Machine vision-based soft package lithium ion battery cell external dimension detection system

Abstract

The invention discloses a soft package lithium ion battery cell appearance size detection system based on machine vision. The system comprises a feeding device, a rotating disc detection device and a discharging device which are used in cooperation. The feeding device is used for conveying the soft package lithium ion battery cell to the turntable detection device; the turntable detection device is used for acquiring and processing an image, measuring the size and judging whether the size is qualified or not; and the discharging device is used for conveying the flexibly-packaged lithium ion battery cells which are detected to be qualified to the next working procedure and removing the flexibly-packaged lithium ion battery cells which are detected to be unqualified from the production line at the same time. The beneficial effects of the invention are that the detection precision is improved; the detection efficiency is improved; the production cost is reduced; the battery quality is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of battery detection, and particularly to an automated detection system for the appearance dimensions of soft-pack lithium-ion battery cells. Background Art

[0002] With the wide application of lithium-ion batteries in fields such as consumer electronics and electric vehicles, soft-pack lithium-ion batteries are favored due to their high energy density, flexible shape, etc. The appearance dimension accuracy of soft-pack lithium-ion battery cells is crucial for the performance, safety, and compatibility with battery modules of the battery. Currently, the detection of the appearance dimensions of soft-pack lithium-ion battery cells mainly relies on manual detection and traditional mechanical measuring tools.

[0003] Manual detection has problems such as strong subjectivity, low efficiency, and easy fatigue. After long-term work, inspectors are prone to misjudgment, resulting in defective products flowing into the next process, affecting the overall quality of the battery. Moreover, the manual detection speed cannot meet the requirements of large-scale production, increasing production costs and production cycles.

[0004] Although traditional mechanical measuring tools such as calipers have high precision, they are complex to operate, require contact measurement with the battery cell, and are prone to causing scratches and other damages to the surface of the battery cell, affecting the performance and appearance of the battery cell. At the same time, this detection method has a low degree of automation and cannot achieve fast and continuous on-line detection. Summary of the Invention

[0005] The purpose of the present invention is to provide a detection system for the appearance dimensions of soft-pack lithium-ion battery cells based on machine vision, so as to achieve fast, accurate, and non-contact detection of the appearance dimensions of soft-pack lithium-ion battery cells, improve detection efficiency and accuracy, reduce production costs, and ensure battery quality.

[0006] To achieve the above purpose, the technical solution provided by the present invention is as follows:

[0007] A detection system for the appearance dimensions of soft-pack lithium-ion battery cells based on machine vision, comprising a feeding device, a turntable detection device, and a discharging device that are used in cooperation;

[0008] The feeding device is used to convey the soft-pack lithium-ion battery cells to the turntable detection device;

[0009] The turntable detection device is used for image acquisition, processing, dimension measurement, and determination of whether it is qualified;

[0010] The discharging device is used to convey the qualified soft-pack lithium-ion battery cells to the next process, and at the same time remove the unqualified soft-pack lithium-ion battery cells from the production line.

[0011] Furthermore, the feeding device includes a feeding frame and an automatic feeding drive servo motor, an automatic feeding linear module, a feeding manipulator, a transfer manipulator, and a transfer platform installed on the feeding frame;

[0012] The automatic feeding drive servo motor is connected to the lead screw in the automatic feeding linear module through a synchronous wheel and a synchronous belt transmission;

[0013] The automatic feeding linear module is a linear module guided by a guide rail and a slider driven by a screw rod. The module load drives two parallel material taking and placing manipulators, a feeding manipulator and a transfer manipulator;

[0014] The loading robot is used to move the battery cells that have been taped from the previous process to the transfer platform;

[0015] The transfer robot is used to move the battery cells after the tape is applied from the transfer platform to the turntable detection device.

[0016] Furthermore, the feeding robot and the transfer robot both include a robot frame connected to the automatic feeding linear module and a telescopic cylinder installed on the robot frame, and a suction plate is provided at the end of the piston rod of the telescopic cylinder.

[0017] Furthermore, the suction cup adopts a three-channel suction plate structure, which can adjust the suction cup hole position according to the large, medium and small sizes of the battery cell. Specifically, the first channel connects to four first suction cups arranged in a row longitudinally, the second channel connects to two rows of second suction cups symmetrically arranged on both sides of the first suction cup, totaling eight, and the third channel connects to eight third suction cups arranged in a square shape on the outside of the second suction cup.

