Detection system for explosion-proof valve assembly

Abstract

A detection system for an explosion-proof valve assembly. The detection system comprises a processor (103), a first image collection member (101), and a second image collection member (102), wherein the first image collection member (101) is configured to perform image collection on an explosion-proof valve assembly (106) to obtain a first image, and send the first image to the processor (103); the second image collection member (102) is configured to perform image collection on the explosion-proof valve assembly (106) to obtain a second image, and send the second image to the processor (103); and the processor (103) is configured to receive the first image and the second image, and when it is determined that defects are present in the first image and the second image, perform defect position detection on the first image and the second image to determine a defect position of the explosion-proof valve assembly (106). The explosion-proof valve assembly (106) comprises an explosion-proof valve (1062) and a protective film (1061) covering the explosion-proof valve (1062).

Description

Explosion-proof valve assembly testing system

[0001] Cross-references to related applications

[0002] This disclosure is based on and claims priority to Chinese Patent Application No. 202510117735.0, filed on January 24, 2025, entitled "Detection System for Explosion-proof Valve Assembly", the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to the field of battery manufacturing technology, and in particular to a detection system for an explosion-proof valve assembly. Background Technology

[0004] During operation, lithium batteries generate significant heat and gas through internal chemical reactions, which can easily lead to explosions or even fires. Applying explosion-proof valves and protective films to the batteries can control the internal gas pressure, allowing for limited gas release and preventing overheating, overcharging, and over-discharging. However, because the protective films for explosion-proof valves are typically transparent, existing detection methods struggle to accurately pinpoint the location of defects relative to the protective film, failing to meet the reliability requirements of industrial testing. Summary of the Invention

[0005] This disclosure provides a testing system for explosion-proof valve assemblies, which can improve the testing accuracy and efficiency of explosion-proof valve assemblies, thereby enhancing the safety of the battery cells under test.

[0006] The technical solution disclosed herein is implemented as follows:

[0007] In a first aspect, this disclosure provides a detection system for an explosion-proof valve assembly. The detection system includes a processor, a first image acquisition unit, and a second image acquisition unit; wherein:

[0008] The first image acquisition unit is used to acquire images of the explosion-proof valve assembly, obtain a first image, and send the first image to the processor;

[0009] The second image acquisition unit is used to acquire images of the explosion-proof valve assembly, obtain a second image, and send the second image to the processor;

[0010] The processor is configured to receive a first image and a second image, and when it is determined that a defect exists in the first image and the second image, to perform defect location detection on the first image and the second image to determine the defect location of the explosion-proof valve assembly; wherein the explosion-proof valve assembly includes an explosion-proof valve and a protective film covering the explosion-proof valve.

[0011] Using the aforementioned technical means, a first image acquisition device is used to capture an image of the explosion-proof valve assembly, and a second image acquisition device is used to capture an image of the same assembly, resulting in a second image. Defect detection is then performed on both images. If a defect is detected, its location within the explosion-proof valve assembly is determined through defect location detection. In this way, when the first and second image acquisition devices capture images of the same explosion-proof valve assembly, defects above or below the protective film of the explosion-proof valve are mapped onto the image. This not only allows for differentiation between defects located on the surface of the protective film and those on the surface of the explosion-proof valve, but also improves the accuracy and efficiency of the explosion-proof valve assembly detection, thereby enhancing the safety of the battery cell under test.

[0012] In some embodiments, the first image acquisition unit and the second image acquisition unit are on the same horizontal plane, and the first image acquisition unit and the second image acquisition unit are symmetrically arranged above the explosion-proof valve assembly.

[0013] Using the aforementioned technical means, a first image acquisition unit and a second image acquisition unit, which are symmetrically arranged on the same horizontal plane, are used to capture images of the same explosion-proof valve assembly, obtaining a first image and a second image respectively. In this way, based on the parallax principle, the location of the defect can be determined from the first and second images, thus distinguishing whether the defect is located on the surface of the protective film or the surface of the explosion-proof valve.

[0014] In some embodiments, the detection system further includes a light source component for turning on or off based on a control signal sent by a processor; a first image acquisition component for acquiring a first image when the light source component is turned on; and a second image acquisition component for acquiring a second image when the light source component is turned on.

[0015] Using the aforementioned technical means, when the light source assembly is turned on, the first image acquisition unit and the second image acquisition unit capture images of the same explosion-proof valve assembly, obtaining a first image and a second image respectively. This improves the clarity of the acquired first and second images, thereby increasing the accuracy of the detection.

[0016] In some embodiments, when the protective film is in a first configuration, the first image acquisition device and the second image acquisition device are arranged on the same horizontal plane along a first direction. The first configuration indicates that the difference between the length of the protective film in the first direction and the length of the protective film in the second direction is greater than a first threshold.

[0017] By employing the aforementioned technical means, when the protective film is in its first form, the first image acquisition unit and the second image acquisition unit are arranged on the same horizontal plane along the first direction. This avoids missing images of the explosion-proof valve assembly, thereby improving the accuracy of defect detection.

[0018] In some embodiments, when the protective film is in the second form, the first image acquisition device and the second image acquisition device are arranged on the same horizontal plane along the second direction. The second form indicates that the difference between the length of the protective film in the first direction and the length of the protective film in the second direction is less than or equal to a first threshold.

[0019] By employing the aforementioned technical means, when the protective film is in its second form, the first image acquisition unit and the second image acquisition unit are arranged on the same horizontal plane along the second direction. This minimizes the impact of the protective film's shadow area on the imaging effect, improves imaging clarity, and thus enhances the accuracy of defect detection.

[0020] In some embodiments, the processor is communicatively connected to a first image acquisition unit and a second image acquisition unit, respectively; wherein: the processor is configured to establish communication connections with the first image acquisition unit and the second image acquisition unit based on a first preset protocol; the first image acquisition unit is configured to encode a first image based on a second preset protocol to generate an encoded first image, and send the encoded first image to the processor through the communication connection; the second image acquisition unit is configured to encode a second image based on a second preset protocol to generate an encoded second image, and send the encoded second image to the processor through the communication connection.

[0021] Through the aforementioned technical means, the processor can receive the first image and the second image, respectively encoded by the first image acquisition device and the second image acquisition device, based on a first preset protocol. This improves the communication efficiency between the processor and the first and second image acquisition devices, thereby enhancing detection efficiency.

[0022] In some embodiments, the processor is further configured to determine depth information of the defect based on the pixel difference between the same point on the defect in the first image and the second image; and to determine, based on the depth information, whether the defect is located on the surface of the protective film in the explosion-proof valve assembly or on the surface of the explosion-proof valve.

[0023] Using the aforementioned techniques, the processor determines the depth information of the defect based on the pixel difference between the same point on the defect in the first and second images, and further determines the location of the defect. Thus, defects of different heights correspond to different depth information, enabling the differentiation between defects on the surface of the protective film and those on the surface of the explosion-proof valve, thereby improving the accuracy of defect detection.

[0024] In some embodiments, the processor is further configured to detect defects based on a first image and / or a second image, determine the characteristic parameters of the defects, and determine whether the explosion-proof valve assembly is abnormal based on the defect location and characteristic parameters of the explosion-proof valve assembly.

[0025] By employing the aforementioned technical means, based on the first and / or second images, the characteristic parameters of the defect are determined. Furthermore, based on the characteristic parameters and location of the defect, it is determined whether the explosion-proof valve assembly exhibits any abnormality. In this way, by comprehensively considering various parameters of the defect, it is possible to determine whether the defect in the explosion-proof valve assembly is abnormal, thereby improving the accuracy and precision of the detection.

[0026] In some embodiments, the processor is further configured to determine that the appearance of the protective film is abnormal when the characteristic parameters of the defect meet the first characteristic condition; or to determine that the appearance of the protective film is not abnormal when the characteristic parameters of the defect do not meet the first characteristic condition.

[0027] By employing the aforementioned technical methods, when the processor determines that a defect is located on the surface of the protective film, it can determine whether the appearance of the protective film of the explosion-proof valve is abnormal based on whether the defect's characteristic parameters meet the first characteristic condition. This improves the accuracy of detecting the protective film of the explosion-proof valve.

