Plate part electrophoresis production line automatic feeding and discharging method and device based on machine vision

Abstract

The invention relates to an automatic feeding and discharging method for a plate part electrophoresis production line, which comprises the following steps: firstly, positioning a hanger to a station table through a slide rail, and placing plate parts in a special stacking bin; images and depth information of the parts and the hanging tool are synchronously collected through an industrial depth camera and a lateral global industrial camera; carrying out preprocessing and edge detection on the image, extracting the minimum circular hole contour of the part, and carrying out angular point and straight line detection on the hanger image to identify the position of a cargo hook; 3D coordinates of the center point of the circular hole are obtained based on Hough circle transformation, the suction position is determined by combining depth information, and suction is completed through a vacuum chuck at the tail end of a six-axis mechanical arm; by comparing the angle between the side edge of the part and the cargo hook and the overlap ratio between the circular hole and the corner point of the cargo hook, the manipulator is adjusted to realize accurate placement; after each row of hanging pieces is completed, the hanging tool transversely moves by one cargo hook interval; and the hanger is turned over after a single face is completed, and the process is repeated. According to the device, automatic identification, positioning and feeding and discharging of the plate parts are achieved, and the production efficiency and electrophoresis quality consistency are improved.

Description

Automatic loading and unloading method and device for electrophoresis production line of sheet metal parts based on machine vision Technical Field

[0001] This invention belongs to the field of parts loading and unloading, specifically relating to an automatic loading and unloading method for electrophoresis production lines of plate parts based on machine vision. Background Technology

[0002] Plate-type parts, as key basic components in industrial manufacturing, are widely used in industries such as automobiles, home appliances, and electronics, resulting in a huge demand for surface electrophoresis treatment, which has now been implemented on production lines. The electrophoresis process mainly includes loading, pretreatment (degreasing, cleaning, phosphating, etc.), electrophoresis, post-cleaning, drying, and unloading. However, the core step—loading and unloading on the production line (including hanging parts on and removing them from the fixtures)—still heavily relies on manual operation, resulting in low efficiency, high labor costs, high labor intensity, and susceptibility to fatigue errors. This severely restricts the overall efficiency and capacity improvement of electrophoresis production lines, hindering the large-scale application and development of electrophoresis technology. Some equipment manufacturers have also recognized the need for automation in loading and unloading plate-type parts for electrophoresis and have developed related semi-automatic or simple automatic loading and unloading devices. However, due to the difficulty in achieving reliable and high-precision gripping and placement, these devices generally suffer from inaccurate part positioning, misaligned clamping, and even collision damage, failing to truly meet the demands of efficient, flexible, and highly reliable fully automated electrophoresis production. Summary of the Invention

[0003] This invention proposes an automated loading and unloading device for an electrophoresis production line of plate-type parts based on machine vision. First, through visual calibration experiments, a precise mapping model is established between the key visual features (such as contours, corners, and specific markings) of plate-type parts and fixtures and their actual physical dimensions and spatial positions. During the operation of the electrophoresis production line, a global industrial camera is set up at the loading and unloading station and illuminated by a ring light source to acquire images of the plate-type parts to be processed or those that have completed electrophoresis from preset angles (such as the top of the part or the side of the fixture). After the global industrial camera acquires the images, they are sent to an industrial control computer for image processing (including edge contour extraction, circular hole recognition, marking of the center point of the circumscribed rectangle, and depth information) to obtain the key features of the parts. Second, based on the pre-calibrated camera parameters (intrinsic and extrinsic parameters) and pixel equivalent, the pixel accuracy is determined, and then the measured pixel coordinates and dimensions are converted into the actual physical dimensions and spatial coordinates (X, Y, Z) of the parts. After obtaining the location and depth information of the parts to be picked up, the control system drives a six-axis robotic arm to perform a picking action on the plate-type parts in the stack and accurately place them in the designated position on the electrophoresis fixture. At the unloading station, the parts that have completed electrophoresis are removed from the fixture and safely transferred to the unloading area or the next process. This process realizes fully automated, high-precision, and repeatable loading and unloading of plate-type parts on the electrophoresis production line.

