Method, device and equipment for detecting positive and negative collector plates of cylindrical battery and medium

Abstract

The invention relates to a cylindrical battery positive and negative collector plate detection method, device and equipment and a medium, and the method comprises the steps: grabbing a collector plate, and adjusting the collector plate to a horizontal state; the collector plates in the horizontal state are conveyed along a preset detection path, and the collector plates are sequentially positioned to a plurality of detection stations; in at least one detection station provided with a liquid lens, driving a rotating jig to perform indexing rotation on a borne collector plate, and acquiring a multi-focus image sequence of a specific area of the collector plate through the liquid lens at a plurality of rotation angles; according to the multi-focus image sequence and detection data obtained at the multiple detection stations, the defects of the current collecting disc are recognized, the qualification of the current collecting disc is comprehensively judged, and sorting, discharging and disc arranging operation is executed according to the judgment result. The purpose of high-speed, high-precision and full-coverage automatic detection of dimensional deviation, appearance flaws and micron-sized burr defects of the cylindrical battery collector plate is achieved.

Description

Technical Field

[0001] This invention relates to the field of new energy battery manufacturing and automated testing technology, and in particular to a method, apparatus, equipment and medium for testing the positive and negative current collectors of cylindrical batteries. Background Technology

[0002] With the rapid development of new energy vehicles and energy storage industries, cylindrical all-tab batteries, as core power units, have attracted much attention regarding their manufacturing quality and safety reliability. The current collector, as a key component connecting the electrodes and current collectors within the battery, is of paramount importance. During production, high-speed stamping and cleaning processes can easily generate various defects on the current collector, such as micron-level burrs, scratches, copper leaks, and dimensional deviations. These minute flaws can easily lead to serious self-discharge, internal short circuits, and even thermal runaway risks during subsequent battery use, posing serious safety hazards. Currently, the industry faces severe challenges in online inspection of current collectors: traditional manual visual inspection is inefficient, highly subjective, and cannot meet the pace of high-speed production lines; while existing automated visual inspection solutions, either due to positioning errors and vibration issues at high speeds with rotary structures, or due to the difficulty in overcoming efficiency bottlenecks in frequent start-stop cycles with translational structures, cannot achieve comprehensive, high-precision dimensional and appearance defect detection of irregularly shaped current collector structures at a pace of ≥100PPM, especially lacking the stable detection capability for microburrs smaller than 100μm. Summary of the Invention

[0003] The main objective of this invention is to provide a method, apparatus, equipment, and medium for detecting the positive and negative current collectors of cylindrical batteries, so as to achieve high-speed, high-precision, and full-coverage automated detection of dimensional deviations, appearance defects, and micron-level burr defects of cylindrical battery current collectors.

[0004] To achieve the above objectives, the present invention provides a method for detecting the positive and negative current collectors of a cylindrical battery, comprising the following steps: Grab the collector disk and adjust it to a horizontal position; The horizontally positioned collector plate is transported along the preset detection path and then sequentially positioned to multiple detection stations. At at least one inspection station equipped with a liquid lens, a rotating fixture is driven to rotate the supported manifold in increments, and at multiple rotation angles, a multi-focus image sequence of a specific area of ​​the manifold is acquired through the liquid lens. Based on the multifocal image sequence and the detection data obtained at the multiple detection stations, defects in the manifold are identified, the qualification of the manifold is comprehensively determined, and sorting, unloading, and tray placement operations are performed according to the determination results.

[0005] Further, the step of grasping the collector disk and adjusting the collector disk to a horizontal state includes: It can grab multiple collection discs arranged in an array in a material basket at one time; The collector plate is flipped from a non-horizontal state to a horizontal state by a flipping mechanism; The horizontally positioned manifolds are placed on a temporary storage fixture, and the spacing between the manifolds is adjusted using a pitch-changing mechanism.