[0028] In some embodiments, the processor is further configured to determine that the appearance of the explosion-proof valve is abnormal when the characteristic parameters of the defect meet the second characteristic condition; or to determine that the appearance of the explosion-proof valve is not abnormal when the characteristic parameters of the defect do not meet the second characteristic condition.

[0029] By employing the aforementioned technical methods, when the processor determines that a defect is located on the surface of the explosion-proof valve, it can determine whether the valve's appearance is abnormal based on whether the defect's characteristic parameters meet the second characteristic condition. This improves the accuracy of explosion-proof valve inspection.

[0030] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the technical solutions of this utility model. Attached Figure Description

[0031] Figure 1 is a schematic diagram of the composition structure of a detection system for an explosion-proof valve assembly provided in an embodiment of this disclosure;

[0032] Figure 2 is a schematic diagram of the composition structure of a detection system for an explosion-proof valve assembly provided in an embodiment of this disclosure;

[0033] Figure 3 is a schematic diagram of a battery cell to be tested according to an embodiment of this disclosure;

[0034] Figure 4 is a schematic diagram of a method for setting up a first image acquisition device and a second image acquisition device according to an embodiment of this disclosure;

[0035] Figure 5 is a schematic diagram of the composition structure of a detection system for an explosion-proof valve assembly provided in an embodiment of this disclosure;

[0036] Figure 6 is a schematic diagram of a method for setting up a first image acquisition device and a second image acquisition device according to an embodiment of this disclosure;

[0037] Figure 7 is a schematic diagram of a triangulation method provided in an embodiment of this disclosure;

[0038] Figure 8 is a schematic diagram of a first image and a second image provided in an embodiment of this disclosure;

[0039] Figure 9 is a schematic diagram of the composition structure of a detection system for an explosion-proof valve assembly provided in an embodiment of this disclosure;

[0040] Figure 10 is a detailed flowchart illustrating a testing method for an explosion-proof valve assembly provided in an embodiment of this disclosure. Detailed Implementation

[0041] To gain a more detailed understanding of the features and technical content of the embodiments of this disclosure, the implementation of the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for reference and illustration only and are not intended to limit the embodiments of this disclosure.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. The terminology used herein is for the purpose of describing embodiments of this disclosure only and is not intended to be limiting of this disclosure.

[0043] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0044] It should also be noted that the terms "first, second, third" used in the embodiments of this disclosure are only used to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of this disclosure described herein can be implemented in an order other than that illustrated or described herein.

[0045] Furthermore, the reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this disclosure. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0046] The following is a description of the relevant technologies disclosed herein.

[0047] New energy batteries are being used more and more widely in daily life and industry. For example, new energy vehicles equipped with batteries are already widely used. In addition, batteries are being used more and more in the field of energy storage.

[0048] Currently, new energy batteries are being used more and more widely in daily life and industry. They are not only used in energy storage systems for hydropower, thermal power, wind power, and solar power plants, but also extensively in electric vehicles such as electric bicycles, electric motorcycles, and electric cars, as well as in aerospace and other fields. With the continuous expansion of the application areas of power batteries, the market demand is also constantly increasing.

[0049] In this embodiment, the battery can be a single battery cell. A single battery cell refers to a basic unit capable of converting chemical energy into electrical energy, and can be used to manufacture battery modules or battery packs to supply power to electrical devices. A single battery cell can be a rechargeable battery, which is a battery cell that can be recharged after discharge to reactivate its active materials and continue to be used. The single battery cell can be a lithium-ion battery, sodium-ion battery, sodium-lithium-ion battery, lithium metal battery, sodium metal battery, lithium-sulfur battery, magnesium-ion battery, nickel-metal hydride battery, nickel-cadmium battery, lead-acid battery, etc., and this embodiment is not limited to these types.

[0050] In embodiments of this disclosure, the battery may also be a single physical module comprising one or more battery cells to provide higher voltage and capacity. When there are multiple battery cells, the multiple battery cells are connected in series, parallel, or mixed via a busbar.

[0051] New energy vehicles widely use batteries as their power source. A battery typically consists of at least one cell. During cell manufacturing, to ensure battery safety, an explosion-proof valve and a protective film are usually installed on the top cover of the cell. The explosion-proof valve, as a key safety design element in the battery system, primarily functions to: control pressure release within the cell, maintain the integrity and stability of the battery structure, assist in the operation of the cooling system, and protect occupants. The protective film maintains a stable internal chemical environment within the cell, assists in pressure relief by the explosion-proof valve, and improves the overall stability of the battery. The protective film prevents tiny particles from entering the valve, which could cause it to fail to open or malfunction, thus extending the lifespan of the explosion-proof valve and improving battery safety performance.

[0052] In addition, corresponding testing equipment needs to be installed to visually inspect the explosion-proof valve and its protective film, identifying the location of defects. This reduces the serious safety impact on the battery cell caused by defects such as damage to the explosion-proof valve and its protective film, electrolyte residue, and scratches. Therefore, accurately detecting and differentiating defects in explosion-proof valves and their protective films is a crucial aspect that the lithium battery automated testing industry needs to focus on.

[0053] In related technologies, no distinguishing measures are taken for the appearance inspection of explosion-proof valves and their protective films. Since the protective films of explosion-proof valves are generally made of transparent materials, current machine vision algorithms have difficulty distinguishing whether the defect is located above or below the protective film. They cannot effectively distinguish between the two types of defects, making it difficult to meet the high detection rate requirements of industrial production.

[0054] Based on this, this disclosure provides a detection system for explosion-proof valve assemblies. A first image acquisition device captures an image of the explosion-proof valve assembly, and a second image acquisition device captures an image of the same assembly, resulting in a second image. Defect detection is then performed on both images. If a defect is detected, its location within the explosion-proof valve assembly is determined through defect location detection. In this way, the first and second image acquisition devices capture images of the same explosion-proof valve assembly, and defects above or below the protective film of the explosion-proof valve are mapped onto the image. This not only allows for differentiation between defects located on the surface of the protective film and those on the surface of the explosion-proof valve, but also improves the accuracy and efficiency of explosion-proof valve assembly detection, thereby enhancing the safety of the battery cell under test.

[0055] The present disclosure will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0056] In one embodiment of this disclosure, a detection system for an explosion-proof valve assembly is provided. Figure 1 is a schematic diagram of the structural composition of a detection system for an explosion-proof valve assembly provided in this embodiment. As shown in Figure 1, the detection system 10 includes a processor 103, a first image acquisition unit 101, and a second image acquisition unit 102; wherein:

[0057] The first image acquisition unit 101 is used to acquire images of the explosion-proof valve assembly, obtain a first image, and send the first image to the processor 103;

[0058] The second image acquisition unit 102 is used to acquire images of the explosion-proof valve assembly, obtain a second image, and send the second image to the processor 103;

[0059] The processor 103 is used to receive a first image and a second image, and when it is determined that there is a defect in the first image and the second image, to perform defect location detection on the first image and the second image to determine the defect location of the explosion-proof valve assembly.

[0060] The explosion-proof valve assembly includes an explosion-proof valve and a protective film covering the explosion-proof valve.

[0061] In this embodiment of the disclosure, both the first image acquisition device 101 and the second image acquisition device 102 are devices capable of image acquisition, and the acquired images can be primary color (Red, Green, Blue, RGB) images. Exemplarily, both the first image acquisition device 101 and the second image acquisition device 102 can be any one of an image acquisition device with a rolling shutter, an image acquisition device with a global shutter, or an image acquisition device with a global reset shutter.

[0062] In this embodiment, an explosion-proof valve assembly is disposed on the top surface of the battery cell under test. The explosion-proof valve assembly includes an explosion-proof valve and a protective film covering the explosion-proof valve. The explosion-proof valve is used to control the gas pressure inside the battery cell under test, allowing limited gas release and preventing abnormal gas pressure in the battery cell. Exemplarily, the explosion-proof valve can be a mechanical valve or a negative temperature coefficient (NTC) thermistor valve, etc. The protective film of the explosion-proof valve is used to prevent external microparticles from getting stuck inside the explosion-proof valve. Exemplarily, the protective film of the explosion-proof valve can be a transparent plastic film made of polyethylene (PE) or polyethylene terephthalate (PET), possessing properties such as high temperature resistance and insulation, which can effectively protect electronic components.

