Display screen slitter edge removing device and method based on visual positioning detection algorithm

Abstract

The invention is applied to the field of liquid crystal display panel production, and relates to a display screen slitter edge removing device and method based on a visual positioning detection algorithm, which can effectively remove glass slitter edges. The waste edge removing device for the display screen comprises a rack, a main control unit, a Y-direction transmission mechanism, a waste box, an edge trimming strip Y, an edge trimming strip X, a large-size area light source, a detection camera, an edge trimming adsorption platen, a DD motor and a drag chain, the device is matched with a large-view visual positioning scheme and is combined with a servo control system according to a visual positioning algorithm, and the positioning precision can reach 0.1 mm. The position coordinates of the glass are calculated visually, the position and angle of the edge trimming strip are adjusted in real time, the slitter edge with the minimum size of 1 mm can be effectively removed, the slitter edge removal rate can reach 99%, and the situation that the slitter edge flows into downstream equipment, so that a glass terminal is scratched, and in-plane crushing is caused is effectively eradicated. The design can effectively remove glass slitter edges, improve the product yield and reduce the downtime rate of equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of liquid crystal display panel production, and particularly to a display screen waste edge removing device and method based on visual positioning detection algorithm. BACKGROUND

[0002] In recent years, with the continuous development of the vehicle display industry, the demand for liquid crystal display panels is increasing. In order to improve the utilization rate of the substrate, multiple liquid crystal panels are usually integrated on a large glass substrate. There are glass residues distributed between each panel or around the substrate. Large-size glass substrates are separated into small-size panels through cutting and cracking processes. During the unloading and sorting, due to static electricity and residual adhesive (such as frame glue), the separated single glass particle will usually be adhered to the adjacent waste edge. If the adhered waste edge is not effectively removed, the waste edge will flow to the downstream equipment (such as a cleaning machine and a detection table) with the liquid crystal screen in the automatic production line. During the mechanical conveying or processing, the size does not meet the requirements due to the carrying of the waste edge, so that the positioning mechanism or the edge grinding mechanism collides with the product, which is easy to cause scratches at the terminal, and even the risk of product fragmentation, ultimately leading to product scrap and equipment downtime.

[0003] At present, the removal of the above-mentioned adhered waste edge in the industry mainly relies on manual operation: the operator manually peels off the waste edge by using tweezers or a scraper, and then checks and confirms the removal effect by visual inspection. However, this method has the following defects: first, the manual operation force is difficult to accurately control, which is easy to cause glass edge damage due to excessive force, or to cause waste edge residue due to insufficient force; second, visual inspection is affected by the experience and fatigue of the operator, and the identification rate of small residual waste edge is low, which leads to unqualified products flowing into the downstream; third, the manual processing efficiency is low, which is difficult to match the pace of the automatic production line; fourth, long-term manual contact with the glass substrate is easy to cause secondary pollution due to hand stains and static electricity, which further increases the scrap rate. SUMMARY

[0004] After the existing display screen sorting mechanism picks up the edge, there may be residual waste edges. The machine can only discharge the products with residual waste edges as NG products from the machine. After the waste edges are removed by manual operation, the products are put into the machine again, which leads to frequent alarms of the equipment, low productivity, and frequent manual intervention. To solve the above technical problems, the present application provides a display screen waste edge removing device and method based on visual positioning detection algorithm.

[0005] This invention is achieved using the following technical solution: a display screen edge removal device based on a visual positioning detection algorithm, comprising a frame and a main control unit; a Y-axis transmission mechanism, an edge-removing strip Y, an edge-removing strip X, and a vision system are mounted on the frame; an adsorption device is mounted on the slider of the Y-axis transmission mechanism; the edge-removing strip X can move along an X-axis guide rail mounted on the frame, the X-axis guide rail being arranged near one end of the Y-axis transmission mechanism, and the edge-removing strip Y being positioned opposite one end of the Y-axis transmission mechanism; the edge removal device has a pre-set edge-removing station positioned by the edge-removing strip Y and the edge-removing strip X; the vision system is located above the edge-removing station, and the adsorption device adsorbs the display screen through negative pressure, and can rotate in the horizontal plane and move along the Y-axis to the edge-removing station under the drive of a configured motor; the Y-axis transmission mechanism, the edge-removing strip Y, the edge-removing strip X, the vision system, and the adsorption device are all controlled by the main control unit, and the images collected by the vision system are transmitted to the main control unit for detection and analysis.