[0004] To achieve the above-mentioned objectives, the technical solution of this invention is as follows: An automatic loading and unloading method for electrophoresis production lines of plate-type parts based on machine vision, comprising the following steps:

[0005] S1. Hanger Placement: The top of the hanger is equipped with a slide rail to hook the hanger, move it to the top of the workstation, and place it on the workstation.

[0006] S2. Preparation for loading sheet metal parts: Place sheet metal parts in a dedicated stacking compartment, ensuring that the initial XY positions of the sheet metal parts are the same;

[0007] S3. Image acquisition of plate parts and fixtures: An industrial depth camera is placed above the fixture to detect the image and depth information of the plate parts in real time. Two global industrial cameras are set on the side of the fixture to acquire the image of the fixture and the real-time alignment of the plate parts.

[0008] S4. Image preprocessing: The images of plate parts and fixtures are converted to grayscale and then enhanced and denoised. The Canny operator is used to detect the edges of the plate parts. The contour detection is combined with area filtering to obtain the minimum circular hole contour of the plate parts.

[0009] S5. Hook corner detection and line detection: The preprocessed hanger image is subjected to corner detection and Hough line detection. Corner detection obtains the position of the hook corner, and Hough line transformation obtains the side line of the hook.

[0010] S6. Obtaining 3D coordinates of circular holes: Using the Hough circle transform algorithm, the outline features of circular holes in plate-like parts are marked, and combined with the depth information of plate-like parts, the 3D coordinates of the center point of the circular hole feature are obtained.

[0011] S7. Obtain the suction position: The depth information of the plate-type parts obtained in S3, combined with the stacking position, yields the three-dimensional coordinates of the plate-type parts. The suction position of the vacuum suction cup is the center point of the bounding rectangle of the plate-type part image. The end effector of the six-axis robot uses the vacuum suction cup to suction the plate-type parts.

[0012] S8. Alignment and Placement: When aligning plate-type parts, the posture of the six-axis robot is adjusted by observing the angle between the side of the plate-type part and the hook, and the coincidence of the circular hole outline obtained in S6 and the corner point of the hook obtained in S5, so as to achieve the alignment work.

[0013] S9. Workbench Motion Control: After the robot arm completes the work of hanging items in a row, the hanger moves laterally by the linear module through the interval between the hooks. After all the hanging items on one side are completed, the hanger is flipped over by the gears, and the above operations S1-S9 are repeated.

[0014] Furthermore, step S2, the six-axis robotic arm's picking up and placing, includes the following steps:

[0015] S2.1 Standardized Stacking Design and Workpiece Positioning: A dedicated stacking compartment is designed and configured for plate-type parts. This stacking compartment ensures that the plane coordinate system of all plate-type parts and the reference coordinate system of the stacking compartment are fixed in the stacked state, thereby ensuring that the initial XY plane position and orientation of each part are consistent before picking.

[0016] Furthermore, step S4, image preprocessing, specifically includes the following steps:

[0017] S4.1 converts the captured images of plate-type parts into grayscale images;

[0018] S4.2 Filter and reduce noise from the grayscale converted image of the plate-type parts;

[0019] S4.3 uses the Canny operator to perform edge detection on the denoised plate-type part image;

[0020] S4.4 uses contour detection combined with area filtering to detect the smallest circular hole contour of plate-type parts.

[0021] Furthermore, the filtering method in step S4.2 above can be mean filtering, Gaussian filtering, median filtering, or bilateral filtering.

[0022] Further, step S7 above, obtaining the suction location, specifically includes the following steps:

[0023] S7.1 Establishment of the suction cup model; Given the part to be picked up, the required diameter Dd of the vacuum suction cup can be obtained by the following formula:

[0024]

[0025] Among them, D d [mm] is the diameter of the suction cup, W[N] is the required suction force, and P v [kPa] is the pressure, representing the absolute value of the relative negative pressure. P is the safety factor for the vacuum degree, and t is the safety factor.

