainly includes a memory and a processor.
  • the memory is mainly used to store the computer program
  • the processor is mainly used to execute the computer program, so as to realize the above-mentioned machine vision-based memory alignment method in the process of executing the computer program.
  • the device may be a server, or may be a terminal device such as a smart phone, a tablet computer, a palmtop computer, or a portable computer.
  • This embodiment also provides a storage medium, on which the aforementioned computer program is stored, so that when the computer program is executed by the processor, the aforementioned method for aligning and inserting memory based on machine vision is implemented.
  • the storage medium may be a USB flash drive, a removable hard disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, etc. medium of program code.
  • ROM Read-Only Memory
  • RAM Random Access Memory

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Manipulator (AREA)
  • Image Analysis (AREA)

Abstract

一种基于机器视觉内存对位接插方法及***、设备、存储介质，包括：通过机械臂带动移动相机运动至初始位置，并获取内存条的初始位态；根据内存条的初始位态与抓取基准位态的偏差值移动机械臂至抓取位置抓取内存条；移动机械臂至固定相机的拍照位置，并通计算内存条的抓后位态，再据其与装配基准位态的偏差值将机械臂移动至待装配位置；对内存插槽进行图像获取并计算其当前位态，再据其与内存插槽的插装基准位态计算补偿位态；根据待装配位置与补偿位态计算当前插装位态，并将机械臂移动至当前插装位置进行内存条插装作业。所述方法能够提高内存插装的作业效率，保证内存条与内存插槽对位准确，防止在插装作业过程中出现碰撞事故。