[0006] Furthermore, the adsorption device includes a horizontally positioned edge-adsorption platform rotatably mounted on a slider of a Y-axis transmission mechanism. The upper surface of the edge-adsorption platform has an installation groove, and a negative pressure adsorption channel is integrated inside. Several adsorption holes are distributed on the surface of the installation groove. The negative pressure adsorption channel is connected to an external negative pressure device through a negative pressure pipeline.

[0007] Furthermore, the slider of the Y-axis transmission mechanism is equipped with a DD motor controlled by the main control unit via a bearing seat. The rotor axis of the DD motor is vertical and is fixedly connected to the bottom of the edge-attaching adsorption platform.

[0008] Furthermore, the vision system includes a detection camera and a large surface light source surrounding the detection camera; the large surface light source is controlled by a main control unit, and the detection camera communicates bidirectionally with the main control unit.

[0009] Furthermore, a cable chain is installed on the side of the frame; the power cables for the Y-axis transmission mechanism, the Y-axis edge trimming strip, the X-axis edge trimming strip, the X-axis guide rail, the DD motor, the signal cable for the detection camera, and the negative pressure pipeline for the edge trimming adsorption platform are all located inside the cable chain; the drive motors for the Y-axis transmission mechanism and the X-axis guide rail are both servo motors and are used in conjunction with encoders to achieve closed-loop position control.

[0010] Furthermore, it also includes a waste box located directly below the edge-trimming station, with the opening size of the waste box being larger than the coverage area of ​​the edge-trimming station.

[0011] This invention also discloses a method for removing unwanted edges from a display screen based on a visual positioning detection algorithm, comprising the following steps:

[0012] 1) Place the glass workpiece to be edged on the adsorption device, and the main control unit controls the adsorption device to adsorb and fix the glass workpiece;

[0013] 2) Then the workstation transfer stage begins. The adsorption device moves along the Y-axis transmission mechanism and rotates horizontally at the same time to move the glass workpiece to the edge-trimming station.

[0014] 3) Next, the precise positioning stage begins. The vision system acquires real-time images of the glass workpiece and transmits them to the main control unit. The main control unit calibrates the position of the glass workpiece using a vision positioning detection algorithm. If there is a deviation, the rotation angle of the adsorption device and the slider position of the Y-axis transmission mechanism are finely adjusted to ensure that the edge of the glass workpiece is precisely aligned with the edge strip.

[0015] 4) Then the edge trimming stage begins. The main control unit controls the edge trimming strip's movement and adjusts its spatial position according to the glass workpiece's size parameters to remove residual waste edges. The edge trimming accuracy is controlled within ±0.5mm.

[0016] 5) After the edge trimming is completed, the waste materials are recycled;

[0017] 6) Finally, the resetting and unloading stage begins. The edge strip is reset, the adsorption device releases negative pressure and resets, completing a single edge stripping cycle.

[0018] Furthermore, the visual positioning detection algorithm in step 3) includes the following steps:

[0019] Edge detection:

[0020] This step uses the Canny operator to extract edges. First, the image of the glass workpiece acquired by the vision system is Gaussian smoothed to filter out noise interference from reflections and minor stains on the glass surface, while preserving edge information. Then, discrete feature points above the edge threshold are extracted.