[0006] Furthermore, the steps of conveying the horizontally positioned collector plate along a preset detection path and sequentially positioning the collector plate to multiple detection stations include: The collector is transported to the detection path, which contains multiple detection channels configured in parallel. The control collector is positioned sequentially in each detection channel to multiple detection stations arranged in a series.

[0007] Furthermore, the multiple testing stations arranged in series include a first type of testing station and a second type of testing station; At the first type of inspection station, the collector plate is photographed by a vertically oriented camera to inspect the surface appearance defects and inner and outer diameter dimensions of the collector plate; In the second type of inspection station, the manifold is scanned by a 3D line scan camera to detect the side wing height, flatness and scratch depth of the manifold.

[0008] Furthermore, prior to the step of acquiring a multifocal image sequence of a specific area of ​​the manifold through the liquid lens, the method further includes: Tilt calibration is performed on the image acquisition device equipped with a liquid lens; The tilt calibration includes: placing the standard calibration plate vertically, taking pictures with the camera lens optical axis at a tilt angle to the normal of the calibration plate, and obtaining the fused calibration image in the tilt state through a multi-focus image fusion algorithm; The corresponding relationship between pixel size and actual physical size under tilted shooting is established by performing calculations based on the fused calibration image and the standard calibration image taken in the vertical state.

[0009] Furthermore, at at least one inspection station equipped with a liquid lens, the step of driving a rotating fixture to rotate the supported manifold in indexing increments, and acquiring a multi-focal image sequence of a specific area of ​​the manifold through the liquid lens at multiple rotation angles, includes: Position the manifold to the testing station equipped with the liquid lens; Drive the rotating fixture used to carry the collector plate to rotate in an indexing motion; At each rotation angle, a green light source is used to illuminate the welded part, spring, and area around the central hole of the manifold. Green light source illumination enhances the contrast of the collector disk defect edges during image acquisition and reduces dispersion effects; By adjusting the voltage applied to the liquid lens, the focal length is adjusted, and multiple images of the green light source-illuminated area at different focal planes are acquired from the same viewpoint. A multifocal image sequence is constructed by collecting images at different rotation angles.

[0010] Further, based on the multi-focus image sequence and the detection data obtained at the multiple detection stations, the defects of the manifold are identified, the passability of the manifold is comprehensively determined, and the sorting, unloading, and tray placement operations are performed according to the determination results, including: The multifocal image sequence is processed by a multifocal image fusion algorithm to generate a fused image with full depth of field. Based on the fused image, the image captured by the first type of inspection station, and the three-dimensional morphology data scanned by the second type of inspection station, at least one of the size defects, appearance defects, and burr defects of the manifold is identified by at least one of the size detection algorithm, appearance deep learning algorithm, and multi-focus image fusion algorithm. Summarize the detection results of manifold size defects, appearance defects, and burr defects, and classify the detected manifold defects according to their categories; The manifold is determined to be a qualified or unqualified product based on the defect category. Based on the judgment results, the collection trays are sorted and placed into the qualified product collection tray and the non-qualified product collection tray corresponding to the defect type.

[0011] The present invention also provides a detection device for the positive and negative current collectors of a cylindrical battery, used for detecting the positive current collector and the negative current collector, comprising: The feeding module is used to grab the collection tray and adjust the collection tray to a horizontal position; The positioning module is used to transport the horizontally positioned collector plate along a preset detection path and to position the collector plate sequentially to multiple detection stations. An image acquisition module is used to drive a rotating fixture to rotate a carrier manifold in increments at at least one inspection station equipped with a liquid lens, and to acquire a multi-focus image sequence of a specific area of ​​the manifold through the liquid lens at multiple rotation angles. The data processing module is used to identify defects in the manifold based on the multifocal image sequence and the detection data obtained at the multiple detection stations, comprehensively determine the qualification of the manifold, and perform sorting, unloading, and tray placement operations based on the determination results.

[0012] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described cylindrical battery positive and negative electrode current collector detection method.

[0013] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method for detecting the positive and negative current collectors of a cylindrical battery.