[0063] In this embodiment of the disclosure, the processor may be located in the host computer.

[0064] In this embodiment, the first image and the second image are images of the same explosion-proof valve assembly taken from different shooting angles by the first image acquisition unit 101 and the second image acquisition unit 102, respectively. The first image acquisition unit 101 and the second image acquisition unit 102 can take pictures sequentially, or simultaneously to save shooting time, and then send the first image and the second image to the processor 103.

[0065] In this embodiment, the processor 103 performs defect identification on the first image and the second image. Specifically, the first image and the second image can be input into a neural network model, such as a Convolutional Neural Network (CNN), Faster RCNN, a semi-supervised learning method, or a weakly supervised learning method, or other algorithms capable of defect identification. Before performing defect identification, the algorithm can be trained by pre-inputting multiple images with defects, or by comparing the input first image and the second image with images of normal explosion-proof valve components, thereby identifying whether defects exist in the first image and the second image. For example, defects can be protective film damage, electrolyte leakage, scratches, dirt on the surface of the explosion-proof valve or the protective film, etc.

[0066] It should be noted that, since the first image and the second image are images taken from different perspectives of the same explosion-proof valve assembly from above (the first image acquisition device can be at different positions on the same horizontal plane or at different positions on different horizontal planes, depending on the defect location detection algorithm), one of the captured images also contains almost all the information of the defect of the explosion-proof valve assembly. Therefore, in some embodiments, the processor 103 can also perform defect identification based on one of the first image and the second image to determine whether there is a defect in the first image and the second image.

[0067] Alternatively, in some embodiments, when performing defect identification based on the first image and the second image, the processor 103 may perform a collaborative evaluation based on the results of defect identification performed on the two captured images respectively, according to the confidence level, thereby further improving the accuracy of defect identification.

[0068] In this embodiment, if the processor 103 determines that there are no defects in the first and second images, the battery cell to be tested is moved out of the current testing station and transferred to the next station. Alternatively, if the processor 103 determines that there are defects in the first and second images, that is, images that are different from those of a normal explosion-proof valve assembly, the defect location is further detected to determine the position of the defect in the explosion-proof valve assembly, and appropriate processing is performed based on the location of the defect. The defect may be located on the surface of the protective film or on the surface of the explosion-proof valve.

[0069] This disclosure provides a detection system that uses a first image acquisition device to capture an image of an explosion-proof valve assembly, and a second image acquisition device to capture an image of the same assembly, obtaining a second image. Defect detection is then performed on both the first and second images. If a defect is detected, its location within the explosion-proof valve assembly is determined through defect location detection. Thus, when the first and second image acquisition devices capture images of the same explosion-proof valve assembly, defects above or below the protective film of the explosion-proof valve are mapped onto the image. This not only allows for differentiation between defects located on the surface of the protective film and those on the surface of the explosion-proof valve based on the image, but also improves the detection accuracy and efficiency of the explosion-proof valve assembly, thereby enhancing the safety of the battery cell under test.

[0070] In some embodiments, referring again to FIG1, the first image acquisition element 101 and the second image acquisition element 102 are on the same horizontal plane, and the first image acquisition element 101 and the second image acquisition element 102 are symmetrically arranged above the explosion-proof valve assembly.

[0071] In this embodiment of the disclosure, the first image acquisition unit 101 and the second image acquisition unit can be arranged on the same horizontal plane and symmetrically arranged in the vertical direction with the explosion-proof valve assembly as the center.

[0072] In this embodiment, the first image acquisition unit 101 and the second image acquisition unit 102 are on the same horizontal plane and positioned at different locations above the explosion-proof valve assembly. The first image acquisition unit 101 captures an image of the explosion-proof valve assembly to obtain a first image; the second image acquisition unit 102 captures an image of the explosion-proof valve assembly to obtain a second image. Thus, the acquired first and second images are images of the explosion-proof valve assembly from different angles, and the first and second images exhibit parallax relative to the explosion-proof valve assembly. Further, the first image acquisition unit 101 sends the first image to the processor 103, and the second image acquisition unit 102 sends the second image to the processor 103. The processor 103 can determine the location of the defect based on this parallax, as detailed in the following embodiments.

[0073] This disclosure provides a detection system that uses a first image acquisition unit and a second image acquisition unit, which are symmetrically arranged on the same horizontal plane, to capture images of the same explosion-proof valve assembly, obtaining a first image and a second image respectively. In this way, based on the parallax principle, the location of a defect can be determined from the first and second images, thereby distinguishing whether the defect is located on the surface of the protective film or the surface of the explosion-proof valve.

[0074] In another embodiment of this disclosure, Figure 2 is a schematic diagram of the composition structure of a detection system for an explosion-proof valve assembly provided in this embodiment. As shown in Figure 2, the detection system 10 further includes a light source assembly 104, which is used to turn on or off based on a control signal sent by the processor 103.

[0075] The first image acquisition unit 101 is used to acquire a first image when the light source assembly is turned on;

[0076] The second image acquisition unit 102 is used to acquire a second image when the light source assembly is turned on.

[0077] In this embodiment, the symmetrical center positions between the first image acquisition unit 101 and the second image acquisition unit 102, the center position of the light source assembly 104, and the center position of the explosion-proof valve assembly 106 are sequentially coaxially arranged in the vertical direction. During installation, the vertical positional deviations of the symmetrical center positions between the first image acquisition unit 101 and the second image acquisition unit 102, the center position of the light source assembly 104, and the center position of the explosion-proof valve assembly 106 meet a preset deviation range. For example, the preset deviation range can be an allowable deviation of ±5 millimeters (mm). That is, the symmetrical center positions between the first image acquisition unit 101 and the second image acquisition unit 102, the center position of the light source assembly 104, and the center position of the explosion-proof valve assembly 106 are aligned in the vertical direction (allowable deviation ±5mm). Furthermore, the first image acquisition unit 101 and the second image acquisition unit 102 are located on the same horizontal plane, and the horizontal plane containing the first image acquisition unit 101 and the second image acquisition unit 102 is perpendicular to the aforementioned vertical direction.

[0078] As shown in Figure 2, an explosion-proof valve assembly 106 is disposed on the top surface of the battery cell 105 to be tested. The explosion-proof valve assembly 106 includes an explosion-proof valve 1062 and a protective film 1061 covering the explosion-proof valve. Additionally, the first image acquisition unit 101 includes a first body 1011 and a first lens 1012; the second image acquisition unit 102 includes a second body 1021 and a second lens 1022.

[0079] In this embodiment, the light source assembly 104 can be a planar coaxial light source assembly (or "planar shadowless light source assembly"). The lighting technology of the planar coaxial light source assembly has significant advantages in machine vision and industrial inspection. First, the planar coaxial light source assembly can provide very uniform illumination, reducing shadows and highlights. The diffuse reflection light of the planar coaxial light source assembly can eliminate the reflection problem caused by the local deformation of the explosion-proof valve 1062, which is particularly effective for detecting surface textures, micro-defects, or transparent objects. Uniform illumination helps improve image quality, making subsequent image processing and analysis more accurate. Furthermore, the lighting system design of the planar coaxial light source assembly is flexible, and the angle, intensity, and color of the light source can be adjusted according to different detection needs, making it suitable for the detection of various materials and surfaces. It should be noted that the planar coaxial light source assembly includes a special optical design with a semi-reflective and semi-transparent film, which can simultaneously filter the light rays incident on the object side and the image side, allowing it to be placed between two image acquisition devices and the explosion-proof valve assembly 106. This not only does not obstruct image capture but also makes the captured images clearer. It should be noted that in some embodiments, the first image acquisition device 101 and the second image acquisition device 102 may be equipped with a light source, which is lit up when the first image acquisition device 101 and the second image acquisition device 102 are taking pictures, and turned off after the pictures are taken.