[0021] b. Linear fitting:

[0022] The least squares method is used for fitting; the glass edge is a straight line, and the equation of the standard edge straight line is given by... The actual edge is obtained by edge detection, which yields N feature points. Fit the actual straight line using the least squares method. Minimize the sum of squared errors The formula is as follows

[0023]

[0024]

[0025]

[0026] in: For feature points Sum of coordinates For feature points Sum of coordinates for and Sum of products For feature points Sum of squared coordinates; the coordinate system described in steps a and b above has its origin at the top left corner of the glass workpiece image;

[0027] c-deviation calculation:

[0028] The following formula quantifies the positional deviation, providing data support for the precise movement of the edge trimmer and ensuring that the positioning accuracy between the edge trimmer and the edge of the glass workpiece is controlled within ±0.5mm. Specific deviation calculations include the quantification of overall deviation, translational deviation, and rotational deviation. The formulas and parameters are explained below:

[0029]

[0030] formula This is used to quantify the overall offset between the actual position of the glass workpiece and a pre-set standard glass workpiece, whereby... , These represent the total deviations in the X and Y axes, respectively. , These represent the X-axis and Y-axis deviations caused by rotational motion, respectively. , These are the X-axis and Y-axis deviations caused by translational motion. By superimposing the rotational deviation and the translational deviation, a comprehensive basis is provided for the X-axis and Y-axis compensation action of the edge strip, ensuring that the positioning accuracy of the edge strip and the glass edge is controlled within the preset range.

[0031]

[0032] formula Used to reflect the parallel movement and offset of the glass workpiece in the X and Y axis directions, where , These are the actual X-axis and Y-axis coordinates of a vertex of a glass workpiece obtained through vision system detection. , The ideal positioning coordinates of the preset standard glass workpiece calibration point in the X and Y axes are calculated. The difference between the actual coordinates and the ideal positioning coordinates is used to quantify the parallel movement offset. This provides data support for the translation compensation of the drive motor of the Y-axis transmission mechanism 1 and the X-axis guide rail, ensuring that the horizontal distance between the edge strip and the edge of the glass workpiece is consistent.

[0033]

[0034] formula Used to quantify the positional shift of a glass workpiece caused by tilting around its rotation center, wherein This refers to the actual rotation angle of the glass workpiece. , Let X and Y be the coordinates of the center of rotation of the adsorption device, and let the matrix be... The plane rotation transformation matrix describes the effect of the rotation angle on the coordinates of the glass workpiece. The rotation deviation calculated by this formula can guide the angle compensation of the rotating shaft of the adsorption device, ensuring that the edge of the glass workpiece is consistent with the direction of the edge strip and improving the edge stripping accuracy. The coordinate system in step c above takes the starting point of the actual mechanical coordinate system of the entire device as the origin.

[0035] Furthermore, in step 4), there are two edge-cutting methods for the edge-cutting strip: from top to bottom or from bottom to top. If the cross-section of the glass workpiece is a flat cross-section, either method is acceptable. If the cross-section of the glass workpiece is a normal stepped cross-section, then edge-cutting from top to bottom is used. If the cross-section of the glass workpiece is an inverted stepped cross-section, then edge-cutting from bottom to top is used.

[0036] The beneficial effects of this invention compared to the prior art are as follows:

[0037] 1. By combining large field-of-view vision algorithms, it can identify residual waste edges as small as 0.1mm and perform precise positioning with an accuracy of up to 0.5mm.

[0038] 2. An integrated X / Y / R edge-forming system is built, which can realize automatic compensation in the X / Y / R directions, ensure that the edge-forming strip is parallel to the waste edge, and realize edge-forming in both directions from top to bottom and from bottom to top to deal with different stepped edges.

[0039] 3. Compared with the existing method of manually removing waste edges, this method improves equipment productivity because it eliminates the need for equipment downtime. Attached Figure Description

[0040] Figure 1 A schematic diagram of the main structure of the waste edge removal device of the present invention;

[0041] Figure 2 A top view of the waste edge removal device of the present invention;

[0042] Figure 3 The flowchart for removing unwanted edges as described in this invention;

[0043] Figure 4 Schematic diagram of three cross-sectional views of the display screen;

[0044] Figure 5 Schematic diagram of the waste edge removal mechanism;

[0045] Figure 6Schematic diagrams of two methods for removing waste edges.