[0026] S7.2 converts the diameter Dd of the vacuum chuck into a pixel-level dimension in the image of the plate-type part, and generates a corresponding binary circular mask in the image space, called the chuck mask. The mask diameter is required to be no greater than the width of the plate-type part.

[0027] S7.3 The depth information of the plate-like part obtained from S3 is combined with S2 to obtain the three-dimensional coordinates of the plate-like part. The position is the center point of the bounding rectangle of the plate-like part, and the center point of the mask is corresponding to it.

[0028] S7.4 converts the three-dimensional coordinates of the pick-up center of the plate-type part into three-dimensional coordinates under the robot's base coordinate system, plans the motion trajectory, and instructs the six-axis robot to move its vacuum suction cup to that point, completing the reliable pick-up of the plate-type part and subsequent placement operations.

[0029] The present invention also proposes an automatic loading and unloading device for an electrophoresis production line of plate parts based on machine vision, including an industrial control computer and a computer for image processing, stacking and lifting, worktable movement and controlling a six-axis robot to adjust in real time according to instructions. The invention is characterized by further including an image acquisition and detection device, a stacking and lifting device, a worktable motion control system and a six-axis robot control system.

[0030] The image acquisition and detection device is used to capture images of the surface of plate-type parts and obtain their depth information, including a global industrial camera, an industrial depth camera, and a ring light source; the global industrial camera and the depth camera are equipped with a ring light source. The global industrial camera is used to acquire 2D images of the parts, the industrial depth camera is used to obtain the depth information of the parts, and the ring light source is used to provide supplementary lighting when the global industrial camera acquires images of the parts.

[0031] The stacking and lifting device is used to control the lifting and lowering of plate-type parts inside the stacking compartment. It includes plate-type parts, a compartment body, a cylinder mounting bracket, and a cylinder. The cylinder mounting bracket is fixed to the compartment body. The cylinder is fixed to the cylinder mounting bracket. The cylinder realizes the lifting and lowering of the plate-type parts.

[0032] The worktable is primarily responsible for the movement and overall rotation of the hanger, and it consists of a linear motion component and a gear steering component. The worktable motion control system controls the lateral fixed-distance movement of the hanger and its overall rotation, reducing the programming difficulty of the robotic arm and the hanging time.

[0033] The six-axis robot control system controls the movement trajectory, speed, and posture adjustment of the six-axis robot to achieve precise gripping and placement of plate-type parts; it ensures safe and stable operation under complex working conditions and effectively improves the automation level and production efficiency of the electrophoresis production line.

[0034] The present invention has the following beneficial effects:

[0035] 1. This invention innovatively utilizes the fixed-distance linear movement of the worktable and the overall rotation to reduce the programming complexity of the six-axis robot's pick-up operation, and achieves constant XY coordinates for plate-type parts through a pushable stack, thus reducing the difficulty of pick-up.

[0036] 2. This invention innovatively uses machine vision to acquire surface images of sheet metal parts under complex backgrounds, eliminating the need for a lighting box to provide a special camera environment, making it suitable for use in complex production workshops. Furthermore, it offers fast acquisition speed and high efficiency, making it applicable to industrialized assembly line operations for sheet metal parts.

[0037] 3. Compared with traditional hanging and feeding devices, this invention achieves precise hanging of parts by positioning the side corners of the hanger, reducing the risk of parts being bumped or damaged, and improving the automation level of hanging of plate parts. Attached Figure Description

[0038] Figure 1 is a flowchart of the automatic loading and unloading device for the electrophoresis production line of plate parts based on machine vision according to the present invention.

[0039] Figure 2 is a partial schematic diagram of the image acquisition and detection device of the automatic loading and unloading device for electrophoresis production line of plate parts based on machine vision according to the present invention.