[0080] On the production line for the battery cell 105 under test, the light source assembly 104 is controlled by the processor 103 to turn on or off. Generally, the light source assembly 104 is constantly lit during the production line's operation. The detection system 10 is set up at the detection station. After the processor detects that the battery cell under test has arrived at the detection station, it sends control commands to the first image acquisition unit 101 and the second image acquisition unit 102, controlling them to take vertical downward images of the explosion-proof valve assembly 106, obtaining a first image and a second image respectively, which are then transmitted to the processor 103. The processor 103 performs defect identification and location determination on the first and second images, and then determines the next processing step, such as moving the battery cell under test out of the detection station and transferring it to the next station, or removing it from the production line.

[0081] This disclosure provides a detection system in which, when the light source assembly is turned on, a first image acquisition unit and a second image acquisition unit capture images of the same explosion-proof valve assembly, respectively obtaining a first image and a second image. This improves the clarity of the acquired first and second images, thereby increasing the detection accuracy.

[0082] In another embodiment of this disclosure, Figure 3 is a schematic diagram of a battery cell to be tested provided in this embodiment. As shown in Figure 3, an explosion-proof valve assembly 106 is disposed on the surface of the battery cell 105 to be tested. The explosion-proof valve assembly 106 includes an explosion-proof valve and a protective film covering the explosion-proof valve. A first image acquisition unit and a second image acquisition unit are disposed above the explosion-proof valve assembly 106 on the surface of the battery cell 105 to be tested and are at the same horizontal plane. The explosion-proof valve assembly 106 is photographed to obtain a first image and a second image, respectively. To further improve the image quality, the first image acquisition unit and the second image acquisition unit can adopt different settings for different types of protective films.

[0083] Figure 4 is a schematic diagram of a setting method for a first image acquisition device and a second image acquisition device provided in an embodiment of this disclosure. As shown in Figure 4, when the protective film is in the first form, the first image acquisition device and the second image acquisition device are arranged and placed on the same horizontal plane along a first direction. The first form indicates that the difference between the length of the protective film in the first direction and the length of the protective film in the second direction is greater than a first threshold.

[0084] In this embodiment, the first direction can refer to a direction from left to right or from right to left along the long side of the upper surface of the battery cell to be tested; the second direction can refer to a direction from top to bottom or from bottom to top along the short side of the upper surface of the battery cell to be tested. In this disclosure and the following embodiments, as shown in FIG4, the example of the first direction being from left to right along the long side of the upper surface of the battery cell to be tested, and the example of the second direction being from bottom to top along the short side of the upper surface of the battery cell to be tested, will be used for illustration.

[0085] In this embodiment, the first form can refer to an elongated shape, wherein, in the first form, the length of the protective film in the first direction is greater than its length in the second direction, and the difference is greater than a first threshold. The first threshold can be specifically determined based on the range of the long side length and the range of the short side length of the protective film during actual classification. Alternatively, it can be determined whether the protective film is in the first form based on the ratio of the long side length to the short side length.

[0086] When the protective film is in its first configuration, the first image acquisition unit 101 and the second image acquisition unit 102, which are on the same horizontal plane, can be placed horizontally along the longer side of their elongated shape, in the first direction. This ensures that the first image acquisition unit 101 and the second image acquisition unit 102 can completely capture images of the protective film. Figure 4 is only a schematic diagram of the setup; the distance between the first image acquisition unit 101 and the second image acquisition unit 102 can be specifically determined based on parameters such as the length of the protective film and the distance between the two image acquisition devices and the protective film.

[0087] In this embodiment of the present disclosure, Figure 5 is a schematic diagram of the composition structure of a detection system for an explosion-proof valve assembly provided in this embodiment. As shown in Figure 5, the first image acquisition component 101 and the second image acquisition component 102 are fixed on the horizontal adjustment rod 107 of the image acquisition component. The two ends of the horizontal adjustment rod 107 are respectively fixed on the first adjustment rod bracket 1081 and the second adjustment rod bracket 1082. The first image acquisition component 101 and the second image acquisition component 102 can be fixed after the horizontal adjustment rod 107 is moved horizontally to the left or right to the target position. The first adjustment rod bracket 1081 and the second adjustment rod bracket 1082 are provided with longitudinal locking positions, so that the horizontal adjustment rod 107 can be moved up or down to the target position and fixed. In this way, when the protective film is in the first form, the first image acquisition component 101 and the second image acquisition component 102 can move in the left and right directions and up and down directions, as well as the image acquisition after fixing. In addition, their movement can be controlled by a processor.

[0088] It should also be noted that the light source assembly 104, the first adjustment rod bracket 1081, and the second adjustment rod bracket 1082 can all be fixed on the light source bracket 110.

[0089] This disclosure provides a detection system in which, when the protective film is in a first form, a first image acquisition unit and a second image acquisition unit are arranged on the same horizontal plane along a first direction. This avoids missing images of the explosion-proof valve assembly and thus improves the accuracy of defect detection.

[0090] In some embodiments, FIG6 is a schematic diagram of a setting method of a first image acquisition device and a second image acquisition device provided in the present disclosure. As shown in FIG6, when the protective film is in the second form, the first image acquisition device 101 and the second image acquisition device 102 are arranged and placed on the same horizontal plane along the second direction. The second form indicates that the difference between the length of the protective film in the first direction and the length of the protective film in the second direction is less than or equal to a first threshold.

[0091] In this embodiment of the disclosure, the second form may refer to a short and stout form. In the second form, the length of the protective film in the first direction may be slightly greater than, equal to, or even less than the length in the second direction, and the difference is less than or equal to a first threshold. Alternatively, whether the protective film is in the second form can be determined based on the ratio of the length of the long side to the length of the short side.

[0092] It should be noted that when the protective film is in its second form, the first image acquisition unit 101 and the second image acquisition unit 102, which are on the same horizontal plane, can be placed horizontally along the second direction, that is, along the shorter side of the shorter, thicker form. In this way, the first image acquisition unit 101 can capture, as much as possible, the image of the explosion-proof valve covered by the shadow of the protective film on the side where the second image acquisition unit 102 is located. Correspondingly, the second image acquisition unit 102 can also capture, as much as possible, the image of the explosion-proof valve covered by the shadow of the protective film on the side where the first image acquisition unit 101 is located. This reduces the impact of the protective film's shadow on the first image 101 and the second image 102, maximizing image quality.

[0093] It should also be noted that Figure 6 is only a schematic diagram of the setting method. The distance between the first image acquisition device 101 and the second image acquisition device 102 can be specifically determined according to parameters such as the width of the protective film and the distance between the two image acquisition devices and the protective film.

[0094] In this embodiment, referring to FIG5, the detection system 10 for the explosion-proof valve assembly further includes a first longitudinal support 1091 and a second longitudinal support 1092. The image acquisition component longitudinal adjustment rod, whose two ends are respectively fixed to the first longitudinal support 1091 and the second longitudinal support 1092, is not shown in the figure. The first image acquisition component 101 and the second image acquisition component 102 are fixed to the image acquisition component longitudinal adjustment rod in the same manner as described above, where they are fixed to the image acquisition component lateral adjustment rod 107. The first image acquisition component 101 and the second image acquisition component 102 can be moved left or right and fixed on the image acquisition component longitudinal adjustment rod. The first longitudinal support 1091 and the second longitudinal support 1092 can be moved up and down and fixed, causing the first image acquisition component 101 and the second image acquisition component 102 on the image acquisition component longitudinal adjustment rod to move up and down. Thus, image acquisition of the protective film can be achieved when the protective film is in its second form. The movement can be controlled by a processor.

[0095] This disclosure provides a detection system in which, when the protective film is in a second form, the first image acquisition unit and the second image acquisition unit are arranged on the same horizontal plane along a second direction. This can minimize the impact of the shadow area of ​​the protective film on the imaging effect, improve the imaging clarity, and thus improve the accuracy of defect detection.