[0014] The present invention provides a method, apparatus, equipment, and medium for detecting the positive and negative current collectors of cylindrical batteries, which has the following beneficial effects: By introducing a collaborative detection architecture of parallel multi-channel and serial multi-station, the present invention overcomes the bottleneck of detection efficiency in traditional single-channel or rotary table solutions, achieving comprehensive detection of current collectors under high-speed production cycles. The use of a liquid lens combined with a rotating fixture and multi-focal image fusion technology solves the industry problem of difficulty in achieving clear imaging from all angles due to depth-of-field limitations and angular obstruction of the complex, irregularly shaped current collector structure, and improves the ability to capture and identify minute burrs. Furthermore, by fusing 2D vision, 3D morphology, and multi-focal fused images for comprehensive analysis of multi-source data, and combining dedicated algorithms and deep learning models, the present invention achieves accurate identification and classification of various defect types such as dimensional deviations, appearance flaws, and microscopic burrs. By integrating multiple detection processes into one unit, the present invention avoids the cumbersome process of multiple devices handling each process separately, and achieves efficient, accurate, and intelligent quality control of current collectors while ensuring a high detection rate and a low over-detection rate, fundamentally improving the safety and reliability of battery products. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the detection method for the positive and negative current collectors of a cylindrical battery in one embodiment of the present invention. Figure 2 This is a layout block diagram of the detection path of the positive and negative current collectors of a cylindrical battery in one embodiment of the present invention; Figure 3 This is a structural block diagram of a portion of the detection stations in the detection path of the positive and negative electrode current collectors of a cylindrical battery in one embodiment of the present invention; Figure 4 This is a structural block diagram and a light source layout example diagram of the multi-angle irregular structure of the detection station in the detection path of the positive and negative electrode current collector of a cylindrical battery in one embodiment of the present invention. Figure 5 This is a structural block diagram of a cylindrical battery positive and negative electrode current collector detection device according to an embodiment of the present invention; Figure 6 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0016] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] Reference Figure 1 This is a flowchart illustrating a method for detecting the positive and negative current collectors of a cylindrical battery proposed in this invention, including the following steps: S1, grab the collector disk and adjust the collector disk to a horizontal state; S2, transport the horizontally positioned collector plate along the preset detection path, and position the collector plate sequentially to multiple detection stations; S3, at at least one inspection station equipped with a liquid lens, drive the rotating fixture to rotate the supported manifold in increments, and at multiple rotation angles, acquire multi-focus image sequences of a specific area of ​​the manifold through the liquid lens; S4. Based on the multi-focus image sequence and the detection data obtained at the multiple detection stations, identify the defects of the manifold, comprehensively determine the qualification of the manifold, and perform sorting, unloading, and tray placement operations according to the determination results.

[0019] In one embodiment, for step S1, The steps of grabbing the collector disk and adjusting the collector disk to a horizontal position include: It can grab multiple collection discs arranged in an array in a material basket at one time; The collector plate is flipped from a non-horizontal state to a horizontal state by a flipping mechanism; The horizontally positioned manifolds are placed on a temporary storage fixture, and the spacing between the manifolds is adjusted using a pitch-changing mechanism.

[0020] In practice, this step begins with the automated integration with the front-end cleaning process. A loading robot picks up multiple collector trays (e.g., arranged in an N×M array) from the output basket of the cleaning machine at once. A flipping mechanism rotates the collector trays from a non-horizontal (e.g., vertical) position in the basket to a horizontal position. This orientation adjustment is the baseline requirement for stable positioning and consistent imaging at all subsequent inspection stations. Horizontal placement effectively avoids the shaking caused by vertical instability, ensuring a fixed position during the rotating fixture and transport process. This guarantees the accuracy and repeatability of the camera's shooting angle, providing the necessary stability and consistency for subsequent dimensional measurements and micro-defect identification. Subsequently, the horizontally positioned collector trays are placed on a temporary storage fixture. A pitch-adjusting mechanism precisely adjusts the spacing between the collector trays, ensuring that the layout accurately matches the camera's field of view, the light source's illumination area, and the depth of field of the optics at each subsequent inspection station. This ensures that each collector tray can be completely and clearly imaged when simultaneously inspecting multiple workpieces, avoiding mutual interference and laying the foundation for subsequent high-precision visual inspection. Step S1 achieves an efficient, smooth, and precise transition of the manifold from the upstream process to the core inspection stage through a continuous automated operation of "batch grabbing - posture flipping - pitch adjustment".