[0040] Figure 3 is a schematic diagram of the stacking of plate parts in the automatic loading and unloading device for electrophoresis production line of plate parts based on machine vision according to the present invention.

[0041] Figure 4 is a schematic diagram of the worktable movement and overall rotation of the automatic loading and unloading device for electrophoresis production line of plate parts based on machine vision according to the present invention.

[0042] Figure 5 is a schematic diagram of the vacuum suction cup mask obtained by the present invention.

[0043] Figure 6 shows the pixel-level edges extracted from the image of the perforated plate-like part of the present invention using the Canny edge detection method.

[0044] Figure 7 shows the maximum outer contour of the perforated plate-type part of the present invention and its center point mark.

[0045] Figure 8 shows the position of the hook vertex obtained by corner point detection using the hanging device of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] As shown in Figure 1, the automatic loading and unloading method for electrophoresis production lines of plate-type parts based on machine vision provided by the present invention includes the following steps:

[0048] S1. Hanger Placement: The top of the hanger is equipped with a slide rail to hook the hanger, move it to the top of the workstation, and place it on the workstation.

[0049] S2. Preparation for loading sheet metal parts: Place sheet metal parts in a dedicated stacking compartment, ensuring that the initial XY positions of the sheet metal parts are the same.

[0050] As a preferred embodiment of the present invention, step S2, the six-axis robotic arm's picking up and placing, specifically includes the following steps:

[0051] S2.1 Standardized Stacking Compartment Design and Workpiece Positioning: A dedicated stacking compartment is designed and configured for the plate-type parts. This compartment ensures that the planar coordinate system of each plate-type part remains fixed with the reference coordinate system of the stacking compartment when the parts are stacked, thereby ensuring that the initial XY plane position and orientation of each part are consistent before picking. This layout provides a stable and predictable source of workpieces for subsequent robot grasping.

[0052] S3. Image acquisition of plate parts and fixtures: An industrial depth camera is placed above the fixture to detect the image and depth information of the plate parts in real time. Two global industrial cameras are set on the side of the fixture to acquire the image of the fixture and the real-time alignment of the plate parts.

[0053] S4. Image Preprocessing: The images of plate parts and fixtures are converted to grayscale and then enhanced and denoised. The Canny operator is used to detect the edges of the plate parts. Contour detection combined with area filtering is used to obtain the minimum circular hole contour of the plate parts.

[0054] In a preferred embodiment of the present invention, step S4. image preprocessing specifically includes the following steps:

[0055] S4.1 converts images of plate-type parts captured by a global industrial camera into grayscale images.

[0056] S4.2 performs noise reduction on the grayscale converted plate-type part image using median filtering.

[0057] As shown in Figure 6, S4.3 uses the Canny operator to perform edge detection on the noise-reduced plate-type parts to obtain pixel-level edges.

[0058] S4.4 uses contour detection combined with area filtering to detect the smallest circular hole contour of plate-type parts.

[0059] S5. Hook corner detection and line detection: The preprocessed hanger image is subjected to corner detection and Hough line detection. The corner detection obtains the position of the hook corner, as shown in Figure 8. The Hough line transformation obtains the side line of the hook.

[0060] S6. Obtaining 3D coordinates of circular holes: Using the Hough circle transform algorithm, the outline features of circular holes in plate-like parts are marked, and combined with the depth information of plate-like parts, the 3D coordinates of the center point of the circular hole feature are obtained.

[0061] S7. Obtain the suction position: The depth information of the plate-type parts obtained in S3, combined with the stacking position, yields the three-dimensional coordinates of the plate-type parts. The suction position of the vacuum suction cup is the center point of the bounding rectangle of the plate-type part image. The end effector of the six-axis robot uses the vacuum suction cup to pick up the plate-type parts.

[0062] In a preferred embodiment of the present invention, step S7, obtaining the suction position, specifically includes the following steps:

[0063] S7.1 Establishment of the suction cup model; Given the part to be picked up, the required diameter Dd of the vacuum suction cup can be obtained by the following formula:

[0064]

[0065] Among them, D d [mm] is the diameter of the suction cup, W[N] is the required suction force, and P v [kPa] is the pressure, representing the absolute value of the relative negative pressure. P is the safety factor for the vacuum degree, and t is the safety factor.