[0018] Furthermore, the turntable detection device includes a turntable and an image acquisition module, an image processing module, a size measurement module and a control module;

[0019] The image acquisition module is used to acquire battery cell images;

[0020] The image processing module is connected to the image acquisition module, and is used to pre-process the received battery cell image using an image processing algorithm, and then perform image segmentation to separate the battery cell from the background, thereby providing an accurate target image for subsequent size measurement;

[0021] The size measurement module is used to extract key feature information of the battery cell through a feature extraction algorithm based on the battery cell image after image processing, and use the key feature information in combination with a pre-set coordinate system and geometric calculation method to accurately measure the length, width, and thickness appearance size parameters of the battery cell;

[0022] The control module is used as the core control unit of the entire system to coordinate the operations of the image acquisition module, image processing module, and dimension measurement module. It is used to receive the measurement results of the dimension measurement module and compare them with the preset dimension standards to determine whether the battery cell is qualified. If the dimensions of the battery cell are within the tolerance range, it is determined to be qualified, and the battery cell is allowed to continue to the next process. If it exceeds the tolerance range, it is determined to be unqualified, and an alarm is issued through the alarm device. At the same time, the corresponding rejection device can be controlled to remove the unqualified battery cell from the production line.

[0023] Furthermore, for image segmentation, a threshold-based segmentation method is adopted, and the Otsu method is combined to automatically determine the optimal threshold. According to the gray-scale characteristics of the battery cell image, the battery cell is separated from the background to obtain a binary image containing only the battery cell.

[0024] The feature extraction algorithm is to use the edge detection algorithm to extract the edge information of the battery cell image. By connecting and fitting the edge pixel points, the complete edge contour of the battery cell is obtained. Then, the corner detection algorithm is used to determine the corner positions of the battery cell.

[0025] The dimension calculation algorithm adopted in the dimension measurement module is to establish a coordinate system based on the extracted edge and corner information. Taking a corner point of the battery cell as the origin, by calculating the distances between the pixel points on the edge and combining the calibration parameters of the camera, the length, width, and thickness appearance dimension parameters of the battery cell are accurately calculated.

[0026] Furthermore, the image acquisition module includes a back bottom tape data acquisition device, a front hot melt tape data acquisition device, a front top tape acquisition device, a front bottom tape data acquisition device, and a back top tape acquisition device.

[0027] The back bottom tape data acquisition device uses a coaxial light source and a CCD camera to take pictures of the battery cell to be detected on the back of the turntable for collecting the surface features of the battery cell. The camera's field of view is adjusted according to the accuracy requirements, and the turntable is used to complete the station exchange work.

[0028] The front hot melt tape data acquisition device uses a combined lighting method of a front strip light source and a backlight, and uses a CCD camera to take pictures of the surface of the battery cell for collecting the surface features of the battery cell.

[0029] The front top tape acquisition device uses a front coaxial light source and a CCD camera to collect the surface features of the battery cell.

[0030] The front bottom tape data acquisition device 13 uses a front coaxial light source and a CCD camera to collect the surface features of the battery cell.

[0031] The back top tape acquisition device uses a back coaxial light source and a CCD camera to collect the surface features of the battery cell.

[0032] Further, the CCD camera is a 12 - megapixel color CCD camera.

[0033] Further, the key feature information includes the edges and corners of the battery cell.

[0034] Further, the blanking device includes a blanking mechanism light source, a blanking robot, a battery cell buffer belt, a blanking sorting manipulator, a sorting linear module, and a sorting NG bin;

[0035] The blanking mechanism light source includes a CCD camera, two strip light sources, and a belt conveyor mechanism, which is used to take pictures and locate the position of the tray during the blanking of good battery cells, and provide position coordinates for the pole group buffer.

[0036] The blanking robot is used for rotary blanking, putting good battery cells into the tray and bad battery cells on the sorting transfer platform.

[0037] The battery cell buffer belt includes a belt, a stepping motor, and a lifting servo motor, which is used to store good battery cell pole groups and empty trays.

[0038] The blanking sorting manipulator includes a suction plate and a Z - axis lifting cylinder to complete the transfer of bad battery cells.

[0039] The sorting linear module includes a servo motor, a slide rail slider, and a lead screw, which is used to complete the movement trajectory of the blanking sorting manipulator.

