Battery seal nail leakage detection method, device and computing device

Abstract

The application discloses a battery sealing nail liquid leakage detection method and device and computing equipment. The method comprises the following steps: obtaining a first image of a liquid leakage detection area of a battery before pressurization, and obtaining a second image of the liquid leakage detection area of the battery after pressurization, wherein the liquid leakage detection area of the battery comprises a sealing nail, a welding seam and a cleaning area; calculating a residual image of the first image and the second image; performing target detection on the residual image by using an intelligent target detection model to locate a liquid leakage area in the residual image, wherein the liquid leakage area comprises a highlight area and a shadow area; performing segmentation calculation on the liquid leakage area to determine an actual contour of the liquid leakage area; and determining whether the sealing nail and the welding seam of the battery after pressurization cause liquid leakage due to defects according to the actual contour of the liquid leakage. Based on the above, liquid leakage caused by external defects and internal defects of the sealing nail and the welding seam can be detected, and the detection precision and the detection efficiency are improved.

Description

Technical Field

[0001] This invention relates to the fields of visual inspection and image processing technology, and in particular to a method, device and computing equipment for detecting leakage of battery sealing nails. Background Technology

[0002] Battery sealing pins are crucial components used to seal the electrolyte filling holes and ensure the long-term sealing of the battery casing after electrolyte filling. Common battery sealing pins are round aluminum caps with protrusions. The manufacturing and welding quality of the sealing pins determines the battery's sealing performance, safety, and lifespan; therefore, sealing pin welding is the final critical process before cell molding.

[0003] Battery seal leakage refers to the seepage of electrolyte along the injection hole. Leakage not only corrodes the module but also poses safety risks such as lithium plating and short circuits. The main causes of battery seal leakage include through-cracks in the weld, pinholes in the weld, weld bursts, and dislodged seals.

[0004] Currently, the main solutions for quality inspection of battery sealing nails include surface visual inspection, CT inspection, and helium testing (helium mass spectrometry leak detection). Among these, surface visual inspection cannot detect internal defects in the weld, such as implosion. CT inspection is very time-consuming and its efficiency is far lower than that of battery production. Although battery helium testing has extremely high sensitivity, its equipment, gas, and recovery costs are high, and its efficiency is low, making it difficult to integrate with high-production lines.

[0005] Therefore, a method for detecting leakage in battery sealing pins is needed to solve the problems existing in the above-mentioned technical solutions. Summary of the Invention

[0006] Therefore, the present invention provides a method and device for detecting leakage of battery sealing pins, so as to solve or at least alleviate the above-mentioned problems.

[0007] According to one aspect of the present invention, a method for detecting leakage in a battery sealing pin is provided, executed in a computing device, comprising: acquiring a first image of a leakage detection area of ​​the battery before pressurization, and acquiring a second image of the leakage detection area of ​​the battery after pressurization, wherein the leakage detection area of ​​the battery includes a sealing pin, a weld, and a cleaning area; calculating a residual image between the first image and the second image; performing target detection on the residual image using an intelligent target detection model to locate a leakage area in the residual image, the leakage area including a highlight area and a shadow area; performing segmentation calculation on the leakage area to determine the actual leakage contour in the leakage area; and determining, based on the actual leakage contour, whether the sealing pin and weld of the pressurized battery are leaking due to defects.

[0008] Optionally, in the battery sealing pin leakage detection method according to the present invention, determining whether the sealing pin and weld of the pressurized battery are leaking due to defects based on the actual leakage profile includes: determining the actual leakage area based on the actual leakage profile; determining whether the actual leakage area exceeds a predetermined area threshold; and if the actual leakage area exceeds the predetermined area threshold, determining that the sealing pin and weld of the pressurized battery are leaking due to defects.

[0009] Optionally, in the battery sealing nail leakage detection method according to the present invention, the intelligent target detection model includes a backbone network, an anchor box generation network, a proposal generation network, and a fully connected network coupled sequentially, wherein the backbone network is also coupled to the proposal generation network; using the intelligent target detection model to perform target detection on the residual image to locate the leakage region in the residual image includes: extracting features from the residual image through the backbone network to obtain a feature map; generating anchor boxes based on the feature map through the anchor box generation network; generating candidate proposal regions based on the feature map and the anchor boxes through the proposal generation network; and classifying and adjusting the position of the candidate proposal regions as leakage regions and background noise respectively through the classifier and position regression module of the fully connected network to predict the location box of the leakage region, so as to locate the leakage region in the residual image according to the location box.

[0010] Optionally, in the battery sealing nail leakage detection method according to the present invention, the first image of the leakage detection area includes a first normal exposure image and a first high exposure image of the leakage detection area, and the second image of the leakage detection area includes a second normal exposure image and a second high exposure image of the leakage detection area; calculating the residual image between the first image and the second image includes: calculating the residual image between the first normal exposure image and the second normal exposure image, and the residual image between the first high exposure image and the second high exposure image, respectively.