[0046] 1-Y-axis transmission mechanism, 2-waste box, 3-edge trimming strip Y, 4-edge trimming strip X, 5-large-size surface light source, 6-detection camera, 7-edge trimming adsorption platform, 8-DD motor, 9-drag chain. Detailed Implementation

[0047] This invention combines wide-field vision technology and a servo control system to achieve fully automated waste edge detection. It can detect residual waste edges at terminals with a detection accuracy of up to 0.1mm. It accurately calculates the edge-cutting position and calibrates positional deviations in real time, achieving a positioning accuracy of up to 0.5mm. It can precisely remove waste edges as small as 1mm, with a waste edge removal rate of up to 99%. This effectively prevents waste edges from flowing into downstream equipment and causing scratches and in-plane pressure damage to glass terminals, reducing equipment downtime, improving production efficiency and product yield, and achieving full automation of the machine.

[0048] The present invention will be further described below with reference to the accompanying drawings.

[0049] This invention provides an automated display screen edge removal device (such as...) Figure 1 Its core lies in achieving high-precision edge trimming and standardized replication of glass of different specifications through modular design and precise transmission mechanism.

[0050] (1) Mechanical structure

[0051] The waste edge removal device for the display screen includes a Y-axis transmission mechanism 1, a waste box 2, edge-removing strips Y3 and X4, a large-size surface light source 5, a detection camera 6, an edge-removing adsorption platform 7, a DD motor 8, and a cable chain 9. All components are supported by a frame, forming a collaborative overall structure. The edge-removing adsorption platform 1 is horizontally positioned on the slider of the Y-axis transmission mechanism 1. The upper surface of the edge-removing adsorption platform 7 has a suitable mounting groove, which integrates a negative pressure adsorption channel. The adsorption surface of the mounting groove has several adsorption holes for negative pressure adsorption and fixation of the glass workpiece placed on the edge-removing adsorption platform 7, preventing displacement of the glass during the edge-removing process. The DD motor 8 is vertically mounted in the middle of the slider via a bearing seat, with its upper end fixedly connected to the bottom of the edge-removing adsorption platform 8. This allows for precise 270-degree rotation and positioning of the rotating shaft, thereby driving the edge-removing adsorption platform 7 and the adsorbed and fixed glass to move along the Y-axis to the preset edge-removing position.

[0052] The Y-axis transmission mechanism 1 is laid on the frame (e.g., Figure 2The bidirectional waste trimming strips are symmetrically arranged on both sides of the trimming station. The trimming strip X4 is installed on the X-axis guide rail, which is arranged perpendicularly to the Y-axis transmission mechanism 1. The trimming strip X4 is driven by a linear module to translate along the X-axis. The end of the trimming strip Y is installed opposite the Y-axis transmission mechanism 1. Through the cooperation of the Y-axis transmission mechanism 1 and the X-axis guide rail, the trimming strip can be flexibly adjusted in three-dimensional space to adapt to various sizes of trimming adsorption tables 7 and glass workpieces.

[0053] A large-size surface light source 5 is fixedly installed on the top of the frame via a bracket and located directly above the edging station. The large-size surface light source 5 surrounds the lens of the inspection camera 6, providing a uniform, shadow-free lighting environment for glass positioning. The inspection camera 6's field of view covers the entire edging station and is connected to the main control unit of the equipment. By acquiring real-time images of the glass and transmitting them to the main control unit, it achieves real-time position detection and positioning calibration of the glass at the edging station, ensuring that the glass is accurately positioned at the preset edging reference position. A waste box 2 is located directly below the edging station, with an opening larger than the coverage area of ​​the edging station, used to collect glass waste material detached during the edging process, enabling centralized waste recycling. A cable chain 9 is installed on the side of the frame and connected to the Y-axis transmission mechanism 1. The cable chain 9 internally houses the power cable of the drive motor, the signal cable of the inspection camera 6, and the negative pressure pipeline of the edging adsorption platform 7. During the movement of the edging strip, it guides and protects the cables and pipelines, preventing entanglement or wear.

[0054] All of the above devices are electrically connected to the main control unit of the equipment. The control signals output by the main control unit enable the coordinated action of each component. The drive motors are all servo motors and are used in conjunction with encoders to achieve closed-loop position control, ensuring the positioning accuracy of the edge strip and glass movement.