[0040] The sorting NG bin includes a lifting motor, a lifting platform, and a box body, which is used to store different types of NG battery cells.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0042] (1) Improve the detection accuracy

[0043] By adopting machine vision technology, the subjectivity of manual detection and the possible damage caused by the contact measurement of traditional mechanical measuring tools are avoided. Through high - precision image acquisition and advanced image - processing algorithms, the appearance dimensions of the soft - package lithium - ion battery cells can be accurately measured, and the measurement accuracy can reach 0.01 mm.

[0044] (2) Improve the detection efficiency

[0045] The system can realize automatic detection, with a fast detection speed, and can meet the requirements of large - scale production. The number of battery cells that can be detected per hour reaches 1000. Compared with the traditional detection method, the detection time is greatly shortened, and the production efficiency is improved.

[0046] (3) Reduce the production cost

[0047] It reduces the labor cost of manual inspection and at the same time reduces the cost of cell damage caused by manual misjudgment and traditional detection methods. By promptly and accurately removing unqualified cells, it avoids waste in subsequent processes and reduces the overall production cost.

[0048] (4) Ensure battery quality

[0049] Strict dimensional inspection can ensure that the appearance dimensions of the cells entering the next process meet the standards, improving the consistency and stability of the battery, thereby ensuring the performance and safety of the battery. Description of the drawings

[0050] Figure 1 It is a schematic structural diagram of the detection system provided by the embodiment of the present application;

[0051] Figure 2 It is a schematic structural diagram of the feeding device provided by the embodiment of the present application;

[0052] Figure 3 It is a schematic structural diagram of the turntable detection device provided by the embodiment of the present application;

[0053] Figure 4 It is a schematic structural diagram of the discharging device provided by the embodiment of the present application;

[0054] Figure 5 It is a schematic structural diagram of the suction tray provided by the embodiment of the present application;

[0055] Figure 6 It is a schematic flow diagram of detection using the detection system provided by the present application. Detailed implementation manners

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with 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, rather than all 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.

[0057] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0058] In the description of this patent, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", and "set" should be understood in a broad sense, for example, it can be fixedly connected, set, or detachably connected, set, or integrally connected, set, and can refer to direct connection or indirect connection. For ordinary technicians in this field, the specific meanings of the above terms in this patent can be understood according to specific circumstances.

[0059] like Figure 1 As shown, this embodiment provides a system for detecting the appearance and size of a soft-pack lithium-ion battery cell based on machine vision, including a loading device, a turntable detection device, and a unloading device used in conjunction with each other;

[0060] The feeding device is used to transport the soft-pack lithium-ion battery cells to the turntable detection device;

[0061] The turntable detection device is used to collect and process images, measure dimensions, and determine whether the image is qualified.

[0062] The unloading device is used to convey the soft-pack lithium-ion battery cells that have passed the inspection to the next process, and remove the soft-pack lithium-ion battery cells that have failed the inspection from the production line.

[0063] like Figure 2 As shown, the feeding device includes a feeding frame and an automatic feeding drive servo motor 1 installed on the feeding frame, an automatic feeding linear module 2, a feeding manipulator 3, a transfer manipulator 4, and a transfer platform 5;

[0064] The automatic feeding drive servo motor 1 is connected to the screw rod in the automatic feeding linear module 2 by means of synchronous wheel and synchronous belt transmission;

[0065] The automatic feeding linear module 2 is a linear module guided by a screw rod driving a slide rail slider. The module load drives two parallel material taking and placing manipulators, a feeding manipulator 3 and a transfer manipulator 4.

[0066] The loading robot 3 is used to move the battery cells after the tape is applied from the previous process to the transfer platform 5;

[0067] The transfer robot 4 is used to move the battery cells after the tape is applied from the transfer platform 5 to the turntable detection device.

[0068] Among them, the feeding robot 3 and the transfer robot 4 both include a robot frame connected to the automatic feeding linear module 2 and a telescopic cylinder installed on the robot frame, and a suction plate is provided at the end of the piston rod of the telescopic cylinder.

[0069] like Figure 5As shown, the suction plate adopts a three-channel suction plate structure, which can adjust the sucker hole positions according to the large, medium, and small models of the battery cells. Specifically, the first channel is connected to 4 first suckers 6 arranged longitudinally in a row, the second channel is connected to two rows of second suckers 7 symmetrically arranged on both sides of the first suckers, a total of 8, and the third channel is connected to 8 third suckers 8 arranged in a square outside the second suckers.