[0011] Optionally, in the battery sealing pin leakage detection method according to the present invention, the battery is arranged in a leakage detection station, the leakage detection station is equipped with an imaging component and a pressure plate, the computing device is communicatively connected to the imaging component and the pressure plate, and the imaging component includes a camera and a lens; acquiring a first image of the leakage detection area of ​​the battery before pressurization, and acquiring a second image of the leakage detection area of ​​the battery after pressurization, includes: imaging the leakage detection area of ​​the battery through the imaging component to obtain a first image of the leakage detection area of ​​the battery; pressurizing the battery through the pressure plate, and imaging the leakage detection area of ​​the battery after pressurization through the imaging component to obtain a second image of the leakage detection area of ​​the battery.

[0012] Optionally, in the battery sealing nail leakage detection method according to the present invention, imaging the leakage detection area of ​​the battery by the imaging component to obtain a first image of the leakage detection area of ​​the battery includes: performing normal exposure imaging and high exposure imaging on the leakage detection area of ​​the battery by the imaging component to obtain a first normal exposure image and a first high exposure image of the leakage detection area; imaging the leakage detection area of ​​the pressurized battery by the imaging component to obtain a second image of the leakage detection area of ​​the battery includes: performing normal exposure imaging and high exposure imaging on the pressurized leakage detection area of ​​the battery by the imaging component to obtain a second normal exposure image and a second high exposure image of the leakage detection area.

[0013] Optionally, in the battery sealing nail leakage detection method according to the present invention, the intelligent target detection model adopts a localization tagging framework based on a region convolutional neural network.

[0014] According to one aspect of the present invention, a battery sealing pin leakage detection device is provided, deployed in a computing device, the device comprising: The acquisition unit is adapted to acquire a first image of the leakage detection area of ​​the battery before pressurization, and to acquire a second image of the leakage detection area of ​​the battery after pressurization, wherein the leakage detection area of ​​the battery includes a sealing nail, a weld and a cleaning area. The residual calculation unit is adapted to calculate the residual image between the first image and the second image; The target detection unit is adapted to use an intelligent target detection model to perform target detection on the residual image in order to locate the leakage area in the residual image, the leakage area including the bright area and the shadow area; The segmentation calculation unit is adapted to segment and calculate the leakage area to determine the actual leakage outline in the leakage area; The leakage detection unit is adapted to determine, based on the actual leakage profile, whether the sealing pins and welds of the pressurized battery are leaking due to defects.

[0015] According to one aspect of the present invention, a computing device is provided, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be executed by the at least one processor, the program instructions including instructions for performing the battery seal leak detection method as described above.

[0016] According to one aspect of the present invention, a computer program product is provided, comprising computer program instructions, wherein the computer program instructions, when executed by a processor, implement the method as described above.

[0017] According to one aspect of the present invention, a readable storage medium storing program instructions is provided, which, when read and executed by a computing device, causes the computing device to perform the battery seal leakage detection method as described above.

[0018] According to the technical solution of the present invention, a method for detecting battery sealing nail leakage is provided. Images of the battery leakage detection area are acquired before and after pressurization, and residual images are calculated between the images before and after pressurization. Then, an intelligent target detection model is used to perform target detection on the residual images to locate the leakage area. The leakage area is then segmented to determine the actual leakage contour. Finally, based on the actual leakage contour, it can be determined whether the battery sealing nail and weld are leaking due to defects after pressurization. Based on this, the present invention focuses the detection target on the battery leakage itself, detecting not only leakage caused by surface defects in the battery sealing nail and weld, but also leakage caused by internal defects in the battery sealing nail and weld, thereby achieving more accurate detection of battery sealing nail quality and improving detection efficiency. Secondly, by calculating the residual images before and after pressurization, background noise is greatly reduced, thereby reducing algorithm complexity and improving algorithm accuracy. Furthermore, the intelligent target detection model enables intelligent and accurate localization of the leakage area, eliminating residual noise interference, thereby further improving detection accuracy.

[0019] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0020] To achieve the foregoing and related objectives, certain illustrative aspects are described herein in conjunction with the following description and accompanying drawings. These aspects indicate various ways in which the principles disclosed herein may be practiced, and all aspects and their equivalents are intended to fall within the scope of the claimed subject matter. The foregoing and other objectives, features, and advantages of the invention will become more apparent from the following detailed description, taken in conjunction with the accompanying drawings. Throughout the invention, the same reference numerals generally refer to the same parts or elements.