[0055] (2) Principles of visual algorithms

[0056] The flowchart for removing waste edges is as follows: Figure 3 As shown, after receiving the product to be edged (the glass display screen after preliminary polishing), the edge-trimming table moves directly below the inspection camera 6 to acquire an image. After taking a picture, the inspection camera 6 performs edge detection, linear fitting, position correction, and deviation calculation on the image. It calculates the offset between the current product position and the reference calibration position and sends this calculation to the main control unit. Upon receiving the offset from the vision system, the main control unit compensates for it in the vectors of each direction of the edge-trimming position and recalculates the adjusted edge-trimming position. The edge-trimming strip is positioned according to the calculated position, ensuring that the pressure knife and the waste edge are horizontal and preventing accidental damage to the product.

[0057] Edge detection:

[0058] This step uses the Canny operator for edge extraction. First, the image is Gaussian smoothed to filter out noise interference such as glass surface reflections and minor stains, while preserving edge information. Next, the gradient operator is used to calculate the gradient magnitude and direction of each pixel in the image to initially locate the approximate edge position. Then, a dual threshold is set for edge filtering: strong edges above the higher threshold are retained, and weak edges that are strong edges and above the lower threshold are connected. Noise edges below the lower threshold are removed, and the coordinates of N feature points of the actual edge are output. ( (This ensures fitting accuracy). This process adapts to the complex interference environment of the glass surface, and the contour formed by the output edge feature points is continuous and unbroken, providing accurate discrete point input for subsequent linear fitting steps.

[0059] Linear fitting:

[0060] The least squares method is used for fitting. The glass edge is a straight line; let the equation of the standard edge line be... (Given equipment calibration parameters), the actual edge is obtained by edge detection using N feature points. Fit the actual straight line using the least squares method. Minimize the sum of squared errors The formula is as follows:

[0061]

[0062]

[0063]

[0064] in: For feature points Sum of coordinates for Sum of coordinates for and Sum of products for Sum of squared coordinates.

[0065] Deviation calculation:

[0066] The following formula quantifies the positional deviation, providing data support for the precise movement of the edge-forming mechanism and ensuring that the positioning accuracy of the edge-forming strip and the glass edge is controlled within ±0.5mm. Specific deviation calculations include the quantification of overall deviation, translational deviation, and rotational deviation. The formulas and parameters are explained below:

[0067]

[0068] This formula is used to quantify the combined offset between the actual position and the standard positioning position of the glass workpiece, where , These represent the total deviations in the X and Y axes, respectively. , These represent the X-axis and Y-axis deviations caused by rotational motion, respectively. , The X-axis and Y-axis deviations caused by translational motion are respectively. By superimposing the rotational deviation and the translational deviation, a comprehensive basis is provided for the X-axis and Y-axis compensation action of the edge strip mechanism, ensuring that the positioning accuracy of the edge strip and the glass edge is controlled within the preset range.

[0069]

[0070] This formula reflects the parallel displacement of the glass workpiece in the X and Y axis directions, where , These are the actual X-axis and Y-axis coordinates of the glass workpiece obtained through vision system detection. , The ideal positioning coordinates (unit: mm, stored in the main control unit through equipment calibration) of the preset standard glass workpiece (without waste edges) in the X and Y axes are calculated. The difference between the actual coordinates and the ideal positioning coordinates is calculated to quantify the parallel movement offset, providing data support for the translation compensation of the Y-axis transmission mechanism 1 and the X-axis drive motor, ensuring that the horizontal distance between the edge strip and the glass edge is consistent.

[0071] ;

[0072] This formula is used to quantify the positional shift of a glass workpiece due to tilting (rotation about a center of rotation), where The actual rotation angle of the glass workpiece (unit: radians, calculated from the difference between the slope of the glass edge detected by the vision system and the standard slope). , The matrix represents the X-axis and Y-axis coordinates of the equipment's rotation center (unit: mm, i.e., the axis center coordinates of the DD motor's rotation shaft, determined through equipment calibration). This is a planar rotation transformation matrix used to describe the effect of the rotation angle on the glass coordinates. The rotation deviation calculated by this formula can guide the angle compensation of the DD motor 8 rotation axis, ensuring that the glass edge is consistent with the direction of the edge strip and improving the edge cutting accuracy.