[0096] In another embodiment of this disclosure, referring to Figures 1 and 2, the processor 103 is communicatively connected to the first image acquisition unit 101 and the second image acquisition unit 102, respectively; wherein:

[0097] Processor 103 is used to establish communication connections with first image acquisition unit 101 and second image acquisition unit 102 respectively based on a first preset protocol;

[0098] The first image acquisition unit 101 is used to encode the first image based on the second preset protocol, generate the encoded first image, and send the encoded first image to the processor 103 through a communication connection;

[0099] The second image acquisition unit 102 is used to encode the second image based on the second preset protocol, generate the encoded second image, and send the encoded second image to the processor 103 through a communication connection.

[0100] In this embodiment, the first preset protocol may be the GigE Vision Control Protocol (GVCP); the second preset protocol may be the GigE Vision Stream Protocol (GVSP). GVCP is used for controlling and configuring the first image acquisition device 101 and the second image acquisition device 102, allowing the processor 103 to send commands, such as trigger signals, to the first image acquisition device 101 and the second image acquisition device 102, and to receive status information from both devices. GVSP is used for transmitting image data, including encoded first and encoded second images, supporting real-time video streaming and ensuring high bandwidth and low latency.

[0101] It should be noted that the processor uses different calling interfaces for different image acquisition devices, depending on the calling interface provided by the image acquisition device, but the communication protocol used is generally GVCP.

[0102] The first image acquisition unit 101 and the second image acquisition unit 102 compress the first and second images based on video compression formats supported by a second preset protocol, such as JPEG or MJPEG. The specific compression format can be selected according to actual bandwidth requirements and image quality to optimize data transmission. The compressed first and second images are then sent to the processor. Upon receiving the compressed images, the processor 103 decodes them according to the second preset protocol to obtain the first and second images.

[0103] This disclosure provides a detection system in which a processor can receive, based on a first preset protocol, encoded first and second images from a first image acquisition device and a second image acquisition device, respectively. This improves the communication efficiency between the processor and the first and second image acquisition devices, thereby enhancing detection efficiency.

[0104] In another embodiment of this disclosure, the processor is further configured to determine depth information of the defect based on the pixel difference between the same point on the defect in the first image and the second image; and to determine, based on the depth information, whether the defect is located on the surface of the protective film in the explosion-proof valve assembly or on the surface of the explosion-proof valve.

[0105] It should be noted that binocular imaging involves two image acquisition devices simultaneously capturing the same scene from different angles. Specifically, in this embodiment, the first and second image acquisition devices capture images of the same explosion-proof valve assembly. Parallax information is used to calculate the depth and three-dimensional position of the explosion-proof valve assembly, and its core principle is based on triangulation. Figure 7 is a schematic diagram of a triangulation method provided in this embodiment. As shown in Figure 7, firstly, the first image acquisition device 101 and the second image acquisition device 102 need to be precisely positioned and calibrated to ensure that their relative positions and angles are known. This includes intrinsic parameter calibration (determining parameters such as focal length and optical center of each lens) and extrinsic parameter calibration (determining the relative distance and angle between the two image acquisition devices). For ease of calculation, as in the aforementioned embodiment, the first image acquisition device 101 and the second image acquisition device 102 can be set on the same horizontal plane. Secondly, the first image acquisition device 101 and the second image acquisition device 102 are controlled to simultaneously capture the same object. Due to the different viewing angles, the images captured by each image acquisition device will have certain differences; this difference is called parallax. For example, in Figure 7, the length of an object in the photographed subject is L3. The image length of this object captured by the first image acquisition unit 101 is L1, and the image length of this object captured by the second image acquisition unit 102 is L2. The difference between L1 and L2 is called the parallax between the image captured by the first image acquisition unit 101 and the image captured by the second image acquisition unit 102. Although the different images have different viewpoints, the objects photographed in the two images are the same. Therefore, corresponding feature points can be found in the two images. These feature points represent different viewpoints of the same object in space. As mentioned above, the same object can be identified and determined in different images through feature point matching. Feature point matching is a key step in binocular imaging. Commonly used methods include scale-invariant feature transform (SIFT) algorithm, speed-up robust features (SURF) algorithm, and oriented fast and rotated brief (ORB) algorithm, etc., for feature detection and matching. By comparing the positional differences of corresponding feature points in two image acquisition devices, the parallax between the first and second images is calculated. This can also be referred to as the pixel difference of the same feature point on the defect in the first and second images. The pixel difference reflects the positional difference of the explosion-proof valve assembly in the two image acquisition devices, and this positional difference is directly related to the depth information of the explosion-proof valve assembly.

[0106] In this embodiment of the disclosure, it should be understood that since the protective film of the explosion-proof valve is made of transparent material, if a defect appears in the first and second images, the defect may be on the surface of the transparent protective film of the explosion-proof valve, or it may be on the surface of the explosion-proof valve below the transparent protective film. Since the surface of the explosion-proof valve and the surface of the protective film of the explosion-proof valve have different depths, based on the aforementioned principle, this depth information is mapped into the first and second images. The pixel difference determined based on the same feature point in the two captured images is also different. Therefore, the location of the defect in the explosion-proof valve assembly can be determined based on the pixel difference between the first and second images.

[0107] The process of determining the pixel difference between the first image and the second image is described below. In this embodiment of the present disclosure, Figure 8 is a schematic diagram of a first image and a second image provided in this embodiment of the present disclosure. As shown in Figure 8(a), a first pixel point P2 and a first reference point P1 in the first image are first determined; as shown in Figure 8(b), further through feature point matching, a second pixel point P4 corresponding to the previously determined first pixel point P2 and a second reference point P2 corresponding to the first reference point P1 in the second image are determined. The first pixel point P2 and the second pixel point P4 are different images of the same point in the defect on the explosion-proof valve assembly on different image acquisition devices; the first reference point P1 and the second reference point P3 are different images of the same selected reference point on the explosion-proof valve assembly on different image acquisition devices.

[0108] In this embodiment, the first reference point P1 and the second reference point P3 are used as origins to provide a reference for calibrating the first pixel P2 and the second pixel P4 in the captured image. It should be noted that the selection of the reference point can be determined according to actual needs. For example, the leftmost pixel on the surface of the protective film of the explosion-proof valve in the explosion-proof valve assembly can be selected as the reference point; alternatively, the leftmost pixel on the surface of the explosion-proof valve under the protective film can also be selected as the reference point. No limitation is made here.

[0109] Further, based on Figure 8(a), a first length is determined from the first reference point P1 to the first reference point P1 in the first image, according to the first reference point P1 and the first pixel point P2. And based on Figure 8(b), a second length is determined from the second reference point P3 to the second pixel point P4 in the second image. Finally, the absolute value of the difference between the first length and the second length is taken as the pixel difference between the same point corresponding to the first pixel point P2 and the second pixel point P4 on the defect in the first image and the second image.

[0110] The pixel difference between the first and second images reflects the positional difference of the explosion-proof valve assembly in the two image acquisition devices. This positional difference is directly related to the depth information of the explosion-proof valve assembly. As mentioned earlier, when two image acquisition devices at different horizontal positions at the same height capture images of objects at different heights, parallax occurs. When this parallax is mapped onto the first and second images, it causes different lengths of the same point in the defect relative to the reference point during imaging, thus affecting the pixel difference of the same point in the defect between the first and second images. For example, when the positions of the first and second reference points are on the same plane as the positions corresponding to the first and second pixels, the pixel difference of the same point corresponding to the first and second pixels in the first and second images is small to approach zero; when the positions of the first and second reference points are not on the same plane as the positions corresponding to the first and second pixels, the pixel difference of the same point corresponding to the first and second pixels in the first and second images is large. Thus, the processor can determine the location of the defect based on the pixel difference of the same point corresponding to the first and second pixels in the first and second images, and the relative positional relationship between the defect and the reference point on the explosion-proof valve assembly.

[0111] For example, in some embodiments, if the positions corresponding to the first reference point and the second reference point are points on the surface of the explosion-proof valve protective film, and the pixel difference between the first pixel point and the second pixel point corresponding to the same point is small in the first image and the second image, then the defect is on the surface of the explosion-proof valve protective film; otherwise, if the pixel difference between the first pixel point and the second pixel point corresponding to the same point is large in the first image and the second image, then the defect is below the explosion-proof valve protective film, that is, on the surface of the explosion-proof valve.