[0021] Figure 1 A schematic diagram of a computing device 100 provided according to an embodiment of the present invention is shown; Figure 2 A schematic flowchart of a battery sealing nail leakage detection method 200 provided according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of the leakage detection area of ​​a battery according to an embodiment of the present invention is shown; Figure 4A schematic diagram is shown of imaging the leakage detection area of ​​the battery at the leakage detection station; Figure 5 This is a schematic diagram illustrating the effect of a second image of the leakage detection area of ​​a battery in the event of leakage after the battery is pressurized, according to some embodiments of the present invention. Figure 6 A schematic diagram of the algorithm flow for detecting leakage of battery sealing nails based on a first image and a second image according to an embodiment of the present invention is shown. Figure 7 A schematic diagram of the structure of an intelligent target detection model 700 according to some embodiments of the present invention is shown; Figure 8 A schematic diagram illustrating the effects of normal exposure imaging and high exposure imaging according to some embodiments of the present invention is shown; Figure 9 A schematic diagram of leakage detection results based on normal exposure imaging and high exposure imaging is shown in some embodiments of the present invention; Figure 10 A schematic diagram of a battery sealing nail leakage detection device 1000 provided according to an embodiment of the present invention is shown. Detailed Implementation

[0022] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0023] To address the problems of existing battery sealing nail quality inspection methods failing to detect internal defects in welds and having low inspection efficiency, this invention proposes a battery sealing nail leakage detection method. By combining machine vision and image processing technology, it can detect leakage caused by external surface defects and internal defects in the sealing nail and weld, and can improve detection accuracy and efficiency.

[0024] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0025] Figure 1 A schematic diagram of a computing device 100 according to an embodiment of the present invention is shown. Figure 1As shown, in a basic configuration, computing device 100 includes at least one processing unit 102 and system memory 104. According to one aspect, depending on the configuration and type of the computing device, the processing unit 102 may be implemented as a processor. System memory 104 includes, but is not limited to, volatile memory (e.g., random access memory), non-volatile memory (e.g., read-only memory), flash memory, or any combination of such memories. According to one aspect, system memory 104 includes an operating system 105.

[0026] According to one aspect, operating system 105 is, for example, suitable for controlling the operation of computing device 100. Furthermore, examples are practiced in conjunction with graphics libraries, other operating systems, or any other applications, and are not limited to any particular application or system. Figure 1 The basic configuration is illustrated by the components within the dashed lines. According to one aspect, the computing device 100 has additional features or functions. For example, according to one aspect, the computing device 100 includes additional data storage devices (removable and / or non-removable), such as disks, optical discs, or magnetic tapes. This additional storage... Figure 1 The middle part is shown by removable storage device 109 and non-removable storage device 110.

[0027] As stated above, according to one aspect, program module 103 is stored in system memory 104. According to one aspect, program module 103 may include one or more applications. The present invention does not limit the type of application; for example, applications may include: email and contact applications, word processing applications, spreadsheet applications, database applications, slideshow applications, drawing or computer-aided applications, web browser applications, etc.

[0028] According to one aspect, program module 103 may include a plurality of program instructions adapted to perform the battery seal leak detection method 200 of the present invention, such that computing device 100 is configured to perform the battery seal leak detection method 200 of the present invention.

[0029] According to one aspect, program module 103 may include a battery seal leak detection device 1000, which may be configured to perform the battery seal leak detection method 200 of the present invention.

[0030] According to one aspect, examples can be practiced on circuits including discrete electronic components, packaged or integrated electronic chips containing logic gates, circuits utilizing microprocessors, or on a single chip containing electronic components or a microprocessor. For example, it can be practiced via wherein... Figure 1Each or many of the components shown can be implemented as an example by integrating a System-on-a-Chip (SOC) on a single integrated circuit. According to one aspect, such an SOC device may include one or more processing units, graphics units, communication units, system virtualization units, and various application functions, all integrated (or “burned in”) as a single integrated circuit onto a chip substrate. When operating via the SOC, the functions described herein can be operated via dedicated logic integrated on a single integrated circuit (chip) with other components of the computing device 100. Embodiments of the invention can also be implemented using other techniques capable of performing logical operations (e.g., AND, OR, and NOT), including but not limited to mechanical, optical, fluid, and quantum technologies. Additionally, embodiments of the invention can be implemented within a general-purpose computer or in any other circuit or system.

[0031] According to one aspect, computing device 100 may also have one or more input devices 112, such as a keyboard, mouse, pen, voice input device, touch input device, etc. It may also include output devices 114, such as a display, speaker, printer, etc. The foregoing devices are examples and other devices may also be used. Computing device 100 may include one or more communication connections 116 that allow communication with other computing devices 118. Examples of suitable communication connections 116 include, but are not limited to: RF transmitter, receiver and / or transceiver circuitry; Universal Serial Bus (USB), parallel and / or serial ports.