[0073] If the overall deviation is less than the maximum permissible error, then edge patching is performed; if it is greater than the maximum permissible error, then the process returns to the edge detection step.

[0074] (3) Side-hitting action

[0075] Due to different manufacturing processes, the A and B sides of the product placed at the edge-cutting station face different directions. During the removal process, because the dimensions of the A and B sides of the product are different, three different cross-sectional situations will occur, including a flush cross-section as shown below. Figure 4 (a) A normal step cross-section is as follows Figure 4 (b) The inverted step cross section is as follows Figure 4 (c) For flush cross sections, the effects of top and bottom edge beating are the same, so only the two cases of stepped cross sections are considered.

[0076] The diagram of removing waste edges is shown below. Figure 5 As shown. After the edge-attaching adsorption platform moves to the adjusted edge-attaching position, the pressure knife structure moves to the edge-attaching preparation position to perform the edge-attaching action. After edge-attaching is completed, the pressure knife structure returns to the safe position, and the edge-attaching is completed. For normal stepped cross-sections, the top edge-attaching method is generally used, that is, edge-attaching from top to bottom (e.g. Figure 6 B), the inverted stepped cross-section is the opposite (as shown in B). Figure 6 A) Different edge-forming methods for different cross-sections can reduce the extrusion damage to the cross-section caused by rotation and improve product yield.

[0077] The working process of this automated glass edging equipment is as follows: First, in the feeding stage, the glass display screen workpiece to be edged is placed on the edging adsorption platform. The main control unit controls the edging adsorption platform 7 to start negative pressure adsorption, achieving stable fixation of the glass. Next, in the station transfer stage, the DD motor 8 drives the rotating shaft to rotate, moving the edging adsorption platform 7 and the glass along the Y-axis to the edging station. During the movement, the encoder provides position feedback signals to complete the initial positioning. Then, in the precision positioning stage, the large-size surface light source 5 is activated, providing illumination. The detection camera 6 captures real-time images of the glass and transmits them to the main control unit. The main control unit calibrates the glass position using image algorithms. If there is a deviation, the motor and the DD motor 8 rotating shaft are finely adjusted to ensure precise alignment of the glass edging edge with the waste edging strip. Finally, the edging operation begins. In the first stage, the main control unit drives the motor to control the edge-removing strip action based on the glass size parameters, adjusts the spatial position of the edge-removing strip, and the pressure knife extends from top to bottom / bottom to top (selectable) to remove the residual edge, with the edge-removing accuracy controlled within ±0.5mm; after the edge removal is completed, the waste material recycling stage begins, and the peeled glass waste falls into the waste box 2 below under the action of gravity, completing the centralized recycling of waste materials; finally, the resetting and unloading stage begins, the edge-removing strip is reset, the waste edge vacuum platform releases negative pressure, the DD shaft and the waste edge vacuum platform are reset, completing a single edge-removing cycle.

[0078] Technical features of the present invention:

[0079] 1. Anti-interference visual positioning system: The Canny operator is used to achieve noise-resistant edge detection. The deviation between the actual and standard edges is quantified by least squares fitting. Then, the deviation is calibrated through closed loop, which solves the defects of traditional visual detection such as inaccuracy, missed detection of small waste edges, and lack of closed loop calibration.

[0080] 2. X / Y / R Fully Automatic Integrated Edge Trimming and Waste Removal Mechanism: The deviation calculated by vision is compensated in real time to the X / Y axis and DD axis. Different edge trimming methods are adopted for different cross sections to reduce glass damage. It can also accurately remove 0.5mm tiny waste edges. It can perform edge trimming in two ways: top to bottom and bottom to top. It solves the problems of traditional edge trimming position offset, easy product damage, and incomplete waste edge removal.

[0081] The error value obtained in the deviation calculation step needs to be converted into the movement value of the slider on the X guide rail and the Y transmission mechanism and the rotation angle value of the DD motor. The coordinate system of each of the above-mentioned moving parts is pre-calibrated, and the transformation matrix between it and the actual mechanical coordinate system of the whole device is also obtained in advance. Therefore, after knowing the error value, it is easy to calculate the motion value that each moving part needs to be adjusted.