Description

基于机器视觉内存对位接插方法与***、设备、存储介质
本申请要求于2020年9月9日提交中国专利局、申请号为202010940989.X、发明名称为“基于机器视觉内存对位接插方法与***、设备、存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及图像处理技术领域，特别涉及一种基于机器视觉内存对位接插方法。本发明还涉及一种基于机器视觉内存对位接插***、一种设备和一种存储介质。
背景技术
随着中国电子技术的发展，越来越多的电子设备已得到广泛使用。
服务器是电子设备中的重要组成部分，是提供计算服务的设备。由于服务器需要响应服务请求，并进行处理，因此一般来说服务器应具备承担服务并且保障服务的能力。根据服务器提供的服务类型不同，分为文件服务器、数据库服务器、应用程序服务器、WEB服务器等。
在大数据时代，大量的IT设备会集中放置在数据中心。这些数据中心包含各类型的服务器、存储、交换机及大量的机柜及其它基础设施。每种IT设备都是有各种硬件板卡组成，如计算模块、内存模块、存储模块、机箱等。内存模块是计算机中重要的部件之一，它是存储模块与CPU进行沟通的桥梁。计算机中所有程序的运行都是在内存中进行的，因此内存的性能对计算机的影响非常大。在服务器架构中内存也是必不可少的元器件，且数量较多。
目前，在产线大批量生产作业中，内存条需要***内存插槽完成内存插装作业。在现有技术中，一般通过人工借助内存安装治具手动完成内存条与内存插槽的插装。然而，由于主板安装密度大，内存条安装数量较多且一般满配运行，因此人工作业效率较低，插拔操作费时费力，工人劳动强度较大且易伤手。并且，在内存条插装数量较多时，容易在装配时因为外界环境影响，比如振动等而导致内存条与内存插槽间的对位不准确，进 而导致内存条在插装时碰撞到内存插槽的槽壁上，造成内存条的金手指损坏或断裂，以及内存插槽的损坏。
因此，如何提高内存插装的作业效率，保证内存条与内存插槽对位准确，防止在插装作业过程中出现碰撞事故，是本领域技术人员面临的技术问题。
发明内容
本发明的目的是提供一种基于机器视觉内存对位接插方法，能够提高内存插装的作业效率，保证内存条与内存插槽对位准确，防止在插装作业过程中出现碰撞事故。本发明的另一目的是提供一种基于机器视觉内存对位接插***。
为解决上述技术问题，本发明提供一种基于机器视觉内存对位接插方法，包括：
通过机械臂带动移动相机运动至初始位置，并使所述移动相机通过图像处理技术获取内存条的初始位态；
根据所述内存条的初始位态与抓取基准位态的偏差值计算所述机械臂的偏移量，并根据所述偏移量移动所述机械臂至抓取位置抓取所述内存条；
移动所述机械臂至固定相机的拍照位置，并通过图像处理技术计算所述内存条的抓后位态，再根据所述抓后位态与装配基准位态的偏差值将所述机械臂移动至待装配位置；
通过所述移动相机对内存插槽进行图像获取并通过图像处理技术计算所述内存插槽的当前位态，再根据所述当前位态与所述内存插槽的插装基准位态计算补偿位态；
根据所述待装配位置与所述补偿位态计算当前插装位态，并将所述机械臂移动至当前插装位置进行内存条插装作业。
优选地，在通过机械臂带动移动相机同步运动至初始位置之前，还包括：
对所述移动相机及所述固定相机的拍照图像坐标与所述机械臂的位置坐标进行标定，以获得所述移动相机的图像与所述机械臂之间的坐标转换关系以及所述固定相机的图像与所述机械臂之间的坐标转换关系。
优选地，使移动相机通过图像处理技术获取内存条的初始位态，具体包括：
使所述移动相机通过连通域分析方法在预设ROI区域内获取所述内存条的轮廓，并通过直线拟合出所述内存条的边缘轮廓线，再根据所述边缘轮廓线计算所述内存条的初始位态。
优选地，在获取内存条的初始位态之前，还包括：
获取所述机械臂的抓取基准位态、所述内存条的装配基准位态与所述内存插槽的插装基准位态。
优选地，获取所述机械臂的抓取基准位态，具体包括：
通过所述机械臂带动所述移动相机运动至可获取所述内存条图像的拍照基准位态；
通过所述移动相机获取所述内存条的图像并计算所述内存条的第一坐标基准位态；
通过所述机械臂抓取所述内存条并垂直上升预设距离至抓取基准位态。
优选地，获取所述内存条的装配基准位态，具体包括：
通过所述机械臂对所述内存条保持抓取并移动至所述固定相机的拍照位置；
通过所述固定相机获取所述内存条的图像并计算所述内存条的装配基准位态。
优选地，获取所述内存插槽的插装基准位态，具体包括：
通过所述机械臂带动所述移动相机运动至可获取所述内存插槽图像的识别基准位态；
通过所述移动相机获取所述内存插槽的图像并计算所述内存插槽的第二坐标基准位态；
通过所述机械臂带动所述移动相机运动至所述内存插槽上方预设距离的插装基准位态。