[0032] As used herein, the term computer-readable medium includes computer storage medium. Computer storage medium can include volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information (e.g., computer-readable instructions, data structures, or program module 103). System memory 104, removable storage device 109, and non-removable storage device 110 are examples of computer storage media (i.e., memory storage). Computer storage media can include random access memory (RAM), read-only memory (ROM), electrically erasable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical storage, magnetic tape, magnetic tape, disk storage or other magnetic storage devices, or any other article of manufacture that can be used to store information and is accessible by computing device 100. According to one aspect, any such computer storage medium can be part of computing device 100. Computer storage media does not include carrier waves or other transmitted data signals.

[0033] According to one aspect, the communication medium is implemented by computer-readable instructions, data structures, program modules 103, or other data in a modulated data signal (e.g., a carrier wave or other transmission mechanism), and includes any information transmission medium. According to one aspect, the term "modulated data signal" describes a signal having one or more sets of characteristics or altered in a manner that encodes information in the signal. By way of example and not limitation, the communication medium includes wired media such as wired networks or direct wired connections, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0034] In an embodiment of the present invention, a computing device 100 is configured to perform the battery seal leak detection method 200 of the present invention. The computing device 100 includes one or more processors and one or more readable storage media storing program instructions that, when configured to be executed by the one or more processors, cause the computing device to perform the battery seal leak detection method 200 of the present invention.

[0035] Figure 2 A schematic flowchart of a battery sealing pin leakage detection method 200 according to an embodiment of the present invention is shown. The battery sealing pin leakage detection method 200 can be executed in a computing device (such as the aforementioned computing device 100).

[0036] In an embodiment of the present invention, the computing device 100 for performing the battery sealing nail leakage detection method 200 of the present invention may be a terminal or a server.

[0037] It should be noted that, in this embodiment of the invention, a leakage detection area is pre-defined for the battery sealing nail to be tested. Figure 3 A schematic diagram of the leakage detection area of ​​a battery according to an embodiment of the present invention is shown. Figure 3 As shown, the battery leakage detection area includes a sealing pin, a weld (i.e., the weld formed by welding the sealing pin), and a cleaning area. In some embodiments, the diameter of the battery leakage detection area is approximately 12 mm.

[0038] In this embodiment of the invention, the battery sealing pin can be leaked by acquiring images of the leakage detection area of ​​the battery before and after pressurization, so as to detect whether the battery sealing pin and weld are leaking due to defects.

[0039] like Figure 2 As shown, the battery sealing nail leakage detection method 200 includes the following steps 210-250.

[0040] Step 210: The computing device 100 can acquire a first image of the leakage detection area of ​​the battery before pressurization, and a second image of the leakage detection area of ​​the battery after pressurization. Figure 3 As shown, the battery leakage detection area includes the sealing nails, welds, and cleaning area.

[0041] In some embodiments, leakage detection can be performed on the battery sealing nails at the leakage detection station. Figure 4 This diagram illustrates the imaging of the battery's leakage detection area at the leakage detection station. Figure 4 As shown, the battery to be tested is arranged at the leakage detection station. An imaging component is arranged above the leakage detection station. The imaging component includes a camera and a lens (specifically a telecentric lens). The leakage detection station is also equipped with a pressure plate for pressurizing the battery.

[0042] The computing device 100 can communicate with the imaging assembly. Furthermore, the computing device 100 can communicate with the pressure plate, for example, through a control system.

[0043] Specifically, in step 210, the computing device 100 first images the battery leakage detection area using an imaging component (i.e., images the battery leakage detection area before pressurization) to obtain a first image of the battery leakage detection area. In other words, the computing device 100 can control the imaging component to image the battery leakage detection area to obtain a first image of the battery leakage detection area, and then acquire the first image of the battery leakage detection area from the imaging component. Subsequently, the computing device 100 pressurizes the battery using a pressure plate, and then images the pressurized battery leakage detection area using the imaging component to obtain a second image of the battery leakage detection area. In other words, the computing device 100 can control the pressure plate to pressurize the battery, and then control the imaging component to image the pressurized battery leakage detection area to obtain a second image of the battery leakage detection area, and then acquire the second image of the battery leakage detection area from the imaging component.

[0044] Figure 5 This diagram illustrates the effect of a second image of the leakage detection area of ​​a battery in the event of leakage after pressurization, according to some embodiments of the present invention. It should be noted that if leakage occurs after pressurizing the battery, such as... Figure 5 As shown, the refraction of the leaking liquid will cause obvious bright and shadow areas in the second image. These bright and shadow areas can be detected by comparing the first and second images before and after pressurization. If no leakage occurs after pressurizing the battery, there will be no obvious difference between the first and second images. Based on this, it can be determined whether leakage occurs after pressurizing the battery. See the following steps for details.