Claims

1. A display screen edge removal device based on a visual positioning detection algorithm, comprising a frame and a main control unit; characterized in that, The frame is equipped with a Y-axis transmission mechanism (1), edge trimming strip Y (3), edge trimming strip X (4) and a vision system; an adsorption device is installed on the slider of the Y-axis transmission mechanism (1); the edge trimming strip X (4) can move along the X-axis guide rail installed on the frame, the X-axis guide rail is arranged near one end of the Y-axis transmission mechanism (1), and the edge trimming strip Y (3) is set opposite one end of the Y-axis transmission mechanism (1); the waste edge removal device is preset with an edge trimming station positioned by the edge trimming strip Y (3) and the edge trimming strip X (4); the vision system is located above the edge trimming station, and the adsorption device can rotate in the horizontal plane and move along the Y-axis to the edge trimming station under the drive of the configured motor through the negative pressure adsorption display screen; the Y-axis transmission mechanism (1), edge trimming strip Y (3), edge trimming strip X (4), vision system and adsorption device are all controlled by the main control unit, and the images collected by the vision system are transmitted to the main control unit for detection and analysis.

2. The display screen edge removal device based on a visual positioning detection algorithm as described in claim 1, characterized in that, The adsorption device includes a horizontal edge adsorption platform (7) that is rotatably mounted on a slider of the Y-axis transmission mechanism (1). The upper surface of the edge adsorption platform (7) is provided with an installation groove, and a negative pressure adsorption channel is integrated inside. Several adsorption holes are distributed on the surface of the installation groove. The negative pressure adsorption channel is connected to an external negative pressure device through a negative pressure pipeline.

3. The display screen edge removal device based on a visual positioning detection algorithm as described in claim 2, characterized in that, The slider of the Y-axis transmission mechanism (1) is equipped with a DD motor (8) controlled by the main control unit through a bearing seat. The rotor axis of the DD motor (8) is vertical and is fixedly connected to the bottom of the edge adsorption platform (7).

4. The display screen edge removal device based on a visual positioning detection algorithm as described in claim 3, characterized in that, The vision system includes a detection camera (6) and a large surface light source (5) surrounding the detection camera (6); the large surface light source (5) is controlled by a main control unit, and the detection camera (6) communicates bidirectionally with the main control unit.

5. The display screen edge removal device based on a visual positioning detection algorithm as described in claim 4, characterized in that, The side of the frame is equipped with a drag chain (9); the Y-axis transmission mechanism (1), the edge strip Y (3), the edge strip X (4), the X-axis guide rail, the power cable of the DD motor (8), the signal cable of the detection camera (6) and the negative pressure pipeline of the edge adsorption platform (7) are all located inside the drag chain (9); the drive motors of the Y-axis transmission mechanism (1) and the X-axis guide rail are all servo motors and are used in conjunction with encoders to achieve closed-loop position control.

6. The display screen edge removal device based on a visual positioning detection algorithm as described in claim 5, characterized in that, It also includes a waste box (2) located directly below the edge-trimming station, with the opening size of the waste box (2) being larger than the coverage area of ​​the edge-trimming station.

7. A method for removing unwanted edges from a display screen based on a visual positioning detection algorithm, implemented using the display screen unwanted edge removal device based on a visual positioning detection algorithm as described in any one of claims 1-6, characterized in that, Includes the following steps: 1) Place the glass workpiece to be edged on the adsorption device, and the main control unit controls the adsorption device to adsorb and fix the glass workpiece; 2) Then the workstation transfer stage begins. The adsorption device moves along the Y-direction transmission mechanism (1) and rotates horizontally to move the glass workpiece to the edge-trimming station. 3) Next, the precise positioning stage begins. The vision system acquires real-time images of the glass workpiece and transmits them to the main control unit. The main control unit calibrates the position of the glass workpiece using a visual positioning detection algorithm. If there is a deviation, the rotation angle of the adsorption device and the slider position of the Y-axis transmission mechanism (1) are finely adjusted to ensure that the edge of the glass workpiece is precisely aligned with the edge strip. 4) Then the edge trimming stage begins. The main control unit controls the edge trimming strip's movement and adjusts its spatial position according to the glass workpiece's size parameters to remove residual waste edges. The edge trimming accuracy is controlled within ±0.5mm. 5) After edge trimming, the waste materials are recycled; 6) Finally, the resetting and unloading stage begins. The edge strip is reset, the adsorption device releases negative pressure and resets, completing a single edge stripping cycle.