[0070] Among them, the turntable detection device includes a turntable 9, an image acquisition module, an image processing module, a dimension measurement module, and a control module;

[0071] The image acquisition module is used to acquire the images of the battery cells;

[0072] Among them, the image acquisition module includes an industrial camera, a lens, and a light source. The industrial camera adopts a high-resolution and high-speed camera, which can quickly capture clear images of the soft-pack lithium-ion battery cells. The lens is selected according to the size of the battery cells and the requirements of detection accuracy to ensure the imaging quality. The light source adopts a specially designed lighting system, such as a combination of a ring light source and a backlight source, to provide uniform and stable lighting conditions for the battery cells and reduce the influence of shadows on the image quality.

[0073] The image processing module is connected to the image acquisition module and is used to preprocess the received battery cell images using image processing algorithms, including operations such as denoising and enhancing contrast. Then, image segmentation is performed to separate the battery cells from the background and provide accurate target images for subsequent dimension measurement.

[0074] Specifically, the image processing process starts from image collection. After collecting the image through the exposure and charge transfer of the binocular linear array charge-coupled device, the analog signal is pre-amplified and filtered. The processed signal is converted between analog and digital signals through a signal conversion system, and then the digital signal is preprocessed, including correcting the dark current and flat field, and then performing image enhancement and optimization, such as gray processing, filtering processing, and sharpening processing on the image. The preprocessed image is subjected to image segmentation, and the image is segmented according to thresholds, regions, and edges. The segmented images are subjected to feature extraction. Geometric features, texture features, or color features can be used according to the actual features of the battery. The extracted features are subjected to target recognition, the features are classified, and finally the results are output and displayed.

[0075] The dimension measurement module is used to extract key feature information such as the edges and corners of the battery cells from the processed battery cell images based on feature extraction algorithms, and use the key feature information to accurately measure the appearance dimension parameters such as the length, width, and thickness of the battery cells in combination with a pre-set coordinate system and geometric calculation methods;

[0076] The control module is used as the core control unit of the entire system to coordinate the work of the image acquisition module, the image processing module, and the dimension measurement module. It is used to receive the measurement results of the dimension measurement module and compare them with the preset dimension standards to determine whether the battery cell is qualified. If the dimensions of the battery cell are within the tolerance range, it is determined to be qualified, and the battery cell is allowed to continue to the next process. If it exceeds the tolerance range, it is determined to be unqualified, and an alarm is issued through the alarm device. At the same time, the corresponding rejection device can be controlled to remove the unqualified battery cell from the production line.

[0077] In addition, as the core control unit of the entire system, the control module also controls the shooting parameters of the industrial camera, such as exposure time, frame rate, etc. At the same time, the control module can also communicate with other devices on the production line to achieve seamless docking between the detection system and the production process.

[0078] It should be noted that

[0079] Denoising algorithm: The denoising algorithm adopts the mean filtering and median filtering algorithms to denoise the collected images. By replacing the value of each pixel point in the image with the median value of the pixel values in its neighborhood, this algorithm effectively removes salt-and-pepper noise in the image while retaining important information such as the edges of the battery cell.

[0080] Contrast enhancement algorithm: The histogram equalization algorithm is used to enhance the contrast of the denoised image. By redistributing the gray values of the image, the gray distribution of the image becomes more uniform, improving the contrast between the battery cell and the background and facilitating subsequent image segmentation operations.

[0081] Image segmentation algorithm: A threshold-based segmentation method is adopted, and the Otsu's method is combined to automatically determine the optimal threshold. According to the gray characteristics of the battery cell image, the battery cell is separated from the background to obtain a binary image containing only the battery cell.

[0082] Feature extraction algorithm: Edge detection algorithms such as the Canny edge detection algorithm are used to extract the edge information of the battery cell image. By connecting and fitting the edge pixel points, the complete edge contour of the battery cell is obtained. Then, corner detection algorithms such as the Harris corner detection algorithm are used to determine the corner positions of the battery cell.

[0083] Dimension calculation algorithm: Based on the extracted edge and corner information, a coordinate system is established. Taking a corner point of the battery cell as the origin, by calculating the distances between pixel points on the edge and combining the calibration parameters of the camera (such as the ratio relationship between pixels and actual dimensions), the dimension parameters such as the length, width, and thickness of the battery cell are accurately calculated.