[0112] Alternatively, in some embodiments, if the positions corresponding to the first reference point and the second reference point are points on the surface of the explosion-proof valve, and the pixel difference between the same point corresponding to the determined first pixel point and the second pixel point in the first image and the second image is large, the defect is on the surface of the explosion-proof valve protective film; otherwise, if the pixel difference between the same point corresponding to the determined first pixel point and the second pixel point in the first image and the second image is small, the defect is below the explosion-proof valve protective film, that is, on the surface of the explosion-proof valve.

[0113] It should be noted that when the pixel difference between the first and second pixels at the same point corresponds to a first pixel and the first image and the second image meets a first preset range, the defect is determined to be on the surface of the protective film; when the pixel difference between the first and second pixels at the same point corresponds to a first pixel and the second image and the first image and the second image meets a second preset range, the defect is determined to be on the surface of the explosion-proof valve. The first and second preset ranges are specifically determined based on multiple prior experiments.

[0114] In some embodiments, taking the position corresponding to the first reference point and the second reference point as a point on the surface of the explosion-proof valve protective film as an example, if the pixel difference between the same point corresponding to the first pixel point and the second pixel point in the first image and the second image is less than or equal to a preset threshold, then it is determined that the defect is on the surface of the explosion-proof valve protective film; otherwise, if the pixel difference between the same point corresponding to the first pixel point and the second pixel point in the first image and the second image is greater than the preset threshold, then it is determined that the defect is below the explosion-proof valve protective film, that is, on the surface of the explosion-proof valve.

[0115] In some embodiments, taking the position corresponding to the first reference point and the second reference point as a point below the explosion-proof valve protective film, i.e., a point on the surface of the explosion-proof valve, as an example, if the pixel difference between the same point corresponding to the first pixel point and the second pixel point in the first image and the second image is less than or equal to a preset threshold, then the defect is determined to be below the explosion-proof valve protective film; otherwise, if the pixel difference between the same point corresponding to the first pixel point and the second pixel point in the first image and the second image is greater than the preset threshold, then the defect is determined to be above the explosion-proof valve protective film. It should be noted that the preset threshold can be specifically determined according to the parameters of different image acquisition devices, the resolution of the captured images, etc. For example, the preset threshold can be 15 pixels or 10 pixels.

[0116] This disclosure provides a detection system in which a processor determines the depth information of a defect based on the pixel difference between the same point on a first image and a second image, and further determines the location of the defect. Thus, defects of different heights correspond to different depth information, thereby enabling the differentiation between defects on the surface of a protective film and those on the surface of an explosion-proof valve, improving the accuracy of defect detection.

[0117] In some embodiments, the processor is further configured to detect defects based on a first image and / or a second image, determine the characteristic parameters of the defects, and determine whether the explosion-proof valve assembly is abnormal based on the defect location and characteristic parameters of the explosion-proof valve assembly.

[0118] In the embodiments of this disclosure, the characteristic parameters of the defect may include a variety of parameters, such as the area of ​​the defect, the type of the defect, etc.

[0119] In some embodiments, the processor can input one of the first image and the second image into the defect detection algorithm, which analyzes the captured image and outputs the defect category and area. Understandably, there can be many defect categories, such as damage, electrolyte residue, scratches, dirt, and missing protective film on explosion-proof valves. The defect detection algorithm can classify defects based on their shape, grayscale, etc., to obtain the defect features corresponding to different defect categories.

[0120] For example, the defect characteristics of a scratch can be strip-shaped, the defect characteristics of electrolyte leakage can be sheet-like and shiny, and the defect characteristics of dirt can be sheet-like and black, etc. In other words, different defect categories have different defect characteristics. In addition, for different image acquisition devices, the correspondence between each pixel in the captured image and the actual area should be different. The defect detection algorithm can also determine this correspondence based on the parameters of the image acquisition device, and further calculate and determine the area of ​​the defect.

[0121] Alternatively, in some embodiments, the processor can input the first image and the second image into the defect detection algorithm, respectively, so that the defect detection algorithm can determine the category and area of ​​the defect corresponding to the first image, and the defect category and area corresponding to the second image. Since the defect detection algorithm is pre-trained on machine learning using labeled sample images of various categories, furthermore, after determining the results corresponding to the two captured images, the defect detection algorithm performs a collaborative evaluation, and determines the specific category of the defect and its area based on the confidence evaluation of the two results.

[0122] In this embodiment of the disclosure, after determining the location of the defect, its characteristic parameters, such as category and area, can be further determined. Finally, by combining multiple parameters of the defect, it can be determined whether the appearance of the explosion-proof valve assembly is abnormal. It should be understood that other parameters of the defect can also be determined using a defect detection algorithm to determine whether the appearance of the explosion-proof valve assembly is abnormal.

[0123] It should be noted that the handling methods for defects of the same type may differ depending on their location. For example, if the defect is dirt, and it is located above the protective film of the explosion-proof valve, then the explosion-proof valve assembly is considered defect-free. Otherwise, if the defect is located below the protective film of the explosion-proof valve, then the defect needs to be addressed.

[0124] Furthermore, for defects of the same location and category, differences in area can also affect defect identification, i.e., defect detection results. For example, a small area can be ignored, while a large area indicates that the explosion-proof valve assembly has failed the inspection.

[0125] This disclosure provides a detection system that, based on a first image and / or a second image, determines the characteristic parameters of a defect, and further determines whether an explosion-proof valve assembly is abnormal based on the defect's characteristic parameters and location. In this way, by comprehensively considering various parameters of the defect, it is possible to determine whether the defect in the explosion-proof valve assembly is abnormal, thus improving the accuracy and precision of the detection.

[0126] In some embodiments, the processor is further configured to determine that the appearance of the protective film is abnormal when it is determined that the characteristic parameters of the defect meet the first characteristic condition; or determine that the appearance of the protective film is not abnormal when it is determined that the characteristic parameters of the defect do not meet the first characteristic condition.

[0127] In the embodiments of the present disclosure, the first characteristic condition may include multiple judgment conditions for different characteristic parameters. As described above, the characteristic parameters may include the defect area, defect category, etc. Based on this, the first characteristic condition includes the judgment condition for the defect area, for example, whether the area of the defect is greater than the threshold allowed for the area defect, and may also include the judgment condition for the defect category, for example, whether the category of the defect belongs to certain categories that do not need to be processed, such as dirt, or categories that must be processed, such as breakage. Further, based on the judgment conditions of multiple characteristic parameters, it is determined whether the appearance of the protective film is abnormal.

[0128] In the embodiments of the present disclosure, when it is determined that the defect is on the surface of the protective film of the explosion-proof valve, different defect types correspond to different thresholds allowed for the area defect, and the category of the defect also affects the determination of the defect detection result, that is, based on the judgment conditions of the defect area and defect category, it is determined whether the appearance of the protective film is abnormal. Among them, the threshold allowed for the area defect may refer to the maximum area that does not affect the use of the explosion-proof valve component, and is specifically set according to the detection accuracy requirements, and will not be specifically limited here.

[0129] Exemplarily, when the area of the defect is less than or equal to the threshold allowed for the area defect, it indicates that the area of the defect is relatively small and the defect can be ignored. Or, when the category of the defect is the first category, for example, the first category may include defects such as dirt that will be processed in subsequent processes. Even if the area of the defect is greater than the threshold allowed for the area defect, the defect can also be ignored. Then the result of this defect detection is qualified, that is, the appearance of the protective film of the explosion-proof valve is not abnormal.

[0130] Exemplarily, when the defect does not belong to the first category and the area of the defect is greater than the threshold allowed for the area defect, for example, the defect category includes defects such as pits, breakages, loss of the protective film of the explosion-proof valve, the protective film is pasted crookedly, scratches on the surface of the protective film, etc. that require repair or replacement of the protective film, and the area of the defect is greater than the threshold allowed for the area defect, it indicates that the area of the defect is relatively large and the defect cannot be ignored. Then the result of this defect detection is unqualified, that is, the appearance of the protective film of the explosion-proof valve is abnormal.