[0021] In one embodiment, for step S2, The steps of conveying the horizontally positioned manifold along a preset detection path and sequentially positioning the manifold to multiple detection stations include: The collector is transported to the detection path, which contains multiple detection channels configured in parallel. The control collector is positioned sequentially in each detection channel to multiple detection stations arranged in a series.

[0022] Reference Figure 2In practical implementation, to achieve a balance between high efficiency and high precision detection, this invention adopts a hybrid layout architecture of "parallel channels and serial workstations." Specifically, horizontally positioned manifolds are transported to a detection path containing multiple detection channels arranged in parallel. This parallel design is a core technical means to overcome the limitations of high-speed production cycles (≥100 PPM), allowing multiple manifolds to be transported and detected simultaneously in different channels, effectively doubling the overall throughput of the system, thereby fundamentally solving the problem of the detection speed limit caused by serial processing in traditional single-channel solutions. Within each independent detection channel, the manifolds flow sequentially and are precisely positioned according to multiple serially arranged detection workstations. The serial layout ensures that each manifold can receive a complete and orderly detection process, where each workstation is specifically optimized to perform a specific detection task (e.g., one workstation specializes in 2D dimensions and appearance, while the next workstation is responsible for 3D morphology measurement). The clearly defined serial detection mode avoids mutual interference between different detection items, ensuring that each detection is performed under optimized imaging conditions.

[0023] In one embodiment, the multiple testing stations arranged in series include a first type of testing station and a second type of testing station; At the first type of inspection station, the collector plate is photographed by a vertically oriented camera to inspect the surface appearance defects and inner and outer diameter dimensions of the collector plate; In the second type of inspection station, the manifold is scanned by a 3D line scan camera to detect the side wing height, flatness and scratch depth of the manifold.

[0024] Reference Figure 3In specific implementation, the multiple serially arranged inspection stations are specifically divided into complementary first and second types of inspection stations. Through the synergy of different imaging technologies, comprehensive quality control of the macroscopic dimensions, appearance, and microscopic three-dimensional morphology of the manifold is achieved. At the first type of inspection station, the system is equipped with a vertically oriented high-resolution area array camera to perform orthophoto imaging on the upper or lower surface of the stationary manifold under uniform and stable lighting conditions. This station mainly undertakes two-dimensional vision inspection tasks: based on the captured high-definition two-dimensional images, image processing algorithms (such as sub-pixel edge extraction) are used to accurately measure key dimensions such as the inner and outer diameters of the manifold to ensure compliance with tolerance requirements; using trained deep learning models or traditional machine vision algorithms, the images are analyzed to efficiently identify and classify macroscopic appearance defects such as dirt, copper leakage, black spots, embossing, and surface yellowing. The advantage of this station is its fast two-dimensional imaging speed and high resolution, making it very suitable for rapid planar dimension measurement and large-area appearance screening. After passing the first type of inspection, the manifold is transported to the second type of inspection station, which integrates a 3D line-scan laser camera. As the manifold passes through the laser line at a constant speed, the camera rapidly acquires continuous 3D point cloud data of its surface using the principle of triangulation. This workstation specializes in 3D parameters that are difficult to quantify or cannot be detected by traditional 2D vision: through the reconstructed 3D topography, the height and thickness of the manifold's side wings are accurately measured to ensure its structural integrity; simultaneously, the flatness of the welding surface can be evaluated with high precision to ensure key indicators of subsequent welding quality; 3D contour data is extremely sensitive to surface defects such as scratches and dents, and can accurately quantify the depth, length, and width of the manifold, thereby effectively identifying microscopic 3D defects that are not obvious in 2D images but actually pose safety hazards. By serially integrating 2D and 3D vision technologies and implementing a detection strategy that allows each to perform its specific function, the advantages of each technology are fully utilized to form a three-dimensional, high-precision detection system from two-dimensional to three-dimensional and from macroscopic to microscopic, ensuring the comprehensiveness of defect detection and the accuracy of measurement data.