[0084] Such as Figure 3As shown in the figure, the image acquisition module includes a back bottom tape data acquisition device 10, a front hot melt tape data acquisition device 11, a front top tape acquisition device 12, a front bottom tape acquisition device 13, and a back top tape acquisition device 14;

[0085] The back bottom tape data acquisition device 10 uses a coaxial light source and a CCD camera to take pictures of the cell to be detected on the back of the turntable, collect the surface features of the cell, adjust the field of view of the camera according to the accuracy requirements, and use the turntable to complete the station exchange work; specifically, collect and measure the image of the tape sticking size on the cell surface.

[0086] The front hot melt tape data acquisition device 11 uses a combination of a front strip light source and a back light source for lighting, and uses a CCD camera to take pictures of the cell surface to collect the surface features of the cell; specifically, collect and detect the image of the front hot melt tape features.

[0087] The front top tape acquisition device 12 uses a front coaxial light source and a CCD camera to collect the surface features of the cell; specifically, collect and detect the image of the bottom tape features on the front of the battery.

[0088] The front bottom tape data acquisition device 13 uses a front coaxial light source and a CCD camera to collect the surface features of the cell; specifically, collect and detect the image of the bottom tape features on the front of the battery.

[0089] The back top tape acquisition device 14 uses a back coaxial light source and a CCD camera to collect the surface features of the cell; specifically, collect and detect the image of the bottom tape features on the front of the battery.

[0090] Among them, the CCD camera is a 12-megapixel color CCD camera.

[0091] It should be noted that

[0092] Industrial camera selection: According to the cell size and detection accuracy requirements, select an industrial camera with appropriate resolution and frame rate, such as the Hikvision MV-CS0060-10UC color camera, whose resolution is 3072×2048 and the frame rate can reach 59.6fps. Install the camera above the detection area, and by adjusting the height and angle of the mounting bracket, ensure that the camera lens is vertically aligned with the cell surface to obtain the best imaging effect.

[0093] Lens selection and installation: Select a lens matching the industrial camera, such as the Hikvision MVL-MF3518M-5MPE lens, with a focal length of 35mm. Install the lens on the industrial camera and perform fine focusing to ensure that the cell is clearly imaged on the camera imaging plane.

[0094] Light source selection and arrangement: According to the shape and surface material of the battery cell, a combination of linear light source and backlight is selected. The linear light source is used to provide uniform illumination on the surface of the battery cell, and the backlight is used to highlight the contour of the battery cell. The ring light source is installed around the camera lens, and the backlight is installed under the battery cell. By adjusting the brightness and angle of the light source, the image of the battery cell has high contrast and low shadow. Or a coaxial light source is used to image a small area on the surface of the battery.

[0095] Installation of other hardware: The control module is installed in the control cabinet and connected to devices such as the image acquisition module, image processing module, sensors on the production line, and rejection device through cables. The algorithm of the dimension measurement module runs on a computer connected to the image processing module to ensure fast data transmission and processing.

[0096] As Figure 4 shown, the blanking device includes a blanking mechanism light source 15, a blanking robot 16, a battery cell buffer belt 17, as well as a blanking sorting manipulator 18, a sorting linear module 19, and a sorting NG bin 20;

[0097] The blanking mechanism light source 15 includes a CCD camera, 2 strip light sources, and a belt conveyor mechanism, which is used to take pictures and locate the position of the tray during the blanking of good battery cells and provide position coordinates for the pole group buffer;

[0098] The blanking robot 16 is used for rotary blanking, putting good battery cells into the tray and placing bad battery cells on the sorting transfer platform;

[0099] The battery cell buffer belt 17 includes a belt, a stepping motor, and a lifting servo motor, which is used to store good battery cell pole groups and store empty trays;

[0100] The blanking sorting manipulator 18 includes a suction plate and a Z-axis lifting cylinder to complete the transfer of bad battery cells;

[0101] The linear sorting module 19 includes a servo motor, a slide rail slider, and a lead screw, which is used to complete the movement trajectory of the blanking sorting manipulator 18;

[0102] The sorting NG bin 20 includes a lifting motor, a lifting platform, and a box body, which is used to store different types of NG battery cells.