本发明还提供一种基于机器视觉内存对位接插***，包括：
初始获取模块，用于通过机械臂带动移动相机运动至初始位置，并使 所述移动相机通过图像处理技术获取内存条的初始位态；
内存抓取模块，用于根据所述内存条的初始位态与抓取基准位态的偏差值计算所述机械臂的偏移量，并根据所述偏移量移动所述机械臂至抓取位置抓取所述内存条；
姿态调整模块，用于移动所述机械臂至固定相机的拍照位置，并通过图像处理技术计算所述内存条的抓后位态，再根据所述抓后位态与装配基准位态的偏差值将所述机械臂移动至待装配位置；
插槽定位模块，用于通过所述移动相机对内存插槽进行图像获取并通过图像处理技术计算所述内存插槽的当前位态，再根据所述当前位态与所述内存插槽的插装基准位态计算补偿位态；
对位插装模块，用于根据所述待装配位置与所述补偿位态计算当前插装位态，并将所述机械臂移动至当前插装位置进行内存条插装作业。
本发明还提供一种电子设备，包括：
存储器，用于存储计算机程序；
处理器，用于执行所述计算机程序时实现如上述任一项所述基于机器视觉内存对位接插方法的步骤。
本发明还提供一种存储介质，所述存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如上述任一项所述基于机器视觉内存对位接插方法的步骤。
本发明所提供的基于机器视觉内存对位接插方法，主要包括五个步骤。其中，在第一步中，首先通过机械臂的运动带动移动相机运动到初始位置处，然后使移动相机通过图像处理技术获取内存条的初始位态(位态包括位置和姿态)。在第二步中，根据计算出的内存条的初始位态与预先设定的抓取基准位态进行偏差值计算，得出机械臂的偏移量，然后根据该偏移量移动机械臂至抓取位置处，再抓取内存条。在第三步中，抓取内存条之后还需要移动到内存插槽处，而内存插槽位于固定相机所在区域，如此移动机械臂到固定相机的拍照位置处，然后通过固定相机利用图像处理技术计算内存条的抓后位态，再根据内存条的抓后位态与预先设定的装配基准位态进行偏差值计算，再根据该偏差值将机械臂移动到待装配位置。在第四 步中，为明确当前内存插槽的位态，首先通过移动相机对内存插槽进行图像获取，然后利用图像处理技术计算内存插槽的当前位态，再根据其当前位态与预先设定的插装基准位态的偏差值计算补偿位态。在第五步中，即可根据待装配位置与补偿位态计算出最终的插装位态，然后将机械臂移动到插装位置即可进行内存条插装作业。如此，本发明所提供的基于机器视觉内存对位接插方法，通过移动相机和固定相机对机械臂、内存条和内存插槽进行图像获取，利用图像处理技术计算机械臂、内存条和内存插槽的坐标位态，再在内存条的插装作业过程中利用其实时位态与预先设定好的基准位态之间的偏差值修正、引导机械臂进行运动，最后引入内存插槽的位态补偿，使得内存条在进行插装前与内存插槽保持对齐。因此，本发明能够提高内存插装的作业效率，保证内存条与内存插槽对位准确，防止在插装作业过程中出现碰撞事故。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的实施例，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，还可以根据提供的附图获得其他的附图。
图1为本发明所提供的一种具体实施方式的方法流程图。
图2为本发明所提供的一种具体实施方式的***结构图。
其中，图2中：
初始获取模块—1，内存抓取模块—2，姿态调整模块—3，插槽定位模块—4，对位插装模块—5。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
请参考图1，图1为本发明所提供的一种具体实施方式的整体结构示意图。
在本发明所提供的一种具体实施方式中，基于机器视觉内存对位接插方法，主要包括五个步骤，分别为：
S1、通过机械臂带动移动相机运动至初始位置，并使移动相机通过图像处理技术获取内存条的初始位态；
S2、根据内存条的初始位态与抓取基准位态的偏差值计算机械臂的偏移量，并根据偏移量移动机械臂至抓取位置抓取内存条；
S3、移动机械臂至固定相机的拍照位置，并通过图像处理技术计算内存条的抓后位态，再根据抓后位态与装配基准位态的偏差值将机械臂移动至待装配位置；
S4、通过移动相机对内存插槽进行图像获取并通过图像处理技术计算内存插槽的当前位态，再根据当前位态与内存插槽的插装基准位态计算补偿位态；
S5、根据待装配位置与补偿位态计算当前插装位态，并将机械臂移动至当前插装位置进行内存条插装作业。
其中，在步骤S1中，由于移动相机设置在机械臂上，因此可通过机械臂的运动带动移动相机运动到初始位置处，然后使移动相机通过图像处理技术获取内存条的初始位态(位态包括位置和姿态)。