[0025] In one embodiment, prior to step S3 Before the step of acquiring a multifocal image sequence of a specific area of ​​the manifold through the liquid lens, the method further includes: Tilt calibration is performed on the image acquisition device equipped with a liquid lens; The tilt calibration includes: placing the standard calibration plate vertically, taking pictures with the camera lens optical axis at a tilt angle to the normal of the calibration plate, and obtaining the fused calibration image in the tilt state through a multi-focus image fusion algorithm; The corresponding relationship between pixel size and actual physical size under tilted shooting is established by performing calculations based on the fused calibration image and the standard calibration image taken in the vertical state.

[0026] In practical implementation, before acquiring multifocal image sequences of a specific area of ​​the manifold, a preprocessing step is required: tilt calibration of the image acquisition device equipped with a liquid lens is performed to ensure the accuracy of the detection results under all subsequent tilted angles, especially the measurement accuracy of minute burr sizes. Because the key areas of the manifold, such as the welded parts and spring contacts, have irregular three-dimensional structures, the camera lens optical axis often needs to be tilted at a certain angle to the normal of the surface being measured in order to fully capture their features. This results in severe perspective distortion, causing a non-linear change in the correspondence between image pixel size and actual physical size (i.e., pixel equivalent). Directly using calibration parameters for vertical shooting will introduce unacceptable measurement errors. To address this, this invention performs tilt calibration: a standard calibration plate with a known linewidth (e.g., 30 μm) is placed vertically, and the camera integrated with the liquid lens is adjusted to a specific tilt angle for operation (e.g., the lens optical axis is at a 30° angle to the normal of the calibration plate), and the calibration plate is then photographed. In this tilted state, due to depth-of-field limitations, a single image is insufficient to clearly image the entire calibration board. Therefore, the rapid zoom capability of the liquid lens is utilized to acquire a series of images with different focal planes. A multi-focal image fusion algorithm is then applied to generate a fused calibration image of the tilted state where the entire calibration board is clearly visible. Finally, by comparing and performing geometric calculations with a standard calibration image taken by the camera in a vertical state with known pixel equivalents, the accurate correspondence between the pixel size and the actual physical size at each location in the image at that specific tilt angle can be precisely calculated, thus establishing a high-precision tilt vision measurement model. This calibration process minimizes the impact of perspective distortion and insufficient depth of field on measurement accuracy, laying a metrological foundation for reliable and accurate detection of micron-level burrs (e.g., ≤50μm) under complex working conditions.

[0027] In one embodiment, for step S3, The steps of driving a rotating fixture to rotate a supported manifold at least one inspection station equipped with a liquid lens, and acquiring a multi-focal image sequence of a specific area of ​​the manifold through the liquid lens at multiple rotation angles, include: Position the manifold to the testing station equipped with the liquid lens; Drive the rotating fixture used to carry the collector plate to rotate in an indexing motion; At each rotation angle, a green light source is used to illuminate the welded part, spring, and area around the central hole of the manifold. Green light source illumination enhances the contrast of the collector disk defect edges during image acquisition and reduces dispersion effects; By adjusting the voltage applied to the liquid lens, the focal length is adjusted, and multiple images of the green light source-illuminated area at different focal planes are acquired from the same viewpoint. A multifocal image sequence is constructed by collecting images at different rotation angles.