[0045] Step 220: The computing device 100 can calculate the residual image between the first image of the leakage detection area of ​​the battery before pressurization and the second image of the leakage detection area of ​​the battery after pressurization.

[0046] It should be understood that if leakage occurs after the battery is pressurized, the residual image between the first and second images will show bright and shadow areas due to the refraction of the leaked liquid (the bright and shadow areas in the residual image may be the leakage areas). It should be noted that bright areas are those with higher grayscale values ​​in the second image compared to the first image, and shadow areas are those with lower grayscale values ​​in the second image compared to the first image.

[0047] Step 230: The computing device 100 can use an intelligent target detection model to perform target detection on the residual image to locate the leakage area (suspected leakage area) in the residual image. Here, the leakage area includes both bright and shadow areas. That is, the intelligent target detection model is used to perform target detection on the residual image in order to locate the bright and shadow areas in the residual image caused by leakage.

[0048] It should be noted that there may be slight offsets and environmental changes in the two images of the battery leakage detection area before and after pressurization. The noise interference cannot be completely eliminated by the residual. Therefore, in this embodiment of the invention, an intelligent target detection model is used to locate the leakage area in the residual image, thereby achieving intelligent and accurate positioning of the leakage area and avoiding noise interference.

[0049] In this embodiment of the invention, the intelligent target detection model adopts a localization and labeling framework based on a region convolutional neural network.

[0050] Step 240: The computing device 100 can perform segmentation calculations on the leakage area to determine the actual outline of the leakage in the leakage area.

[0051] Here, the segmentation calculation for the leakage area includes the segmentation calculation for the highlighted area and the segmentation calculation for the shadow area. Specifically, when performing the segmentation calculation for the leakage area, the segmentation threshold can be calculated using the following formula: In the formula, This represents the average grayscale value of the leaking area.

[0052] Step 250: The calculation device 100 can determine whether the sealing nails and welds of the pressurized battery are leaking due to defects based on the actual leakage profile, thereby obtaining the leakage detection results.

[0053] here, Figure 6A schematic diagram of the algorithm flow for detecting leakage of battery sealing nails based on a first image and a second image according to an embodiment of the present invention is shown. The algorithm flow can correspond to the aforementioned steps 220-250.

[0054] Based on this, leaks caused by external and internal defects in sealing nails and welds can be detected.

[0055] In some embodiments, in step 250, when determining whether the sealing pins and welds of the pressurized battery are leaking due to defects based on the actual leakage profile, the computing device 100 first determines the actual leakage area based on the actual leakage profile, and then determines whether the actual leakage area exceeds a predetermined area threshold. If the actual leakage area exceeds the predetermined area threshold, it can be determined that the sealing pins and welds of the pressurized battery are leaking due to defects (including external surface defects and internal defects of the sealing pins and welds).

[0056] Figure 7 A schematic diagram of the structure of an intelligent target detection model 700 according to some embodiments of the present invention is shown.

[0057] like Figure 7 As shown, in some embodiments, the intelligent target detection model 700 includes a backbone network, an anchor box generation network, a proposal generation network, and a fully connected network that are coupled in sequence, wherein the backbone network is also coupled to the proposal generation network.

[0058] In step 230, the specific process of using an intelligent target detection model to perform target detection on the residual image to locate the leakage area in the residual image is as follows: First, the residual image is input into the backbone network of the intelligent target detection model. The feature map is obtained by extracting features from the residual image through the backbone network.

[0059] Subsequently, the feature map output by the backbone network is input into the anchor box generation network, which generates sparse anchor boxes based on the feature map.

[0060] Next, the feature map output by the backbone network and the anchor boxes output by the anchor box generation network are input into the proposal generation network. The proposal generation network can generate candidate proposal regions based on the feature map and the anchor boxes.

[0061] Next, the candidate proposal regions output by the suggestion generation network are input into a fully connected network, which includes a classifier and a location regression module. Finally, the classifier and location regression module of the fully connected network can be used to classify the leakage area and background noise, and adjust their positions (coordinate refinement) to predict the bounding box of the leakage area. Based on the bounding box, the leakage area in the residual image can be located.

[0062] It should be noted that the intelligent target detection model in this embodiment of the invention is obtained by training a pre-trained target detection model based on a training dataset.

[0063] In some embodiments, the training dataset includes multiple positive samples (i.e., residual images with leakage areas) and multiple negative samples (i.e., residual images without leakage areas and containing only background noise). The ratio of the number of positive samples to the number of negative samples can be 4:1. The positive samples contain ground truth labels for marking the leakage areas within them, while the negative samples have no labels.