[0066] S7.2 Convert the diameter Dd of the vacuum chuck into a pixel-level dimension in the image of the plate-type part, and generate a corresponding binary circular mask in the image space, called the chuck mask, as shown in Figure 5; the mask diameter is required to be no more than the width of the plate-type part.

[0067] S7.3 The depth information of the plate-like part obtained from S3 is combined with S2 to obtain the three-dimensional coordinates of the plate-like part. The position is extracted as the center point of the outer rectangle of the plate-like part as shown in Figure 7, and the center point of the mask is corresponding to it.

[0068] S7.4 converts the three-dimensional coordinates of the pick-up center of the plate-type part into three-dimensional coordinates under the robot's base coordinate system, plans the motion trajectory, and instructs the six-axis robot to move its vacuum suction cup to that point, completing the reliable pick-up of the plate-type part and subsequent placement operations.

[0069] S8. Alignment and Placement: When aligning plate-type parts, the posture of the six-axis robot is adjusted by observing the angle between the side of the plate-type part and the hook, and the overlap between the outline of the circular hole obtained in S7 and the corner point of the hook obtained in S5, so as to achieve the alignment work.

[0070] S9. Workbench Motion Control: After the robot arm completes the work of hanging items in a row, the hanger moves laterally by the linear module through the interval between the hooks. After all the hanging items on one side are completed, the hanger is flipped over by the gears, and the above operations S1-S9 are repeated.

[0071] Figure 2 is a schematic diagram of the image acquisition and detection device of the automatic loading and unloading device for electrophoresis production line of plate parts based on machine vision according to the present invention, including an industrial depth camera (101), a side global industrial camera (102), a lens (103), and a ring light source (104). The lens (103) is mounted on the industrial depth camera (101) and the side global industrial camera (102), and the ring light source (104) is set outside the lens. When image acquisition begins, the lens (103), the industrial depth camera (101), the side global industrial camera (102), and the ring light source (104) are respectively oriented towards the top of the stacking bin and the right side of the hanger, acquiring images of the upper surface of the plate parts and the side of the hanger.

[0072] Figure 3 is a schematic diagram of the plate part stacking bin of the automatic loading and unloading device for electrophoresis production line of plate parts based on machine vision according to the present invention. It includes a plate part stacking bin (201), a cylinder mounting bracket (202), a cylinder (203), and plate parts (204). Among them, the cylinder mounting bracket (202) is fixed on the plate part stacking bin (201); the cylinder (203) is fixed on the cylinder mounting bracket (202); the plate parts (204) are placed inside the plate part stacking bin (201); the plate parts (204) face the opening of the plate part stacking bin (201), and the cylinder (203) is placed directly below the plate parts (204). When the loading process starts, the cylinder (203) pushes the plate parts (204) upward by the thickness of one plate part (204) each time until all the plate parts (204) are taken away by the six-axis robot, and the cylinder (203) will return to the initial state.

[0073] Figure 4 is a schematic diagram of the horizontal movement and overall rotation of the worktable of the automatic loading and unloading device for electrophoresis production line of plate parts based on machine vision according to the present invention. It includes a hanger (301), a linear module (302), a worktable (303), a motor (304), a pinion (305), and a gear (306). The hanger (301) is placed on the worktable (303) above the linear module, and the gear (306) is welded to the worktable (303). After the six-axis robot hangs a row of parts, the linear module (302) will move laterally by the interval between the hooks until one side of the hanger (301) is completely covered with plate parts. Then, the motor (304) drives the pinion (305), and the pinion (305) drives the gear (306) to flip the entire worktable (303). The linear module (302) then moves in the opposite direction.

[0074] Figure 5 shows the vacuum mask obtained by the present invention based on the size of the plate-like part and the diameter Dd of the vacuum chuck. The diameter of the mask does not exceed the width of the plate-like part.