在步骤S2中，根据计算出的内存条的初始位态与预先设定的抓取基准位态进行偏差值计算，得出机械臂的偏移量，然后根据该偏移量移动机械臂至抓取位置处，再抓取内存条。
在步骤S3中，抓取内存条之后还需要移动到内存插槽处，而内存插槽位于固定相机(固定不动)所在区域，如此移动机械臂到固定相机的拍照位置处，然后通过固定相机利用图像处理技术计算内存条的抓后位态，再根据内存条的抓后位态与预先设定的装配基准位态进行偏差值计算，再根据该偏差值将机械臂移动到待装配位置。
在步骤S4中，为明确当前内存插槽的位态，首先通过移动相机对内存插槽进行图像获取，然后利用图像处理技术计算内存插槽的当前位态，再根据其当前位态与预先设定的插装基准位态的偏差值计算补偿位态。
在步骤S5中，即可根据待装配位置与补偿位态计算出最终的插装位态，然后将机械臂移动到插装位置即可进行内存条插装作业。
如此，本实施例所提供的基于机器视觉内存对位接插方法，通过移动相机和固定相机对机械臂、内存条和内存插槽进行图像获取，利用图像处理技术计算机械臂、内存条和内存插槽的坐标位态，再在内存条的插装作业过程中利用其实时位态与预先设定好的基准位态之间的偏差值修正、引导机械臂进行运动，最后引入内存插槽的位态补偿，使得内存条在进行插装前与内存插槽保持对齐。因此，本实施例能够提高内存插装的作业效率，保证内存条与内存插槽对位准确，防止在插装作业过程中出现碰撞事故。
此外，考虑到机械臂的坐标可根据其驱动机构构建的空间坐标系进行精确定位，而移动相机和固定相机对于内存条的坐标定位方式仅能通过拍照并进行图像处理技术进行计算获得，如此，为方便移动相机和固定相机通过图像获取方式确定内存条的坐标位态，本实施例中在对内存条进行插装作业之前首先进行了标定流程。该标定流程主要用于根据实际设备和环境确定机械臂的坐标与移动相机和固定相机所拍摄的图像坐标之间的转换关系。
具体的，标定流程中移动相机可采用眼在手上“Eye-in-Hand”的手眼标定方法，而固定相机由于固定在机械臂外，不随着机械臂的运动而运动，因此采用眼在手外的“Eye-on-Hand”手眼标定方法，两者只是相机的安装方式不同，在本文中的标定的原理同样都是用到的九点标定的方式。九点标定原理是首先移动机械臂使移动相机和固定相机的图像视野中存在一个Mark点，然后使用模板匹配或连通域分析方法找到Mark点的位置，之后移动机械手九次，使得每次Mark点在图像中位于不同的位置且都能在图像视野中找到Mark点，分别保存九次的机械臂坐标和图像坐标。设机械臂的坐标为(x，y，z)，因为机械臂在进行移动时可保持Z轴不变，因此只需标定x、y即可。即图像坐标和机械手坐标之间的关系为：
Figure PCTCN2021073258-appb-000001
其中，旋转矩阵T为：
Figure PCTCN2021073258-appb-000002
平移矩阵M为：
Figure PCTCN2021073258-appb-000003
如此可得方程组：
Figure PCTCN2021073258-appb-000004
由方程组可知有六个未知数，需要至少三组点才能解出来这六组数，然而为了保证数值的精度，一般会选择更多的点使用最小二乘法求解出来一个精度较高的解，参考机械臂精度和像素精度，九组点的时候就可以保证得出的矩阵精度达到要求，所以一般使用九组点去求解旋转和平移矩阵。为方便区分移动相机与固定相机，可设移动相机的坐标转换关系矩阵为Hs，固定相机的坐标转换关系矩阵为Hx。
在步骤S1中，使移动相机通过图像处理技术获取内存条的初始位态，具体包括：
首先可用机械臂上的移动相机先进行内存条有无的判断，然后确认内存条存在时，使用连通域分析的方法在特定的ROI区域中找到内存条的轮廓，之后可ROI区域中根据内存条的轮廓分割为多个子区域，然后分别得到这些子区域的x(长度尺寸)和y(宽度尺寸)的值，再根据x的值去判断各个子区域在ROI区域中的位置；之后选择最左边或者最右边的位置为当前所需的内存条位置，并找到对应子区域的最小外接矩形，再把该最小外界矩形的中心点右移分割成为多个小的子矩形，并在每个小子矩形内使用图像处理的canny算子得到梯度变化较大的点，最后根据计算出的多个点进行直线的拟合，拟合出来的直线即为内存条的边缘轮廓线。接下来只需根据内存条的边缘轮廓线，由直线的点斜式方程可以求出内存条的角度，再找到内存条最小外接矩形的中心点，沿着中心点以内存条的角度方向两 边扩散可以求出内存条在图像中的最边缘的中心点位置，最后根据边缘轮廓线的位置、长度和角度即可算出内存条的初始位态。
另外，为提高机械臂的抓取基准位态、内存条的装配基准位态和内存插槽的插装基准位态的精确性，本实施例在获取内存条的初始位态之前，还可事先根据实际设备和环境对各个基准位态进行作业前修正和确定。