[0064] It should be understood that the pre-trained object detection model has a similar structure to the aforementioned intelligent object detection model. When training the pre-trained object detection model based on the training dataset, each positive and negative sample can be input into the model. After processing, the model yields the probability that the anchor box is predicted as a leaking region (output by the classifier) ​​and the predicted offset of the anchor box relative to the true label of the localized box (output by the location regression module). Then, based on the probability of the anchor box being predicted as a leaking region and the true label of the localized box, the classification loss (the core of the classification loss is cross-entropy loss) is calculated. Finally, based on the predicted offset of the anchor box relative to the true label of the localized box and the actual offset of the anchor box relative to the true label of the localized box, the regression loss is calculated. Afterward, the pre-trained object detection model can be trained based on the classification loss and regression loss, adjusting its parameters to ultimately obtain the trained intelligent object detection model.

[0065] In some embodiments, the classification loss and regression loss can be weighted and summed to obtain the total loss, and then the pre-trained object detection model can be trained based on the total loss.

[0066] In some embodiments, the formula for calculating the classification loss is as follows: In the formula, It is the actual label of the positioning frame. It represents the probability that the anchor frame is predicted to be a leakage area. It is a normalized parameter.

[0067] In some embodiments, the regression loss uses the Smooth L1 Loss function, which is calculated as follows: In the formula, It is the predicted offset of the anchor frame relative to the actual label of the positioning frame. It is the actual offset of the anchor frame relative to the actual label of the positioning frame.

[0068] In some embodiments, the formula for obtaining the total loss by weighted summation of classification loss and regression loss is as follows: In some embodiments, given the low reflectivity of the cleaning area (which appears black under normal exposure), high-exposure imaging can be performed simultaneously with normal exposure imaging to detect leakage in the cleaning area. Figure 8 A schematic diagram illustrating the effects of normal exposure imaging and high exposure imaging according to some embodiments of the present invention is shown.

[0069] Specifically, in step 210, the first image of the leakage detection area of ​​the battery before pressurization includes a first normal exposure image and a first high exposure image of the leakage detection area, and the second image of the leakage detection area of ​​the battery after pressurization includes a second normal exposure image and a second high exposure image of the leakage detection area.

[0070] Specifically, before pressurization, when imaging the leakage detection area of ​​the battery using the imaging component, the imaging component can perform normal exposure imaging and high exposure imaging on the leakage detection area of ​​the battery to obtain a first normal exposure image and a first high exposure image of the leakage detection area. Correspondingly, after pressurizing the battery using the pressurizing plate, when imaging the leakage detection area of ​​the pressurized battery using the imaging component, the imaging component can perform normal exposure imaging and high exposure imaging on the leakage detection area of ​​the pressurized battery to obtain a second normal exposure image and a second high exposure image of the leakage detection area. That is to say, normal exposure imaging and high exposure imaging are performed before and after pressurizing the battery, respectively. Therefore, the first image obtained in step 210 may include the first normal exposure image and the first high exposure image, and the second image obtained may include the second normal exposure image and the second high exposure image.

[0071] When calculating the residual image between the first image and the second image in step 220, the residual image between the first normal exposure image and the second normal exposure image, as well as the residual image between the first high exposure image and the second high exposure image, can be calculated respectively.

[0072] Subsequently, for the residual images between the first and second normal exposure images (residual images based on normal exposure imaging) and the residual images between the first and second high-exposure images (residual images based on high-exposure imaging), subsequent steps 230-250 can be performed respectively: Target detection is performed on the residual images using an intelligent target detection model to locate the leakage area in the residual images; the leakage area is segmented and calculated to determine the actual leakage contour within the leakage area; based on the actual leakage contour, it is determined whether the sealing nails and welds of the pressurized battery are leaking due to defects, thus obtaining the corresponding leakage detection results. Here, Figure 9 The diagram illustrates leakage detection results based on normal exposure imaging and high exposure imaging according to some embodiments of the present invention. It should be noted that the specific implementation methods of steps 230-250 can be found in the descriptions of the preceding embodiments, and will not be repeated here.

[0073] In step 250, when determining whether the sealing pins and welds of the pressurized battery are leaking due to defects based on the actual leakage profile, the actual leakage profile of the sealing pins and welds of the pressurized battery can be used for both normal exposure imaging and high exposure imaging to determine whether the actual leakage area is leaking due to defects. That is, for both normal exposure imaging and high exposure imaging, the actual leakage area is determined based on the actual leakage profile of the corresponding leakage area, and it is determined whether the actual leakage area exceeds a predetermined area threshold. If the actual leakage area exceeds the predetermined area threshold, it is determined that the sealing pins and welds of the pressurized battery are leaking due to defects (including external surface defects and internal defects of the sealing pins and welds).

