WO2022052404A1

WO2022052404A1 - Memory alignment and insertion method and system based on machine vision, device, and storage medium

Info

Publication number: WO2022052404A1
Application number: PCT/CN2021/073258
Authority: WO
Inventors: 刘彬; 苑森康
Original assignee: 苏州浪潮智能科技有限公司
Priority date: 2020-09-09
Filing date: 2021-01-22
Publication date: 2022-03-17
Also published as: CN111815634A

Abstract

A memory alignment and insertion method and system based on machine vision, a device, and a storage medium. The method comprises: driving, by a mechanical arm, a movable camera to move to an initial position and obtain an initial position and posture of a memory bank; according to a deviation value between the initial position and posture of the memory bank and a reference grabbing position and posture, moving the mechanical arm to a grabbing position to grab the memory bank; moving the mechanical arm to a photographing position of a fixed camera, calculating a post-grabbing position and posture of the memory bank, and according to a deviation value between the post-grabbing position and posture and a reference assembly position and posture, moving the mechanical arm to a position for assembly; obtaining an image of a memory slot and calculating the current position and posture of the memory slot, and calculating a compensation position and posture according to the current position and posture and a reference insertion and assembly position and posture of the memory slot; and calculating a current insertion and assembly position and posture according to the position for assembly and the compensation position and posture, and moving the mechanical arm to a current insertion and assembly position for an insertion and assembly operation of the memory bank. The method can improve the operation efficiency of insertion and assembly of a memory, ensure accurate alignment of the memory bank and the memory slot, and prevent collision accidents during an insertion and assembly operation.

Description

Machine vision-based memory alignment method and system, device, and storage medium

This application claims the priority of the Chinese patent application filed on September 9, 2020, with the application number 202010940989.X and the invention titled "Method and System, Equipment, and Storage Medium for Alignment of Memory Based on Machine Vision" , the entire contents of which are incorporated herein by reference.

technical field

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.

Background technique

With the development of electronic technology in China, more and more electronic devices have been widely used.

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.

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.

At present, in the mass production operation of the production line, the memory module needs to be inserted into the memory slot to complete the memory insertion operation. In the prior art, the insertion of the memory stick and the memory slot is generally done manually by means of a memory installation jig. However, due to the high installation density of the motherboard, the large number of installed memory modules and generally full configuration, 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. In addition, when 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. On the slot wall of the slot, the gold fingers of the memory module are damaged or broken, and the memory slot is damaged.

Therefore, how to 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 is a technical problem faced by those skilled in the art.

SUMMARY OF THE INVENTION

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.

In order to solve the above-mentioned technical problems, 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;

Calculate the offset of the robotic arm according to the deviation between the initial position state of the memory stick and the grab reference position, and move the robotic arm to the grasping position to grab the memory stick according to the offset ;

Move the mechanical 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 mechanical arm according to the deviation between the post-grab position and the assembly reference position to the position to be assembled;

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.

Preferably, 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.

Preferably, the mobile camera obtains the initial position of the memory module through image processing technology, which specifically includes:

Make the mobile camera obtain the outline of the memory bank in the preset ROI area through a connected domain analysis method, and fit the edge contour of the memory bank through a straight line, and then calculate the memory according to the edge outline. The initial position of the bar.

Preferably, 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.

Preferably, acquiring the grasping reference position of the robotic arm specifically includes:

Drive the mobile camera to move to a photographing reference position where the image of the memory stick can be obtained by the mechanical arm;

Obtaining the image of the memory stick by the mobile camera and calculating the first coordinate reference position of the memory stick;

The memory stick is grabbed by the mechanical arm and vertically raised by a preset distance to a grabbing reference position.

Preferably, 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.

Preferably, acquiring the insertion reference position of the memory slot specifically includes:

Drive the mobile camera to move to a recognition reference position where the image of the memory slot can be obtained by the mechanical arm;

Obtain an image of the memory slot by the mobile camera and calculate the second coordinate reference position of the memory slot;

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:

memory for storing computer programs;

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 provided by the present invention 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. In 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. In the fourth step, in order to clarify the position of the current memory slot, 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. In the fifth step, 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. In this way, the machine vision-based memory alignment and plugging method provided by the present invention 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.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without creative work.

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.

Among them, in Figure 2:

Initial acquisition module-1, memory grab module-2, attitude adjustment module-3, slot positioning module-4, alignment insertion module-5.

detailed description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Please refer to FIG. 1 . FIG. 1 is a schematic diagram of the overall structure of a specific embodiment provided by the present invention.

In 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:

S1. 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 module through image processing technology;

S2. 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 grasping position to grab the memory stick according to the offset;

S3. 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 move the manipulator to the position to be assembled according to the deviation between the post-grab position and the assembly reference position;

S4. Acquire an 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;

S5. Calculate the current insertion position according to the position to be assembled and the compensation position, and move the mechanical arm to the current insertion position to perform the memory module insertion operation.

Among them, in 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).

In 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.

In 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.

In 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.

In 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.

In this way, in the machine vision-based memory alignment method provided in this embodiment, 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. Position compensation to keep the memory module aligned with the memory slot before insertion. Therefore, the present embodiment 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.

In addition, considering that the coordinates of the robotic arm can be accurately positioned according to the spatial coordinate system constructed by its driving mechanism, and the coordinate positioning method of the mobile camera and the fixed camera for the memory stick can only be obtained by taking pictures and performing image processing technology to calculate, so, In order to facilitate the mobile camera and the fixed camera to determine the coordinate position of the memory module through image acquisition, in this embodiment, 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.