具体的，获取机械臂的抓取基准位态和内存条的装配基准位态的方法具体包括：
首先带动机械臂带动移动相机运动到预设的第一个拍照位置处进行拍照，此时保存机械臂的坐标p6(x6，y6，q6)和在图像中内存条的坐标p7(x7，y7，q7)，然后手动移动机械臂带动移动相机抓取内存条，抓取到内存条之后，只改变Z坐标到拍照高度，记录此时的机械臂坐标p8(x8，y8，q8)为机械臂需要抓取的基准位置。然后手动移动机械臂抓取内存条移动至固定相机的拍照位置处进行拍照处理，确定内存条的坐标p9(x9，y9，q9)，该坐标为机械臂抓取内存条之后要放进内存插槽的基准位置。如此，即可分别确定机械臂的拍照基准位态p6、内存条的第一坐标基准位态p7、机械臂的抓取基准位态p8、内存条的装配基准位态p9。其中，q为姿态角度。
获取内存插槽的插装基准位态的方法具体包括：
使用机械臂引导移动相机运动内存插槽的上方预设距离位置处，缓缓引导机械臂把内存条***内存插槽，***内存插槽之后，再抬高到移动相机的拍照高度，保证此时机械臂的x，y坐标值不变，并记录此时的机械臂坐标为p1(x1，y1，q1)，之后移动机械臂到移动相机视野中可以清楚拍摄内存插槽的位置，找到此时内存插槽的位置进行图像处理和识别，计算出内存插槽的坐标，分别保存此时机械臂的坐标p2(x2，y2，q2)和内存插槽的坐标p3(x3，y3，q3)。其中，点p1即为插装内存条时的插装基准位态、点p2即为识别内存插槽的识别基准位态、p3即为识别内存插槽时的第二坐标基准位态。
在步骤S4中，首先通过移动相机计算出内存插槽的当前位态为p4(x4，y4，q4)，然后与前述第二坐标基准位态p3进行差值计算，再根据该差值计算得到较基准的补偿位态p5(x5，y5，q5)：
Figure PCTCN2021073258-appb-000005
同理，在步骤S2中，再计算机械臂的偏移量(xp、yp、qp)时，可根据内存条的初始位态依次与拍照基准位态、第一坐标基准位态和抓取基准位态之间的差值计算总的偏移量，然后对机械臂的抓取位置进行修正，抓取位置的坐标为p10(x10、y10、q10)：
Figure PCTCN2021073258-appb-000006
此外，在步骤S3中，通过固定相机计算内存条的抓后位态为p11(x11，y11，q11)，之后将其与装配基准位态p9(x9，y9，q9)进行差值计算，即可获得内存条的待装配位置的坐标为p12(x12，y12，q12)：
Figure PCTCN2021073258-appb-000007
在步骤S4中，补偿位态的计算方法为将内存条的待装配位置的坐标p12与补偿位态p5的坐标相结合，即p13(x13，y13，q13)＝p13(x12+x5，y12+y5，q12+q5)。
如图2所示，图2为本发明所提供的一种具体实施方式的***结构图。
本实施例还提供一种基于机器视觉内存对位接插***，主要包括初始获取模块1、内存抓取模块2、姿态调整模块3、插槽定位模块4和对位插装模块5。
其中，初始获取模块1主要用于通过机械臂带动移动相机运动至初始位置，并使移动相机通过图像处理技术获取内存条的初始位态。内存抓取模块2主要用于根据内存条的初始位态与抓取基准位态的偏差值计算机械臂的偏移量，并根据偏移量移动机械臂至抓取位置抓取内存条。姿态调整 模块3主要用于移动机械臂至固定相机的拍照位置，并通过图像处理技术计算内存条的抓后位态，再根据抓后位态与装配基准位态的偏差值将机械臂移动至待装配位置。插槽定位模块4主要用于通过移动相机对内存插槽进行图像获取并通过图像处理技术计算内存插槽的当前位态，再根据当前位态与内存插槽的插装基准位态计算补偿位态。对位插装模块5主要用于根据待装配位置与补偿位态计算当前插装位态，并将机械臂移动至当前插装位置进行内存条插装作业。
本实施例还提供一种设备，主要包括存储器和处理器。其中，存储器主要用于存储计算机程序，而处理器主要用于执行该计算机程序，以在执行计算机程序的过程中实现如前所述的基于机器视觉内存对位接插方法。
在本实施例中，设备可以是服务器，也可以是智能手机、平板电脑、掌上电脑、便携计算机等终端设备。
本实施例还提供一种存储介质，该存储介质上存储有前述计算机程序，以便该计算机程序被处理器执行时实现如前所述的基于机器视觉内存对位接插方法。
在本实施例中，该存储介质可以为U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本发明。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况下，在其它实施例中实现。因此，本发明将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (10)