[0074] In addition, if the actual leakage area corresponding to normal exposure imaging does not exceed the predetermined area threshold, and the actual leakage area corresponding to high exposure imaging does not exceed the predetermined area threshold, then it can be further determined whether the total area of ​​the actual leakage area corresponding to normal exposure imaging and the actual leakage area corresponding to high exposure imaging exceeds the predetermined area threshold. If the total area exceeds the predetermined area threshold, it is determined that the sealing nails and welds of the pressurized battery are leaking due to defects; if the total area does not exceed the predetermined area threshold, it is determined that the sealing nails and welds of the pressurized battery are not leaking due to defects.

[0075] Figure 10 A schematic diagram of a battery seal leak detection device 1000 according to an embodiment of the present invention is shown. The battery seal leak detection device 1000 can be deployed in a computing device 100, and the battery seal leak detection device 1000 is configured to perform the battery seal leak detection method 200 of the present invention.

[0076] like Figure 10 As shown, in an embodiment of the present invention, the battery sealing nail leakage detection device 1000 includes an acquisition unit 1010, a residual calculation unit 1020, a target detection unit 1030, a segmentation calculation unit 1040, and a leakage detection unit 1050 that are sequentially connected in communication.

[0077] The acquisition unit 1010 can acquire a first image of the leakage detection area of ​​the battery before pressurization, and acquire a second image of the leakage detection area of ​​the battery after pressurization. The leakage detection area of ​​the battery includes a sealing nail, a weld, and a cleaning area.

[0078] The residual calculation unit 1020 can calculate the residual image between the first image and the second image.

[0079] The target detection unit 1030 can use an intelligent target detection model to perform target detection on the residual image in order to locate the leakage area in the residual image, which includes a bright area and a shadow area.

[0080] The segmentation calculation unit 1040 can perform segmentation calculations on the leakage area to determine the actual outline of the leakage in the leakage area.

[0081] The leakage detection unit 1050 can determine whether the sealing nails and welds of the pressurized battery are leaking due to defects based on the actual leakage profile.

[0082] It should be noted that the acquisition unit 1010, residual calculation unit 1020, target detection unit 1030, segmentation calculation unit 1040, and leakage detection unit 1050 are respectively used to execute the aforementioned steps 210-250. Here, the specific execution logic of each unit can be found in the description of steps 210-250 in method 200 above, and will not be repeated here.

[0083] According to the battery sealing nail leakage detection method 200 of this embodiment, images of the battery leakage detection area are acquired before and after pressurization of the battery, and the residual images before and after pressurization are calculated. Then, an intelligent target detection model is used to perform target detection on the residual images to locate the leakage area in the residual images. The leakage area is then segmented to determine the actual leakage contour. Finally, based on the actual leakage contour, it can be determined whether the battery sealing nail and weld are leaking due to defects after pressurization. Based on this, this invention focuses the detection target on the battery leakage itself, detecting not only leakage caused by surface defects in the battery sealing nail and weld, but also leakage caused by internal defects in the battery sealing nail and weld, thereby achieving more accurate detection of battery sealing nail quality and improving detection efficiency. Secondly, by calculating the residual images of the two images before and after pressurization, background noise is greatly reduced, thereby reducing algorithm complexity and improving algorithm accuracy. Furthermore, the intelligent target detection model enables intelligent and accurate positioning of the leakage area, eliminating residual noise interference, thereby further improving detection accuracy.

[0084] The various techniques described herein can be implemented in combination with hardware or software, or a combination thereof. Thus, the methods and apparatus of the present invention, or certain aspects or portions thereof, can take the form of program code (i.e., instructions) embedded in a tangible medium, such as a removable hard disk, USB flash drive, floppy disk, CD-ROM, or any other machine-readable storage medium, wherein when the program is loaded into and executed by a machine such as a computer, the machine becomes an apparatus for practicing the present invention.

[0085] When the program code is executed on a programmable computer, the mobile terminal generally includes a processor, a processor-readable storage medium (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. The memory is configured to store program code; the processor is configured to execute the battery sealing pin leakage detection method of the present invention according to instructions in the program code stored in the memory.

[0086] By way of example, and not limitation, readable media include readable storage media and communication media. Readable storage media stores information such as computer-readable instructions, data structures, program modules, or other data. Communication media generally embodies computer-readable instructions, data structures, program modules, or other data in the form of modulated data signals such as carrier waves or other transmission mechanisms, and includes any information delivery medium. Any combination of the above is also included within the scope of readable media.

[0087] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with the examples of this invention. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0088] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0089] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various aspects of the invention, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0090] Those skilled in the art will understand that modules, units, or components of the devices disclosed in the examples herein can be arranged in the devices described in this embodiment, or alternatively, can be located in one or more devices different from the devices in this example. The modules in the foregoing examples can be combined into a single module or, in addition, can be divided into multiple sub-modules.