[0103] It should be noted that after the system of the present invention is installed, the hardware is first debugged. By adjusting the parameters of the industrial camera, lens, and light source, images of different battery cell samples are collected, and the image quality is observed to ensure that the images are clear, shadow-free, and have good contrast. The camera is calibrated to determine the conversion relationship between pixels and actual dimensions, improving the accuracy of dimension measurement.

[0104] Then, the software algorithm is debugged. Input battery cell images of different qualities and check the running results of the image processing and size measurement algorithms. Optimize the parameters of algorithms such as denoising, contrast enhancement, image segmentation, feature extraction, and size calculation to ensure that the algorithm can accurately process battery cell images in various situations and obtain correct size measurement results.

[0105] After the debugging is completed, the system is put into formal operation. The battery cells are continuously conveyed on the production line. When the battery cells reach the detection area, the system automatically performs image acquisition, processing, and size measurement. The control module judges in real time whether the battery cells are qualified and controls the action of the rejection device according to the judgment result. At the same time, the system can record and statistically analyze the detection data, generate a detection report, and provide data support for production quality control.

[0106] As Figure 6 shown, the detection process provided by this application is specifically as follows:

[0107] First, build a test system and make corresponding adjustments to the light source according to the requirements of extracting features of the battery.

[0108] The pouch lithium-ion battery cells are conveyed to the detection area through the battery cell feeding module. When the battery cells reach the specified position, the sensor signal is triggered. After receiving the sensor signal, the control module starts the image acquisition module to perform image acquisition.

[0109] The acquired image is transmitted to the image processing module for processing. After preprocessing and image segmentation, a clear battery cell target image is obtained.

[0110] The size measurement module extracts features from the target image, detects the corresponding sizes of the adhesive tapes, and transmits the measurement results to the control module.

[0111] The control module compares the measurement results with the preset standards. If the battery cell size is within the tolerance range, it is judged as qualified, and the battery cell is allowed to continue to the next process; if it exceeds the tolerance range, it is judged as unqualified, and an alarm is issued through the alarm device. At the same time, the corresponding rejection device can be controlled to remove the unqualified battery cells from the production line.

[0112] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A soft-pack lithium-ion battery cell appearance size detection system based on machine vision, characterized in that: It includes a loading device, a turntable detection device and a unloading device used in conjunction with each other; The feeding device is used to transport the soft-pack lithium-ion battery cells to the turntable detection device; The turntable detection device is used to collect and process images, measure dimensions, and determine whether the image is qualified. The unloading device is used to convey the soft-pack lithium-ion battery cells that have passed the inspection to the next process, and remove the soft-pack lithium-ion battery cells that have failed the inspection from the production line.

2. According to the machine vision-based soft-pack lithium-ion battery cell appearance size detection system of claim 1, it is characterized in that: The feeding device comprises a feeding frame, an automatic feeding drive servo motor (1) installed on the feeding frame, an automatic feeding linear module (2), a feeding manipulator (3), a transfer manipulator (4), and a transfer platform (5); The automatic feeding drive servo motor (1) is connected to the inner screw of the automatic feeding linear module (2) by means of a synchronous wheel and a synchronous belt transmission; The automatic feeding linear module (2) is a linear module guided by a guide rail and a slider driven by a screw rod, and the module load drives two parallel material taking and placing manipulators, a feeding manipulator (3) and a transfer manipulator (4); The loading robot (3) is used to move the battery cells after the tape is applied from the previous process to the transfer platform (5); The transfer robot (4) is used to move the battery cells after the adhesive tape is applied from the transfer platform (5) to the turntable detection device.

3. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 2, characterized in that: The feeding robot (3) and the transfer robot (4) both comprise a robot frame connected to the automatic feeding linear module (2) and a telescopic cylinder mounted on the robot frame, wherein a suction disc is provided at the end of the piston rod of the telescopic cylinder.

4. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 3, characterized in that: The suction cup adopts a three-channel suction plate structure, and the suction cup hole position can be adjusted according to the large, medium and small sizes of the battery cell. Specifically, the first channel is connected to four first suction cups (6) arranged in a row longitudinally, the second channel is connected to two rows of second suction cups (7) symmetrically arranged on both sides of the first suction cup, a total of eight, and the third channel is connected to eight third suction cups (8) arranged in a square shape outside the second suction cup.

5. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 1, characterized in that: The turntable detection device comprises a turntable (9) and an image acquisition module, an image processing module, a dimension measurement module and a control module; The image acquisition module is used to acquire battery cell images; The image processing module is connected to the image acquisition module, and is used to pre-process the received battery cell image using an image processing algorithm, and then perform image segmentation to separate the battery cell from the background, thereby providing an accurate target image for subsequent size measurement; The size measurement module is used to extract key feature information of the battery cell through a feature extraction algorithm based on the battery cell image after image processing, and use the key feature information in combination with a pre-set coordinate system and geometric calculation method to accurately measure the length, width, and thickness appearance size parameters of the battery cell; The control module is used as the core control unit of the entire system, coordinating the work of the image acquisition module, the image processing module and the size measurement module, and is used to receive the measurement results of the size measurement module and compare them with the preset size standard to determine whether the battery cell is qualified; if the battery cell size is within the tolerance range, it is determined to be qualified and the battery cell is allowed to continue to enter the next process; if it exceeds the tolerance range, it is determined to be unqualified, and an alarm is issued through the alarm device, and the corresponding rejection device can be controlled to remove the unqualified battery cell from the production line.

6. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 5, characterized in that: The image segmentation adopts the threshold-based segmentation method, combined with the Otsu method to automatically determine the optimal threshold; according to the grayscale characteristics of the battery cell image, the battery cell is separated from the background to obtain a binary image containing only the battery cell; The feature extraction algorithm uses an edge detection algorithm to extract edge information of the battery cell image; connects and fits edge pixels to obtain a complete edge contour of the battery cell; and then uses a corner detection algorithm to determine the corner position of the battery cell. The size calculation algorithm used in the size measurement module is to establish a coordinate system based on the extracted edge and corner point information; take a corner point of the battery cell as the origin, calculate the distance between the pixel points on the edge, and combine the calibration parameters of the camera to accurately calculate the length, width and thickness appearance size parameters of the battery cell.

7. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 5, characterized in that: The image acquisition module comprises a back bottom adhesive tape data acquisition device (10), a front hot melt adhesive tape data acquisition device (11), a front top adhesive tape acquisition device (12), a front bottom adhesive tape acquisition device (13), and a back top adhesive tape acquisition device (14); The back bottom tape data acquisition device (10) uses a coaxial light source and a CCD camera to take pictures of the battery cell to be inspected on the back of the turntable to collect the surface features of the battery cell, adjusts the camera's field of view according to the accuracy requirements, and uses the turntable to complete the workstation exchange work; The front hot melt tape data acquisition device (11) uses a front strip light source and a backlight source combined lighting method, and uses a CCD camera to take pictures of the surface of the battery cell to collect the surface characteristics of the battery cell; The front top tape collection device (12) uses a front coaxial light source and a CCD camera to collect the surface characteristics of the battery cell; The front bottom tape data acquisition device 13 uses a front coaxial light source and a CCD camera to collect the surface characteristics of the battery cell; The back top adhesive tape collection device (14) uses a back coaxial light source and a CCD camera to collect the surface features of the battery cell.

8. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 7, characterized in that: The CCD camera is a 12-megapixel color CCD camera.

9. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 8, characterized in that: The key feature information includes the edges and corners of the battery cell.

10. The machine vision-based soft-pack lithium-ion battery cell appearance size detection system according to claim 1, characterized in that: The unloading device comprises an unloading mechanism light source (15), an unloading robot (16), a battery cell buffer belt (17), an unloading and sorting manipulator (18), a sorting linear module (19), and a sorting NG material box (20); The unloading mechanism light source (15) comprises a CCD camera, two strip light sources, and a belt conveyor mechanism, and is used to take photos and locate the position of the tray when unloading good quality battery cells, and provide position coordinates for the electrode group buffer; The unloading robot (16) is used for unloading materials from a turntable, placing good quality batteries into a tray, and placing bad quality batteries onto a sorting transfer platform; The battery cell buffer belt (17) comprises a belt, a stepping motor, and a lifting servo motor, and is used to store good battery pole groups and empty trays; The material sorting robot (18) comprises a suction plate and a Z-axis lifting cylinder to complete the transfer of defective battery cells; The sorting linear module (19) comprises a servo motor, a slide rail and a lead screw, and is used to complete the moving track of the blanking sorting robot (18); The sorting NG material box (20) comprises a lifting motor, a lifting platform, and a box body, and is used for storing different types of NG battery cells.

Images