  1. 一种基于机器视觉内存对位接插方法，其特征在于，包括：
    通过机械臂带动移动相机运动至初始位置，并使所述移动相机通过图像处理技术获取内存条的初始位态；
    根据所述内存条的初始位态与抓取基准位态的偏差值计算所述机械臂的偏移量，并根据所述偏移量移动所述机械臂至抓取位置抓取所述内存条；
    移动所述机械臂至固定相机的拍照位置，并通过图像处理技术计算所述内存条的抓后位态，再根据所述抓后位态与装配基准位态的偏差值将所述机械臂移动至待装配位置；
    通过所述移动相机对内存插槽进行图像获取并通过图像处理技术计算所述内存插槽的当前位态，再根据所述当前位态与所述内存插槽的插装基准位态计算补偿位态；
    根据所述待装配位置与所述补偿位态计算当前插装位态，并将所述机械臂移动至当前插装位置进行内存条插装作业。
  2. 根据权利要求1所述的基于机器视觉内存对位接插方法，其特征在于，在通过机械臂带动移动相机同步运动至初始位置之前，还包括：
    对所述移动相机及所述固定相机的拍照图像坐标与所述机械臂的位置坐标进行标定，以获得所述移动相机的图像与所述机械臂之间的坐标转换关系以及所述固定相机的图像与所述机械臂之间的坐标转换关系。
  3. 根据权利要求1所述的基于机器视觉内存对位接插方法，其特征在于，使移动相机通过图像处理技术获取内存条的初始位态，具体包括：
    使所述移动相机通过连通域分析方法在预设ROI区域内获取所述内存条的轮廓，并通过直线拟合出所述内存条的边缘轮廓线，再根据所述边缘轮廓线计算所述内存条的初始位态。
  4. 根据权利要求1所述的基于机器视觉内存对位接插方法，其特征在于，在获取内存条的初始位态之前，还包括：
    获取所述机械臂的抓取基准位态、所述内存条的装配基准位态与所述内存插槽的插装基准位态。
  5. 根据权利要求4所述的基于机器视觉内存对位接插方法，其特征在于，获取所述机械臂的抓取基准位态，具体包括：
    通过所述机械臂带动所述移动相机运动至可获取所述内存条图像的拍照基准位态；
    通过所述移动相机获取所述内存条的图像并计算所述内存条的第一坐标基准位态；
    通过所述机械臂抓取所述内存条并垂直上升预设距离至抓取基准位态。
  6. 根据权利要求5所述的基于机器视觉内存对位接插方法，其特征在于，获取所述内存条的装配基准位态，具体包括：
    通过所述机械臂对所述内存条保持抓取并移动至所述固定相机的拍照位置；
    通过所述固定相机获取所述内存条的图像并计算所述内存条的装配基准位态。
  7. 根据权利要求6所述的基于机器视觉内存对位接插方法，其特征在于，获取所述内存插槽的插装基准位态，具体包括：
    通过所述机械臂带动所述移动相机运动至可获取所述内存插槽图像的识别基准位态；
    通过所述移动相机获取所述内存插槽的图像并计算所述内存插槽的第二坐标基准位态；
    通过所述机械臂带动所述移动相机运动至所述内存插槽上方预设距离的插装基准位态。
  8. 一种基于机器视觉内存对位接插***，其特征在于，包括：
    初始获取模块，用于通过机械臂带动移动相机运动至初始位置，并使所述移动相机通过图像处理技术获取内存条的初始位态；
    内存抓取模块，用于根据所述内存条的初始位态与抓取基准位态的偏差值计算所述机械臂的偏移量，并根据所述偏移量移动所述机械臂至抓取位置抓取所述内存条；
    姿态调整模块，用于移动所述机械臂至固定相机的拍照位置，并通过图像处理技术计算所述内存条的抓后位态，再根据所述抓后位态与装配基准位态的偏差值将所述机械臂移动至待装配位置；
    插槽定位模块，用于通过所述移动相机对内存插槽进行图像获取并通过图像处理技术计算所述内存插槽的当前位态，再根据所述当前位态与所述内存插槽的插装基准位态计算补偿位态；
    对位插装模块，用于根据所述待装配位置与所述补偿位态计算当前插装位态，并将所述机械臂移动至当前插装位置进行内存条插装作业。
  9. 一种设备，其特征在于，包括：
    存储器，用于存储计算机程序；
    处理器，用于执行所述计算机程序时实现如权利要求1至7任一项所述基于机器视觉内存对位接插方法的步骤。
  10. 一种存储介质，其特征在于，所述存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1至7任一项所述基于机器视觉内存对位接插方法的步骤。
PCT/CN2021/073258 2020-09-09 2021-01-22 基于机器视觉内存对位接插方法与***、设备、存储介质 WO2022052404A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010940989.X 2020-09-09
CN202010940989.XA CN111815634A (zh) 2020-09-09 2020-09-09 基于机器视觉内存对位接插方法与***、设备、存储介质