[0131] This disclosure provides a detection system in which, when a processor determines that a defect is located on the surface of a protective film, it can determine whether the appearance of the protective film of the explosion-proof valve is abnormal based on whether the characteristic parameters of the defect meet a first characteristic condition. This improves the accuracy of detecting the protective film of the explosion-proof valve.

[0132] In some embodiments, the processor is further configured to determine that the appearance of the explosion-proof valve is abnormal when the characteristic parameters of the defect meet the second characteristic condition; or to determine that the appearance of the explosion-proof valve is not abnormal when the characteristic parameters of the defect do not meet the second characteristic condition.

[0133] In this embodiment of the disclosure, referring to the content of the first feature condition mentioned above, the second feature condition may also include judgment conditions set for multiple feature parameters, such as whether the area of ​​the defect is greater than or equal to the threshold allowed for area defects, or whether the category of the defect belongs to the category that needs to be processed, such as dirt.

[0134] In this embodiment, when the defect is determined to be located below the protective film of the explosion-proof valve, i.e., on the surface of the explosion-proof valve, the type of defect can include electrolyte leakage, dirt, damage, dents, scratches on the surface of the explosion-proof valve, etc., which are generally non-negligible defects. Therefore, it can be determined whether the defect can be ignored by comparing the area of ​​the defect with the allowable area defect. The second set threshold can be set according to the detection accuracy requirements and is not specifically limited here.

[0135] For example, if the area of ​​a defect is less than or equal to the allowable threshold for area defects, it indicates that the area of ​​the defect is relatively small and can be ignored. In this case, the defect detection result is qualified, meaning that the appearance of the explosion-proof valve is not abnormal.

[0136] For example, if the area of ​​a defect exceeds the allowable threshold for area defects, it indicates that the area of ​​the defect is relatively large and cannot be ignored. In this case, the defect detection result is unqualified, meaning that the appearance of the explosion-proof valve is abnormal.

[0137] This disclosure provides a detection system in which, when a processor determines that a defect is located on the surface of an explosion-proof valve, it can determine whether the appearance of the explosion-proof valve is abnormal based on whether the characteristic parameters of the defect meet a second characteristic condition. This improves the accuracy of detecting explosion-proof valves.

[0138] In another embodiment of this disclosure, the type of the light source assembly, the type of the first image acquisition device, and the type of the second image acquisition device are related to the shooting time.

[0139] In some embodiments, when the type of the light source component is planar coaxial light and the types of the first image acquisition device and the second image acquisition device are roller shutter image acquisition devices, the planar coaxial light can be kept constantly lit. The first image acquisition device and the second image acquisition device need to use a fixed shooting method to acquire images. That is, when the battery cell to be tested on the production line flows to the bottom of the first image acquisition device and the second image acquisition device in the testing station, it needs to be paused for a period of time to take pictures. In this case, the shooting time is relatively long, for example, 1000us, in order to obtain a clear first image and second image.

[0140] In some embodiments, when the type of the light source component is a strobe light source and the type of the first image acquisition unit and the type of the second image acquisition unit are global image acquisition units, in this case, if it is kept constantly lit or a planar coaxial light is used, motion blur may occur. Therefore, the strobe light source is only turned on when shooting. The first image acquisition unit and the second image acquisition unit can use a flying shooting method to acquire images. That is, when the battery cell to be tested on the production line flows to the bottom of the first image acquisition unit and the second image acquisition unit in the inspection station, a clear first image and a second image can be obtained without pausing. In this case, the shooting time is short, for example, 100us, which greatly shortens the shooting time and improves the defect detection efficiency on the production line.

[0141] This disclosure provides a detection system that can select different types of light source components, a first image acquisition device, and a second image acquisition device according to the shooting time requirements, thereby improving the flexibility and applicability of the detection system.

[0142] In another embodiment of this disclosure, a detection system for an explosion-proof valve assembly based on the foregoing embodiments is described in Figure 9, which is a schematic diagram of the composition structure of a detection system for an explosion-proof valve assembly provided in this embodiment. As shown in Figure 9, the detection system includes a first image acquisition unit 101, a second image acquisition unit 102, a light source assembly 104, a protective film for the explosion-proof valve, and the explosion-proof valve itself. The first image acquisition unit 101 and the second image acquisition unit 102 are on the same horizontal plane and are positioned above the light source assembly.

[0143] The center of symmetry between the first image acquisition unit 101 and the second image acquisition unit 102 is aligned with the center of the light source assembly 104 (allowable deviation ±5mm). When the battery cell to be tested moves to the photo-taking point on the testing station, the center of symmetry between the first image acquisition unit 101 and the second image acquisition unit 102, the center of the light source assembly 104, and the center of the explosion-proof valve assembly are aligned (allowable deviation ±5mm). The light source assembly 104 is pre-set to be lit, and the first image acquisition unit 101 and the second image acquisition unit 102 take pictures simultaneously. Finally, the photo-taking ends, the battery cell to be tested is moved away, and the processor acquires the first and second images and processes them based on the algorithm.

[0144] The light source component 104 can be a planar coaxial light source. Planar coaxial lighting technology has significant advantages in machine vision and industrial inspection. First, coaxial lighting can provide very uniform illumination, reducing shadows and highlights, which is particularly effective for detecting surface textures, minute defects, or transparent objects. Uniform illumination helps improve image quality, making subsequent image processing and analysis more accurate. Furthermore, the coaxial lighting system is flexible in design, allowing the angle, intensity, and color of the light source to be adjusted according to different inspection needs, making it suitable for the inspection of various materials and surfaces.

[0145] Furthermore, depending on the shape of the protective patch for different explosion-proof valves, the first image acquisition unit 101 and the second image acquisition unit 102 can be arranged in different ways: For the first shape, that is, the protective patch for a long and thin explosion-proof valve, the first image acquisition unit 101 and the second image acquisition unit 102 can be placed horizontally along the long side of the protective patch. Alternatively, for the second shape, that is, the protective patch for a short and thick explosion-proof valve, the first image acquisition unit 101 and the second image acquisition unit 102 can be placed vertically along the width of the protective patch to avoid the influence of shadow areas on the imaging effect.

[0146] In another embodiment of this disclosure, based on the defect detection method for explosion-proof valve assemblies of the foregoing embodiments, Figure 10 is a detailed flowchart of a detection method for explosion-proof valve assemblies provided by an embodiment of this disclosure. This method can be applied to the aforementioned detection system. As shown in Figure 10, the detailed flowchart of the defect detection method for the explosion-proof valve assembly may include:

[0147] S201, acquire the first image and the second image.

[0148] It's important to note that binocular imaging uses two image acquisition devices to simultaneously capture the same scene from different angles, utilizing parallax information to calculate the depth and 3D position of objects. Its core principle is based on triangulation. First, precise positioning and calibration of the two image acquisition devices are necessary to ensure their relative positions and angles are known. This includes intrinsic parameter calibration (determining parameters such as focal length and optical center of each lens) and extrinsic parameter calibration (determining the relative distance and angle between the two image acquisition devices). Second, since the two image acquisition devices simultaneously capture the same scene, the images captured by each device will differ due to the different viewpoints; this difference is called parallax. Corresponding feature points are found in the two images; these feature points represent different viewpoints of the same object in space. Feature point matching is a crucial step in binocular imaging, and commonly used methods include feature detection and matching algorithms such as SIFT, SURF, and ORB. By comparing the positional differences of corresponding feature points in the two image acquisition devices, the parallax value is calculated. The parallax value reflects the positional difference of the object in the two image acquisition devices and is directly related to the object's depth information. Next, using the parallax value and known image acquisition device and lens parameters (such as focal length and baseline distance), the depth information of each feature point in space is calculated through triangulation, thereby generating a depth map of the entire scene. Finally, based on the depth map, 3D reconstruction can be further performed, converting the 2D image into a 3D model, enabling a 3D understanding of the scene or object.