[0028] Reference Figure 4 In specific implementation, step S3 aims to achieve 360-degree full coverage and clear panoramic imaging of irregular three-dimensional structures on the manifold (such as welded parts, spring clips, and the area around the central hole). After precisely positioning the manifold to a dedicated inspection station equipped with a liquid lens, the rotating fixture carrying the manifold is driven to perform precise indexing rotation (e.g., 120° rotation each time, for a total of three angular positions). This rotation mechanism ensures that even features with complex three-dimensional contours can achieve full coverage without blind spots through multi-angle shooting, which is impossible with fixed-angle shooting. At each rotation angle, a specific wavelength of green light is used to illuminate a specific area (such as welded parts, spring clips, and the area around the central hole). Compared to other wavelengths, green light produces sharper image edge contrast on the surface of the manifold (usually made of metal) and effectively reduces the dispersion effect that may occur during imaging by the liquid lens. Through the combined effect of the rotation angle and green light, imaging features such as tiny burrs and scratches are significantly highlighted.

[0029] Under optimized lighting conditions, by changing the control voltage applied to the liquid lens, the lens curvature is dynamically and precisely adjusted within milliseconds based on the electrowetting effect, thereby rapidly changing the focal length. This process enables the system to instantly acquire multiple images of a series of different focal planes within the target area from the same camera perspective. Ultimately, the images of all different focal planes acquired at all rotation angles are combined to form a complete multifocal image sequence. This sequence not only contains 360-degree orientation information of the target area but also complete depth (depth of field) information for each orientation. This embodiment solves the imaging blurring problem caused by insufficient depth of field due to the irregular shape of the collector disk, providing a high-quality image foundation for subsequent accurate identification.

[0030] In one embodiment, for step S4, Based on the multi-focus image sequence and the detection data obtained at the multiple detection stations, the defects of the manifold are identified, the passability of the manifold is comprehensively determined, and the sorting, unloading, and tray placement operations are performed according to the determination results, including: The multifocal image sequence is processed by a multifocal image fusion algorithm to generate a fused image with full depth of field. Based on the fused image, the image captured by the first type of inspection station, and the three-dimensional morphology data scanned by the second type of inspection station, at least one of the size defects, appearance defects, and burr defects of the manifold is identified by at least one of the size detection algorithm, appearance deep learning algorithm, and multi-focus image fusion algorithm. Summarize the detection results of manifold size defects, appearance defects, and burr defects, and classify the detected manifold defects according to their categories; The manifold is determined to be a qualified or unqualified product based on the defect category. Based on the judgment results, the collection trays are sorted and placed into the qualified product collection tray and the non-qualified product collection tray corresponding to the defect type.

[0031] In specific implementation, a multifocal image fusion algorithm is used to perform depth-of-field fusion processing on the multifocal image sequence obtained in step S3. This algorithm, through intelligent computation, seamlessly stitches and synthesizes the clearest regions from multiple images at different focal planes from the same viewpoint, ultimately generating a panoramic depth-of-field fused image that is clear regardless of whether the entire target area is above or tilted relative to the focal plane. Then, a multi-source information collaborative defect identification stage is entered. Based on the fused image, the two-dimensional images captured by the first type of inspection station, and the three-dimensional morphology data scanned by the second type of inspection station, a three-dimensional, complementary dataset is formed. Different algorithms are invoked or fused for identification based on different defect types: a size detection algorithm is used to perform sub-pixel-level precise measurement of key dimensions to determine size defects; a trained appearance deep learning model (such as a convolutional neural network) is used to intelligently analyze the two-dimensional images to efficiently identify appearance defects such as dirt, copper leakage, and black spots; simultaneously, the detection capability for minute burr defects is further enhanced by combining the three-dimensional information and features of the fused image. Through the fusion strategy of multiple algorithms and multiple data sources, the coverage and accuracy of defect identification are improved, effectively reducing false positives and false negatives. After all defect identification is completed, the inspection results from each workstation are summarized and automatically classified according to predefined categories such as "dimensional defects," "appearance defects," and "burr defects." Based on this classification result, a final conformity judgment is performed: if no defects are detected or the defects are within the allowable range, the product is judged as conforming (OK); otherwise, it is judged as non-conforming (NG). Finally, the unloading and sorting mechanism performs precise physical sorting and palletizing operations based on this judgment result, neatly placing conforming products into a dedicated conforming product collection tray, while placing non-conforming products of different defect types into their corresponding NG collection trays. This refined sorting and palletizing strategy not only achieves quality isolation but also provides valuable data support for process traceability and quality analysis in the production process, realizing full automation and intelligence from inspection and decision-making to execution.