Publications (1)

Publication Number Publication Date
WO2022052404A1 true WO2022052404A1 (zh) 2022-03-17

Family

ID=72860785

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/073258 WO2022052404A1 (zh) 2020-09-09 2021-01-22 基于机器视觉内存对位接插方法与***、设备、存储介质

Country Status (2)

Country Link
CN (1) CN111815634A (zh)
WO (1) WO2022052404A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115682926A (zh) * 2022-09-07 2023-02-03 广东爱吉尔机器人科技有限公司 一种一对多快速手眼标定的方法
CN115890654A (zh) * 2022-10-09 2023-04-04 北京微链道爱科技有限公司 基于三维特征点的深度相机自动标定算法
CN116141340A (zh) * 2023-04-20 2023-05-23 中国信息通信研究院 基于电路板装配的双机器人柔性协同机械手
CN117021121A (zh) * 2023-10-09 2023-11-10 浪潮(山东)计算机科技有限公司 一种基于机器视觉的内存条自动安装设备及方法
CN117900810A (zh) * 2024-03-15 2024-04-19 中电鹏程智能装备有限公司 一种内存条自动化柔性无损插装方法及插装装置

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111815634A (zh) * 2020-09-09 2020-10-23 苏州浪潮智能科技有限公司 基于机器视觉内存对位接插方法与***、设备、存储介质
US11867755B2 (en) 2021-03-10 2024-01-09 Changxin Memory Technologies, Inc. Memory device test method, apparatus, and system, medium, and electronic device
CN113393447B (zh) * 2021-06-24 2022-08-02 四川启睿克科技有限公司 基于深度学习的针尖正位度检测方法及***
CN113562465B (zh) * 2021-09-26 2021-12-03 成都新西旺自动化科技有限公司 片状物放置视觉引导方法及***
CN114227187B (zh) * 2021-11-30 2023-03-28 浪潮(山东)计算机科技有限公司 一种插接部件安装方法、***及相关组件
CN114725753B (zh) * 2022-02-28 2023-07-18 福建星云电子股份有限公司 一种基于视觉引导的自动对插方法及***