[0075] Figure 6 shows the pixel-level edges of plate-like parts obtained by the present invention through a series of visual processing steps, specifically including: first, converting the plate-like part image captured by the global industrial camera into a grayscale image; then, using median filtering for noise reduction; and finally, using the Canny operator to perform edge detection on the denoised plate-like part to obtain pixel-level edges.

[0076] Figure 7 shows a schematic diagram of the center point of the bounding rectangle of the plate-type part image based on the present invention, which is mainly achieved by combining contour area filtering with bounding rectangle operation.

[0077] Figure 8 is a schematic diagram of the vertices of each hook of the hanger in this invention, which is mainly implemented by the corner detection function of OpenCV.

Claims

1. A machine vision-based automatic loading and unloading method for electrophoresis production lines of plate-type parts, characterized in that, The system includes a hanger, a workstation, sheet metal parts, a six-axis robot, a global industrial camera, an industrial depth camera, a ring light source, and cylinders. The method includes the following steps: S1. Hanger Placement: A slide rail is installed on the top of the hanger to hook the hanger and move it to the top of the workstation, placing it on the workstation. S2. Sheet Metal Part Loading Preparation: Sheet metal parts are placed in a dedicated stacking bin, ensuring that the initial XY positions of the sheet metal parts are the same. S3. Sheet Metal Part and Hanger Image Acquisition: An industrial depth camera is placed above the hanger to detect the image and depth information of the sheet metal parts in real time. Two global industrial cameras are set on the side of the hanger to acquire the hanger image and the real-time alignment of the sheet metal parts. S4. Image Preprocessing: The sheet metal part image and the hanger image are converted to grayscale and enhanced and denoised. The Canny operator is used for edge detection of the sheet metal parts. Contour detection combined with area filtering is used to obtain the minimum circular hole contour of the sheet metal parts. S5. Hook Corner Detection and Line Detection: The preprocessed hanger image is subjected to corner detection and Hough line detection. Corner detection yields the hook corner. S6. Hough Circle Transform: Using the Hough Circle Transform algorithm, the outline features of the circular holes in plate-like parts are marked. Combined with the depth information of the plate-like parts, the 3D coordinates of the center point of the circular hole feature are obtained. S7. Pickup Position: The depth information of the plate-like parts obtained in S3, combined with the stacking position, yields the three-dimensional coordinates of the plate-like parts. The vacuum suction cup pickup position is the center point of the bounding rectangle of the plate-like part image. The six-axis robot end effector uses the vacuum suction cup to pick up the plate-like parts. S8. Alignment and Placement: When aligning plate-like parts, the six-axis robot adjusts its posture by observing the angle between the side of the plate-like part and the hook, and the overlap between the outline of the circular hole obtained in S6 and the corner point of the hook obtained in S5, to achieve alignment; S9. Workbench Motion Control: After the robot completes the work of hanging a row of parts, the hanger moves laterally by the linear module through the interval between the hooks. After all the hanging work on one side is completed, the hanger is flipped over by the gears, and the above operations S1-S9 are repeated.

2. The automatic loading and unloading method for electrophoresis production lines of plate-type parts based on machine vision as described in claim 1, characterized in that, Step S2 includes: standardized stacking design and workpiece positioning: a dedicated stacking compartment is designed and configured for plate-type parts. This stacking compartment ensures that the plane coordinate system of all plate-type parts in the stacked state is fixed with the reference coordinate system of the stacking compartment, thereby ensuring that the initial XY plane position and orientation of each part are consistent before picking.

3. The automatic loading and unloading method for electrophoresis production lines of plate-type parts based on machine vision as described in claim 1, characterized in that, The image preprocessing in step S4 includes the following steps: S4.1 Converting the image of the plate-like part taken by the industrial depth camera above the fixture into a grayscale image of the plate-like part; S4.2 Filtering and denoising the grayscale image of the plate-like part; S4.3 Using the Canny operator to perform edge detection on the denoised image of the plate-like part; S4.4 Detecting the smallest circular hole contour of the plate-like part by combining contour detection with area filtering.