[0032] Reference Figure 5 Here is a structural block diagram of a cylindrical battery positive and negative electrode current collector detection device according to an embodiment of the present invention, comprising: The feeding module is used to grab the collection tray and adjust the collection tray to a horizontal position; The positioning module is used to transport the horizontally positioned collector plate along a preset detection path and to position the collector plate sequentially to multiple detection stations. An image acquisition module is used to drive a rotating fixture to rotate a carrier manifold in increments at at least one inspection station equipped with a liquid lens, and to acquire a multi-focus image sequence of a specific area of ​​the manifold through the liquid lens at multiple rotation angles. The data processing module is used to identify defects in the manifold based on the multifocal image sequence and the detection data obtained at the multiple detection stations, comprehensively determine the qualification of the manifold, and perform sorting, unloading, and tray placement operations based on the determination results.

[0033] For the specific implementation of each module in the above device example, please refer to the above method embodiments, which will not be repeated here.

[0034] Reference Figure 6 This invention also provides a computer device, which can be a server, and its internal structure can be as follows: Figure 6 As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0035] Those skilled in the art will understand that Figure 6 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0036] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0037] In summary, this invention achieves high-speed, high-precision, and full-coverage automated inspection of cylindrical battery current collectors by grasping the current collector and adjusting it to a horizontal position; conveying the horizontal current collector along a preset inspection path and sequentially positioning it at multiple inspection stations; at at least one inspection station equipped with a liquid lens, driving a rotating fixture to rotate the supported current collector in increments, and acquiring multi-focus image sequences of specific areas of the current collector through the liquid lens at multiple rotation angles; identifying defects in the current collector based on the multi-focus image sequences and the inspection data obtained at the multiple inspection stations, comprehensively determining the passability of the current collector, and performing sorting and unloading operations based on the determination results.

[0038] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0039] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0040] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for detecting the positive and negative current collectors of a cylindrical battery, characterized in that, Includes the following steps: Grab the collector disk and adjust it to a horizontal position; The horizontally positioned collector plate is transported along the preset detection path and then sequentially positioned to multiple detection stations. At at least one inspection station equipped with a liquid lens, a rotating fixture is driven to rotate the supported manifold in increments, and at multiple rotation angles, a multi-focus image sequence of a specific area of ​​the manifold is acquired through the liquid lens. Based on the multifocal image sequence and the detection data obtained at the multiple detection stations, defects in the manifold are identified, the qualification of the manifold is comprehensively determined, and sorting, unloading, and tray placement operations are performed according to the determination results.

2. The method for detecting the positive and negative current collectors of a cylindrical battery according to claim 1, characterized in that, The step of grabbing the collector disk and adjusting the collector disk to a horizontal state includes: It can grab multiple collection discs arranged in an array in a material basket at one time; The collector plate is flipped from a non-horizontal state to a horizontal state by a flipping mechanism; The horizontally positioned manifolds are placed on a temporary storage fixture, and the spacing between the manifolds is adjusted using a pitch-changing mechanism.

3. The method for detecting the positive and negative current collectors of a cylindrical battery according to claim 1, characterized in that, The step of conveying the horizontally positioned collector plate along a preset detection path and sequentially positioning the collector plate to multiple detection stations includes: The collector is transported to the detection path, which contains multiple detection channels configured in parallel. The control collector is positioned sequentially in each detection channel to multiple detection stations arranged in a series.