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040102911A1 (en) * 2002-11-21 2004-05-27 Samsung Electronics Co., Ltd. Hand/eye calibration method using projective invariant shape descriptor of 2-dimensional image
CN102692183A (zh) * 2011-03-23 2012-09-26 比比威株式会社 多台摄像机的初始位置和姿势的计量方法
CN106670763A (zh) * 2017-01-10 2017-05-17 荣旗工业科技（苏州）有限公司 一种高精度自动组装机的计算方法
CN110125926A (zh) * 2018-02-08 2019-08-16 比亚迪股份有限公司 自动化的工件取放方法及***
CN110374312A (zh) * 2019-07-03 2019-10-25 广东博智林机器人有限公司 搬运***及其控制方法、地砖铺贴***
CN111815634A (zh) * 2020-09-09 2020-10-23 苏州浪潮智能科技有限公司 基于机器视觉内存对位接插方法与***、设备、存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107443362A (zh) * 2017-09-29 2017-12-08 张浩洋 电脑主板内存条自动插装机
CN108453727B (zh) * 2018-01-11 2020-08-25 中国人民解放军63920部队 基于椭圆特征的机械臂末端位姿误差校正方法及***
CN110480615B (zh) * 2019-08-30 2020-11-10 河南省机械设计研究院有限公司 机器人拆垛定位修正方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040102911A1 (en) * 2002-11-21 2004-05-27 Samsung Electronics Co., Ltd. Hand/eye calibration method using projective invariant shape descriptor of 2-dimensional image
CN102692183A (zh) * 2011-03-23 2012-09-26 比比威株式会社 多台摄像机的初始位置和姿势的计量方法
CN106670763A (zh) * 2017-01-10 2017-05-17 荣旗工业科技（苏州）有限公司 一种高精度自动组装机的计算方法
CN110125926A (zh) * 2018-02-08 2019-08-16 比亚迪股份有限公司 自动化的工件取放方法及***
CN110374312A (zh) * 2019-07-03 2019-10-25 广东博智林机器人有限公司 搬运***及其控制方法、地砖铺贴***
CN111815634A (zh) * 2020-09-09 2020-10-23 苏州浪潮智能科技有限公司 基于机器视觉内存对位接插方法与***、设备、存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WANG HEPING: "Research of Location and Inspection Technology of Plug-in Machine Based on Machine Vision", CHINESE MASTER’S THESES FULL-TEXT DATABASE (ELECTRONIC JOURNALS), no. 7, 31 July 2018 (2018-07-31), XP055910517 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115682926A (zh) * 2022-09-07 2023-02-03 广东爱吉尔机器人科技有限公司 一种一对多快速手眼标定的方法
CN115682926B (zh) * 2022-09-07 2023-08-29 广东爱吉尔机器人科技有限公司 一种一对多快速手眼标定的方法
CN115890654A (zh) * 2022-10-09 2023-04-04 北京微链道爱科技有限公司 基于三维特征点的深度相机自动标定算法
CN115890654B (zh) * 2022-10-09 2023-08-11 北京微链道爱科技有限公司 基于三维特征点的深度相机自动标定算法
CN116141340A (zh) * 2023-04-20 2023-05-23 中国信息通信研究院 基于电路板装配的双机器人柔性协同机械手
CN116141340B (zh) * 2023-04-20 2023-07-14 中国信息通信研究院 基于电路板装配的双机器人柔性协同机械手
CN117021121A (zh) * 2023-10-09 2023-11-10 浪潮(山东)计算机科技有限公司 一种基于机器视觉的内存条自动安装设备及方法
CN117900810A (zh) * 2024-03-15 2024-04-19 中电鹏程智能装备有限公司 一种内存条自动化柔性无损插装方法及插装装置

Also Published As

Publication number Publication date
CN111815634A (zh) 2020-10-23

Similar Documents

Publication Publication Date Title
WO2022052404A1 (zh) 基于机器视觉内存对位接插方法与***、设备、存储介质
US11049280B2 (en) System and method for tying together machine vision coordinate spaces in a guided assembly environment
CN110666798B (zh) 一种基于透视变换模型的机器人视觉标定方法
WO2019114339A1 (zh) 机械臂运动的校正方法和装置
TW200402117A (en) Carriage robot system and controlling method thereof
KR20100065073A (ko) 기판검사방법, 기판검사장치 및 기억매체
CN109788277B (zh) 防抖机芯的光轴偏差的补偿方法、装置和存储介质
US20190149788A1 (en) Calibration method of depth image capturing device
CN112621711B (zh) 机器人及其相机固定于机架的手眼标定方法以及存储介质
JP2941617B2 (ja) 電子部品の部品データ記録装置およびそれを用いた電子部品の搬送組み付け装置
CN113643384B (zh) 坐标系标定方法、自动组装方法及装置
CN112308925A (zh) 可穿戴设备的双目标定方法、设备及存储介质
CN112621743A (zh) 机器人及其相机固定于末端的手眼标定方法及存储介质
WO2022141324A1 (zh) 摄像头的硬件在环标定、靶点设置方法、***及相关设备
CN111591474B (zh) 一种航天器在轨操作***对准式手眼标定方法
CN114952856A (zh) 机械臂手眼标定方法、***、计算机及可读存储介质
CN114664721A (zh) 一种晶圆位置预对准方法、电子设备及晶圆传输***
CN117173254A (zh) 一种相机标定方法、***、装置和电子设备
JP2633147B2 (ja) 部品実装方法
CN111856895B (zh) Ccd曝光机的对位方法、装置及其计算机可读存储介质
CN109615658B (zh) 机器人的物品拿取方法、装置、计算机设备及存储介质
CN115937321B (zh) 一种电子设备的姿态检测方法和装置
JP2012013841A (ja) 露光装置
CN112446928B (zh) 一种拍摄装置的外参确定***和方法
CN117218191A (zh) 一种视觉定位装置、定位方法及可读存储介质

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: 21865468

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: 21865468

Country of ref document: EP

Kind code of ref document: A1