[0091] Unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

Claims

1. A method for detecting leakage of battery sealing pins, executed in a computing device, comprising: A first image of the leakage detection area of ​​the battery before pressurization is obtained, and a second image of the leakage detection area of ​​the battery after pressurization is obtained, wherein the leakage detection area of ​​the battery includes a sealing nail, a weld and a cleaning area. Calculate the residual image between the first image and the second image; Using an intelligent target detection model, target detection is performed on the residual image to locate the leakage area in the residual image, the leakage area including bright areas and shadow areas; The leakage area is segmented and calculated to determine the actual leakage outline within the leakage area. Based on the actual outline of the leak, determine whether the sealing pins and welds of the pressurized battery are leaking due to defects.

2. The method as described in claim 1, wherein, Based on the actual leakage profile, determine whether the sealing pins and welds of the pressurized battery are leaking due to defects, including: Determine the actual area of ​​the leak based on the actual outline of the leak. Determine whether the actual area of ​​the leakage exceeds a predetermined area threshold; If the actual leakage area exceeds a predetermined area threshold, it is determined that the leakage is caused by defects in the sealing nails and welds of the pressurized battery.

3. The method as described in claim 1 or 2, wherein, The intelligent target detection model includes a backbone network, an anchor box generation network, a proposal generation network, and a fully connected network that are coupled in sequence, wherein the backbone network is also coupled to the proposal generation network. Using an intelligent target detection model, target detection is performed on the residual image to locate the leakage area in the residual image, including: The feature map is obtained by extracting features from the residual image through the backbone network. Anchor boxes are generated based on the feature map using the anchor box generation network. The proposal generation network generates candidate proposal regions based on the feature map and the anchor box. The classifier and location regression module of the fully connected network are used to classify the leakage area and background noise and adjust their positions respectively to predict the location box of the leakage area, so as to locate the leakage area in the residual image according to the location box.

4. The method according to any one of claims 1-3, wherein, The first image of the leakage detection area includes a first normal exposure image and a first high exposure image of the leakage detection area, and the second image of the leakage detection area includes a second normal exposure image and a second high exposure image of the leakage detection area; Calculating the residual image between the first image and the second image includes: Calculate the residual images between the first normal exposure image and the second normal exposure image, and the residual images between the first high exposure image and the second high exposure image, respectively.

5. The method according to any one of claims 1-4, wherein, The battery is arranged at the leakage detection station, which is equipped with an imaging component and a pressure plate. The computing device is communicatively connected to the imaging component and the pressure plate. The imaging component includes a camera and a lens. Acquiring a first image of the leakage detection area of ​​the battery before pressurization, and acquiring a second image of the leakage detection area of ​​the battery after pressurization, including: The imaging component is used to image the leakage detection area of ​​the battery to obtain a first image of the leakage detection area of ​​the battery. The battery is pressurized by a pressure plate, and the leakage detection area of ​​the pressurized battery is imaged by the imaging component to obtain a second image of the leakage detection area of ​​the battery.

6. The method of claim 5, wherein, Imaging the leakage detection area of ​​the battery using the imaging component to obtain a first image of the leakage detection area of ​​the battery includes: The imaging component performs normal exposure imaging and high exposure imaging on the leakage detection area of ​​the battery to obtain a first normal exposure image and a first high exposure image of the leakage detection area. The imaging component images the leakage detection area of ​​the pressurized battery to obtain a second image of the leakage detection area of ​​the battery, including: The imaging component performs normal exposure imaging and high exposure imaging on the leakage detection area of ​​the pressurized battery to obtain a second normal exposure image and a second high exposure image of the leakage detection area.

7. The method according to any one of claims 1-6, wherein, The intelligent target detection model adopts a localization and labeling framework based on a region convolutional neural network.

8. A battery sealing pin leakage detection device, deployed in a computing device, the device comprising: The acquisition unit is adapted to acquire a first image of the leakage detection area of ​​the battery before pressurization, and to acquire a second image of the leakage detection area of ​​the battery after pressurization, wherein the leakage detection area of ​​the battery includes a sealing nail, a weld and a cleaning area. The residual calculation unit is adapted to calculate the residual image between the first image and the second image; The target detection unit is adapted to use an intelligent target detection model to perform target detection on the residual image in order to locate the leakage area in the residual image, the leakage area including the bright area and the shadow area; The segmentation calculation unit is adapted to segment and calculate the leakage area to determine the actual leakage outline in the leakage area; The leakage detection unit is adapted to determine, based on the actual leakage profile, whether the sealing pins and welds of the pressurized battery are leaking due to defects.

9. A computing device, comprising: At least one processor; and A memory storing program instructions, wherein the program instructions are configured to be processed by the at least one processor, the program instructions including instructions for processing the method as described in any one of claims 1-7.

10. A computer program product comprising computer program instructions, wherein, When the computer program instructions are executed by the processor, they implement the method as described in any one of claims 1-7.

Images