4. The method for detecting the positive and negative current collectors of a cylindrical battery according to claim 3, characterized in that, The multiple detection stations arranged in series include a first type of detection station and a second type of detection station; At the first type of inspection station, the collector plate is photographed by a vertically oriented camera to inspect the surface appearance defects and inner and outer diameter dimensions of the collector plate; In the second type of inspection station, the manifold is scanned by a 3D line scan camera to detect the side wing height, flatness and scratch depth of the manifold.

5. The method for detecting the positive and negative current collectors of a cylindrical battery according to claim 1, characterized in that, Before the step of acquiring a multifocal image sequence of a specific area of ​​the manifold through the liquid lens, the method further includes: Tilt calibration is performed on the image acquisition device equipped with a liquid lens; The tilt calibration includes: placing the standard calibration plate vertically, taking pictures with the camera lens optical axis at a tilt angle to the normal of the calibration plate, and obtaining the fused calibration image in the tilt state through a multi-focus image fusion algorithm; The corresponding relationship between pixel size and actual physical size under tilted shooting is established by performing calculations based on the fused calibration image and the standard calibration image taken in the vertical state.

6. The method for detecting the positive and negative current collectors of a cylindrical battery according to claim 1, characterized in that, The step of driving a rotating fixture to rotate a supported manifold at least one inspection station equipped with a liquid lens, and acquiring a multi-focal image sequence of a specific area of ​​the manifold through the liquid lens at multiple rotation angles, includes: Position the manifold to the testing station equipped with the liquid lens; Drive the rotating fixture used to carry the collector plate to rotate in an indexing motion; At each rotation angle, a green light source is used to illuminate the welded part, spring, and area around the central hole of the manifold. Green light source illumination enhances the contrast of the collector disk defect edges during image acquisition and reduces dispersion effects; By adjusting the voltage applied to the liquid lens, the focal length is adjusted, and multiple images of the green light source-illuminated area at different focal planes are acquired from the same viewpoint. A multifocal image sequence is constructed by collecting images at different rotation angles.

7. The method for detecting the positive and negative current collectors of a cylindrical battery according to claim 4, characterized in that, The steps of identifying defects in the manifold based on the multi-focus image sequence and the detection data obtained at the multiple detection stations, comprehensively determining the pass / fail status of the manifold, and performing sorting, unloading, and tray placement operations based on the determination results include: The multifocal image sequence is processed by a multifocal image fusion algorithm to generate a fused image with full depth of field. Based on the fused image, the image captured by the first type of inspection station, and the three-dimensional morphology data scanned by the second type of inspection station, at least one of the size defects, appearance defects, and burr defects of the manifold is identified by at least one of the size detection algorithm, appearance deep learning algorithm, and multi-focus image fusion algorithm. Summarize the detection results of manifold size defects, appearance defects, and burr defects, and classify the detected manifold defects according to their categories; The manifold is determined to be a qualified or unqualified product based on the defect category. Based on the judgment results, the collection trays are sorted and placed into the qualified product collection tray and the non-qualified product collection tray corresponding to the defect type.

8. A device for detecting the positive and negative current collectors of a cylindrical battery, used for detecting the positive current collector and the negative current collector, characterized in that... include: The feeding module is used to grab the collection tray and adjust the collection tray to a horizontal position; The positioning module is used to transport the horizontally positioned collector plate along a preset detection path and to position the collector plate sequentially to multiple detection stations. An image acquisition module is used to drive a rotating fixture to rotate a carrier manifold in increments at at least one inspection station equipped with a liquid lens, and to acquire a multi-focus image sequence of a specific area of ​​the manifold through the liquid lens at multiple rotation angles. The data processing module is used to identify defects in the manifold based on the multifocal image sequence and the detection data obtained at the multiple detection stations, comprehensively determine the qualification of the manifold, and perform sorting, unloading, and tray placement operations based on the determination results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the cylindrical battery positive and negative electrode current collector detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the cylindrical battery positive and negative electrode current collector detection method according to any one of claims 1 to 7.

Images