4. The automatic loading and unloading method for electrophoresis production lines of plate-type parts based on machine vision as described in claim 1, characterized in that, The step S7 of obtaining the suction position includes the following steps: S7.1 Establishing the suction cup model; given the part to be suctioned, the required diameter Dd of the vacuum suction cup can be obtained by the following formula: Among them, D d [mm] is the diameter of the suction cup, W[N] is the required suction force, and P v [kPa] is the pressure, representing the absolute value of the relative negative pressure; P is the safety factor of the vacuum degree; t is the safety factor. S7.2 Convert the diameter Dd of the vacuum suction cup into a pixel-level dimension in the image of the plate-like part, generating a corresponding binary circular mask in the image space, called the suction cup mask. The mask diameter must not exceed the width of the plate-like part. S7.3 Combine the depth information of the plate-like part obtained in S3 with S2 to obtain the three-dimensional coordinates for placing the plate-like part. The suction position is the center point of the bounding rectangle of the plate-like part, and the center point of the mask is aligned with it. S7.4 Convert the three-dimensional coordinates of the suction center of the plate-like part into three-dimensional coordinates under the robot's base coordinate system, plan the motion trajectory, and instruct the six-axis robot to move its vacuum suction cup to this point, completing the reliable suction and subsequent placement operation of the plate-like part.

5. The automatic loading and unloading device for an electrophoresis production line of plate-type parts based on machine vision as described in any one of claims 1-4, comprising an industrial control computer and a computer, used for image processing, stacking and lifting, worktable movement, and controlling a six-axis robot to adjust in real time according to instructions, characterized in that, It also includes an image acquisition and detection device, a stacking and lifting device, a worktable motion control system, and a six-axis robot control system; the image acquisition and detection device is used to capture images of the surface of plate parts and obtain their depth information, including a global industrial camera, an industrial depth camera, and a ring light source; the global industrial camera and the industrial depth camera are equipped with a ring light source, the global industrial camera is used to acquire 2D images of the parts, the industrial depth camera is used to obtain the depth information of the parts, and the ring light source is used to provide supplementary lighting when the global industrial camera acquires images of the parts; the stacking and lifting device is used to control the lifting and lowering of plate parts in the stacking compartment, including a plate part stacking compartment (201), a cylinder mounting bracket (202), a cylinder (203), and a plate part (204), the cylinder (203) being fixed to the plate part stacking compartment (201) by the cylinder mounting bracket (202); the plate part (204) is placed inside the plate part stacking compartment (201); the plate part (204) faces the opening of the plate part stacking compartment (201), and the cylinder (203) is placed... Directly below the plate-like part (204); the workbench motion control system is used to control the lateral fixed-distance movement of the hanger and the overall rotation action. The workbench includes a hanger (301), a linear module (302), a workbench (303), a motor (304), a pinion (305), and a gear (306); the hanger (301) is placed on the workbench (303) above the linear module, and the gear (306) is fixedly connected to the workbench (303); the six-axis robot arm hangs the hanger each time... After a row of parts, the linear module (302) will move laterally by the interval between the hooks until one side of the hanger (301) is completely covered with plate-like parts. Then, the motor (304) drives the pinion (305), which in turn drives the gear (306) to flip the entire worktable (303). The linear module (302) then moves in the opposite direction. The six-axis robot control system includes controlling the movement trajectory, speed, and posture adjustment of the six-axis robot to achieve precise gripping and placement of plate-like parts.

6. The automatic loading and unloading device for electrophoresis production line of plate-type parts based on machine vision as described in claim 5, characterized in that, The industrial depth camera and its ring light source are directed toward the upper surface of the plate-like parts to collect the depth information of the uppermost plate-like parts; the global industrial camera and its ring light source are placed on the side of the fixture for real-time alignment.

Images