Specifically, 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. Let the coordinates of the manipulator be (x, y, z), because the manipulator can keep the Z axis unchanged when it moves, so it only needs to calibrate x and y. That is, the relationship between the image coordinates and the robot coordinates is:

Among them, the rotation matrix T is:

The translation matrix M is:

The system of equations can thus be obtained:

It can be seen from the equation system that there are six unknowns, and at least three sets of points are required to solve these six sets of numbers. However, in order to ensure the accuracy of the numerical value, more points are generally selected to use the least squares method to solve a higher precision solution. Refer to The accuracy of the robot arm and the pixel accuracy can be guaranteed to meet the requirements of the obtained matrix when there are nine sets of points, so nine sets of points are generally used to solve the rotation and translation matrices. In order to distinguish the mobile camera from the fixed camera conveniently, the coordinate transformation matrix of the mobile camera can be set as Hs, and the coordinate transformation matrix of the fixed camera can be set as Hx.

In step S1, the mobile camera is used to obtain the initial position of the memory module through image processing technology, which specifically includes:

First, 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. Next, just according to the edge contour of the memory stick, 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.

In addition, in order to improve the accuracy of the grasping reference position of the robotic arm, the assembling reference position of the memory module, and the insertion reference position of the memory slot, in this embodiment, before acquiring the initial position of the memory module, the Each reference position is corrected and determined before operation according to the actual equipment and environment.

Specifically, the method for obtaining the grasping reference position of the robotic arm and the assembly reference position of the memory module specifically includes:

First, drive the robotic arm to move the mobile camera to the preset first photographing position to take pictures. At this time, save the coordinates p6 (x6, y6, q6) of the robotic arm and the coordinates p7 (x7, y7) of the memory stick in the image. q7), and then manually move the robotic arm to drive the mobile camera to grab the memory stick. After grabbing the memory stick, only change the Z coordinate to the photographing height, and record the coordinate p8 (x8, y8, q8) of the robotic arm at this time for the needs of the robotic arm. The base position of the grab. Then 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. Among them, q is the attitude angle.

The method of obtaining the insertion reference position of the memory slot specifically includes:

Use the robotic arm to guide the mobile camera to move the preset distance above the memory slot, and slowly guide the robotic arm to insert the memory stick into the memory slot. After inserting the memory slot, raise it to the height of the mobile camera to ensure that The x and y coordinates of the robotic arm remain unchanged, and the coordinates of the robotic arm at this time are recorded as p1 (x1, y1, q1), and then move the robotic arm to the field of view of the mobile camera to clearly capture the location of the memory slot, and find the location at this time. 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. Wherein, 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, and p3 is the second coordinate reference position when the memory slot is identified.

In 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:

Similarly, in 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):

In addition, in 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):

In step S4, the calculation method of the compensation position is to combine the coordinates p12 of the to-be-assembled position of the memory stick with the coordinates of the compensation position p5, that is, p13(x13, y13, q13)=p13(x12+x5, y12+ y5, q12+q5).

As shown in FIG. 2 , 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 .

Wherein, 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. Wherein, the memory is mainly used to store the computer program, and 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.

In this embodiment, 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.

In this embodiment, 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.

The above description of the disclosed embodiments enables any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

A machine vision-based memory alignment method, characterized in that, comprising:

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;

Calculate the offset of the robotic arm according to the deviation between the initial position state of the memory stick and the grab reference position, and move the robotic arm to the grasping position to grab the memory stick according to the offset ;

Move the mechanical 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 mechanical arm according to the deviation between the post-grab position and the assembly reference position to the position to be assembled;

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 for aligning and inserting memory based on machine vision according to claim 1, characterized in that, before driving the mobile camera to move synchronously to the initial position by the mechanical arm, further comprising:

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.
The method for aligning and inserting memory based on machine vision according to claim 1, characterized in that, enabling the mobile camera to obtain the initial position of the memory module through image processing technology, specifically comprising:

Make the mobile camera obtain the outline of the memory bank in the preset ROI area through a connected domain analysis method, and fit the edge contour of the memory bank through a straight line, and then calculate the memory according to the edge outline. The initial position of the bar.
The method for aligning and inserting memory based on machine vision according to claim 1, characterized in that, before acquiring the initial position state of the memory stick, further comprising:

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.
The method for aligning and inserting memory based on machine vision according to claim 4, wherein acquiring the grasping reference position of the robotic arm specifically includes:

Drive the mobile camera to move to a photographing reference position where the image of the memory stick can be obtained by the mechanical arm;

Obtaining the image of the memory stick by the mobile camera and calculating the first coordinate reference position of the memory stick;

The memory stick is grabbed by the mechanical arm and vertically raised by a preset distance to a grabbing reference position.
The method for aligning and inserting memory based on machine vision according to claim 5, wherein 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.
The method for aligning and inserting memory based on machine vision according to claim 6, characterized in that, obtaining the insertion reference position of the memory slot, specifically comprising:

Drive the mobile camera to move to a recognition reference position where the image of the memory slot can be obtained by the mechanical arm;

Obtain an image of the memory slot by the mobile camera and calculate the second coordinate reference position of the memory slot;

The mobile camera is driven by the mechanical arm to move to an insertion reference position at a preset distance above the memory slot.
A machine vision-based memory alignment plugging system, characterized in that it includes:

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 module 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.
A device, characterized in that it includes:

memory for storing computer programs;

The processor is configured to implement the steps of the machine vision-based memory alignment method according to any one of claims 1 to 7 when executing the computer program.
A storage medium, characterized in that, a computer program is stored on the storage medium, and when the computer program is executed by a processor, the computer program according to any one of claims 1 to 7 is implemented based on the machine vision memory alignment method. step.