[0149] In this embodiment of the disclosure, a first image acquisition unit and a second image acquisition unit at different horizontal positions at the same height acquire images of the explosion-proof valve assembly, respectively obtaining a first image and a second image, and transmit them to the processor.

[0150] In addition, regarding the selection of equipment type, images of explosion-proof valves and protective films can be captured using a fixed-shot method, employing planar coaxial light and roller shutter image acquisition devices; alternatively, images of explosion-proof valves and protective films can be captured using a flying-shot method, employing strobe light sources and global image acquisition devices to improve detection efficiency.

[0151] S202, Image preprocessing.

[0152] S203, whether the pixel difference of the same point on the defect in the first image and the second image meets the second preset range.

[0153] If the pixel difference between the same point on the defect in the first image and the second image meets the first preset range, then the defect is above the protective film; if the pixel difference between the same point on the defect in the first image and the second image meets the second preset range, then the defect is below the protective film.

[0154] In this embodiment of the disclosure, when the first image acquisition device and the second image acquisition device at different horizontal positions at the same height capture images of objects at different heights, a viewing angle difference will be generated. When this parallax is mapped onto the imaging of the two image acquisition devices, parallax will be generated. Therefore, defects on and under the membrane will also have different heights. The same point on the defect will generate a pixel difference in the first image and the second image. Then, by calibration and algorithm, the two images are fused to determine whether the pixel difference of the same point in the first image and the second image meets the second preset range, thereby distinguishing defects on and under the membrane.

[0155] S204, the defect is located on the surface of the explosion-proof valve.

[0156] S205, Defect Classification.

[0157] S206, whether the area of ​​the defect is greater than the first set threshold.

[0158] In this embodiment of the disclosure, if the defect is located on the surface of the explosion-proof valve, the defect category needs to be determined based on defect classification. If the area of ​​the defect is greater than a first set threshold, the characteristic parameters of the defect are determined to meet the second characteristic condition, and the appearance of the explosion-proof valve is abnormal. Alternatively, if one of them is not met, the characteristic parameters of the defect are determined not to meet the second characteristic condition, and the appearance of the explosion-proof valve is not abnormal.

[0159] The first set threshold is the maximum area of ​​allowable defects on the surface of the explosion-proof valve, which can be determined specifically according to the defect type.

[0160] S207, the defect is located on the surface of the protective film.

[0161] S208, Defect Classification.

[0162] S209, whether the area of ​​the defect is greater than the second set threshold.

[0163] In this embodiment of the disclosure, if the defect is located on the surface of the protective film, and the defect category is determined to be not a category that does not need to be considered, such as dirt, based on defect classification, and the area of ​​the defect is greater than a second set threshold, then the characteristic parameters of the defect are determined to meet the first characteristic condition, and the appearance of the protective film is abnormal; or, if one of them is not met, then the characteristic parameters of the defect are determined not to meet the first characteristic condition, and the appearance of the protective film is not abnormal.

[0164] The second set threshold is the maximum area of ​​defects allowed on the surface of the protective film, which can be determined specifically according to the defect type.

[0165] This disclosure provides a defect detection method for explosion-proof valve assemblies. It utilizes a first image acquisition unit and a second image acquisition unit at different horizontal positions at the same height for image acquisition, effectively distinguishing whether the defect is on the surface of the protective film or the surface of the explosion-proof valve. Furthermore, it employs a planar coaxial light source assembly to eliminate reflections caused by deformation of the explosion-proof valve assembly. Additionally, it utilizes a rapid imaging method to increase the equipment's detection cycle time, thereby improving detection efficiency.

[0166] It should be understood that those skilled in the art will recognize that this disclosure may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0167] It should also be understood that the phrase "an embodiment" or "one embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this disclosure. Therefore, "in one embodiment" or "one embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this disclosure, the sequence numbers of the above steps / processes do not imply a sequential order of execution; the execution order of each step / process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this disclosure. The sequence numbers of the above embodiments of this disclosure are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0168] It should be noted that, in this disclosure, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0169] In the several embodiments provided in this disclosure, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components may be combined, or integrated into another system, or some features may be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0170] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the embodiments of this disclosure, all functional units may be integrated into one processing unit, or each unit may be a separate unit, or two or more units may be integrated into one unit; the integrated unit may be implemented in hardware or in a combination of hardware and software functional units.

[0171] The above are merely preferred embodiments of this disclosure and are not intended to limit the scope of protection of this disclosure. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A detection system for an explosion-proof valve assembly, the detection system comprising a processor, a first image acquisition unit, and a second image acquisition unit; wherein: The first image acquisition unit is used to acquire images of the explosion-proof valve assembly, obtain a first image, and send the first image to the processor; The second image acquisition unit is used to acquire images of the explosion-proof valve assembly, obtain a second image, and send the second image to the processor; The processor is configured to receive the first image and the second image, and when it is determined that there is a defect in the first image and the second image, to perform defect location detection on the first image and the second image to determine the defect location of the explosion-proof valve assembly; wherein, the explosion-proof valve assembly includes an explosion-proof valve and a protective film covering the explosion-proof valve.

2. The detection system according to claim 1, wherein, in: The first image acquisition device and the second image acquisition device are on the same horizontal plane, and the first image acquisition device and the second image acquisition device are symmetrically arranged above the explosion-proof valve assembly.

3. The detection system according to claim 1 or 2, wherein, The detection system also includes a light source assembly, which is used to turn on or off based on a control signal sent by the processor. The first image acquisition device is used to acquire the first image when the light source assembly is turned on; The second image acquisition device is used to acquire the second image when the light source assembly is turned on.

4. The detection system according to any one of claims 1-3, wherein, When the protective film is in the first form, the first image acquisition device and the second image acquisition device are arranged on the same horizontal plane along the first direction. The first form indicates that the difference between the length of the protective film in the first direction and the length of the protective film in the second direction is greater than a first threshold.

5. The detection system according to claim 4, wherein, When the protective film is in the second form, the first image acquisition device and the second image acquisition device are arranged on the same horizontal plane along the second direction. The second form indicates that the difference between the length of the protective film in the first direction and the length of the protective film in the second direction is less than or equal to a first threshold.

6. The detection system according to any one of claims 1-5, wherein, The processor is communicatively connected to both the first image acquisition device and the second image acquisition device; wherein: The processor is configured to establish communication connections with the first image acquisition device and the second image acquisition device respectively based on a first preset protocol; The first image acquisition device is used to encode the first image based on a second preset protocol to generate an encoded first image, and send the encoded first image to the processor through the communication connection; The second image acquisition device is used to encode the second image based on a second preset protocol, generate an encoded second image, and send the encoded second image to the processor through the communication connection.

7. The detection system according to claim 6, wherein, The first preset protocol is used to control and configure the first image acquisition device and the second image acquisition device; the second preset protocol is used to transmit the encoded first image and the encoded second image.

8. The detection system according to any one of claims 1-7, wherein, The processor is further configured to determine the depth information of the defect based on the pixel difference between the same point on the defect in the first image and the second image; Based on the depth information, it is determined whether the defect is located on the surface of the protective film in the explosion-proof valve assembly or on the surface of the explosion-proof valve.

9. The detection system according to claim 8, wherein, Feature point matching is performed on the first image and the second image to determine the same point on the defect.

10. The detection system according to claim 8 or 9, wherein, The processor is further configured to detect the defect based on the first image and / or the second image, determine the characteristic parameters of the defect, and determine whether the explosion-proof valve assembly is abnormal based on the defect location of the explosion-proof valve assembly and the characteristic parameters.

11. The detection system according to claim 10, wherein, The processor is further configured to determine that the appearance of the protective film is abnormal when the characteristic parameters of the defect meet the first characteristic condition; or to determine that the appearance of the protective film is not abnormal when the characteristic parameters of the defect do not meet the first characteristic condition.

12. The detection system according to claim 10, wherein, The processor is further configured to determine that the appearance of the explosion-proof valve is abnormal when the characteristic parameters of the defect meet the second characteristic condition; or to determine that the appearance of the explosion-proof valve is not abnormal when the characteristic parameters of the defect do not meet the second characteristic condition.

Images