8. The method for removing unwanted edges from a display screen based on a visual positioning detection algorithm as described in claim 7, characterized in that, The visual positioning detection algorithm in step 3) includes the following steps: Edge detection: This step uses the Canny operator to extract edges. First, the image of the glass workpiece acquired by the vision system is Gaussian smoothed to filter out noise interference from reflections and minor stains on the glass surface, while preserving edge information. Then, discrete feature points above the edge threshold are extracted. b. Linear fitting: The least squares method is used for fitting; the glass edge is a straight line, and the equation of the standard edge straight line is given by... The actual edge is obtained by edge detection, which yields N feature points. Fit the actual straight line using the least squares method. Minimize the sum of squared errors The formula is as follows ； ； ； in: For feature points Sum of coordinates For feature points Sum of coordinates for and Sum of products For feature points Sum of squared coordinates; the coordinate system described in steps a and b above has its origin at the top left corner of the glass workpiece image; c-deviation calculation: The following formula quantifies the positional deviation, providing data support for the precise movement of the edge trimmer and ensuring that the positioning accuracy between the edge trimmer and the edge of the glass workpiece is controlled within ±0.5mm. Specific deviation calculations include the quantification of overall deviation, translational deviation, and rotational deviation. The formulas and parameters are explained below: ； formula This is used to quantify the overall offset between the actual position of the glass workpiece and a pre-set standard glass workpiece, whereby... , These represent the total deviations in the X and Y axes, respectively. , These represent the X-axis and Y-axis deviations caused by rotational motion, respectively. , These are the X-axis and Y-axis deviations caused by translational motion. By superimposing the rotational deviation and the translational deviation, a comprehensive basis is provided for the X-axis and Y-axis compensation action of the edge strip, ensuring that the positioning accuracy of the edge strip and the glass edge is controlled within the preset range. ； formula Used to reflect the parallel movement and offset of the glass workpiece in the X and Y axis directions, where , These are the actual X-axis and Y-axis coordinates of a vertex of a glass workpiece obtained through vision system detection. , The ideal positioning coordinates of the calibration point of the preset standard glass workpiece in the X and Y axes are calculated. The difference between the actual coordinates and the ideal positioning coordinates is calculated to quantify the parallel movement offset, which provides data support for the translation compensation of the drive motor of the Y-axis transmission mechanism (1) and the X-axis guide rail, and ensures that the horizontal distance between the edge strip and the edge of the glass workpiece is consistent. ； formula Used to quantify the positional shift of a glass workpiece caused by tilting around its rotation center, wherein This refers to the actual rotation angle of the glass workpiece. , Let X and Y be the coordinates of the center of rotation of the adsorption device, and let the matrix be... The plane rotation transformation matrix describes the effect of the rotation angle on the coordinates of the glass workpiece. The rotation deviation calculated by this formula can guide the angle compensation of the rotating shaft of the adsorption device, ensuring that the edge of the glass workpiece is consistent with the direction of the edge strip and improving the edge stripping accuracy. The coordinate system in step c above takes the starting point of the actual mechanical coordinate system of the entire device as the origin.

9. The method for removing unwanted edges from a display screen based on a visual positioning detection algorithm as described in claim 7, characterized in that, In step 4), there are two edge-cutting methods for the edge-cutting strip: top to bottom or bottom to top. If the cross-section of the glass workpiece is a flat cross-section, either method is acceptable. If the cross-section of the glass workpiece is a normal stepped cross-section, then top to bottom edge-cutting is used. If the cross-section of the glass workpiece is an inverted stepped cross-section, then bottom to top edge-cutting is used.

Images