CN117781914A

CN117781914A - Tab dislocation detection system and method

Info

Publication number: CN117781914A
Application number: CN202410218403.7A
Authority: CN
Inventors: 李建宁
Original assignee: Contemporary Amperex Technology Co Ltd
Current assignee: Contemporary Amperex Technology Co Ltd
Priority date: 2024-02-28
Filing date: 2024-02-28
Publication date: 2024-03-29

Abstract

The application relates to a tab dislocation detection system and a tab dislocation detection method. The method comprises the following steps: according to the conversion relation among the first exposure type image, the real space and the image space corresponding to the first exposure type image of the tab to be detected, determining the initial tab edge distance of the tab to be detected, and according to the detection model and the second exposure type image of the tab to be detected, determining the target tab layer interval corresponding to the second exposure type image, so as to correct the initial tab edge distance according to the target tab layer interval, and obtaining the target tab edge distance of the tab to be detected. The electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer; the first exposure type image and the second exposure type image are shot by the vision sensor along the thickness direction of the electrode lug to be detected, the field of view range of the vision sensor covers the electrode lug to be detected, and the target electrode lug layer interval is the interval with the maximum offset of the edge in the thickness direction of the electrode lug to be detected. By adopting the method, the accuracy of the dislocation detection of the tab can be improved.

Description

Tab dislocation detection system and method

Technical Field

The application relates to the technical field of batteries, in particular to a tab dislocation detection system and a tab dislocation detection method.

Background

In the battery production process, in order to prevent the bad battery core of tab dislocation from flowing out, the tab is detected.

The manual work detects the tab and needs to consume more manpower and time, and detection efficiency is lower. Therefore, at present, an image of the tab to be measured is generally obtained through the vision sensor, so as to calculate the tab margin of the tab to be measured according to the image, and thus whether the tab to be measured has tab dislocation or not is judged according to the tab margin.

However, the accuracy of the tab misalignment detection result determined in the above method is not high.

Disclosure of Invention

Accordingly, it is desirable to provide a tab misalignment detection system and method that can improve accuracy in response to the above-described problems.

In a first aspect, the present application provides a tab misalignment detection method, including:

determining an initial tab edge distance of the tab to be measured according to a conversion relation among the first exposure type image, the real space and the image space corresponding to the first exposure type image of the tab to be measured; the electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer;

determining a target tab layer interval corresponding to the second exposure type image according to the detection model and the second exposure type image of the tab to be detected; the first exposure type image and the second exposure type image are shot by a visual sensor along the thickness direction of the tab to be detected, the field of view range of the visual sensor covers the tab to be detected, and the target tab layer interval is the interval with the maximum offset of the edge in the thickness direction of the tab to be detected;

Correcting the initial tab edge distance according to the target tab layer interval to obtain a target tab edge distance of the tab to be measured; the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

In the above embodiment, according to the conversion relationship between the first exposure type image, the real space and the image space corresponding to the first exposure type image of the tab to be detected, the initial tab edge distance of the tab to be detected is determined, and according to the detection model and the second exposure type image of the tab to be detected, the target tab layer interval corresponding to the second exposure type image is determined. Because the first exposure type image and the second exposure type image are shot by the vision sensor along the thickness direction of the tab to be detected, the field of view range of the vision sensor covers the tab to be detected, the tab to be detected comprises a plurality of tab layer intervals, the tab layer interval comprises at least one tab layer, the target tab layer interval is the interval with the largest offset of the edge in the thickness direction of the tab to be detected, therefore, the determined target tab layer interval can reflect the tab layer interval in which the most convex edge position of the tab is actually located, the initial tab edge distance is corrected according to the target tab layer interval, and after the target tab edge distance of the tab to be detected is obtained, the influence caused by the inconsistency between the tab layer interval in which the most convex edge position is located and the tab layer interval in which the calibration piece is located can be reduced, so that the error of the initial tab edge distance is reduced, the accuracy of the target tab edge distance is improved, and the accuracy of tab dislocation detection is improved.

In one embodiment, correcting the initial tab edge distance according to the target tab layer interval to obtain the target tab edge distance of the tab to be measured includes:

determining a height difference between the target tab layer interval and the calibration tab layer interval; calibrating the tab layer interval, namely calibrating the real space and the image space to obtain the tab layer interval where the calibration sheet is positioned in the process of converting the relationship;

determining a correction value according to the field angle and the height difference value of the vision sensor;

and correcting the initial tab edge distance according to the corrected value to obtain the target tab edge distance.

In the above embodiment, the calibration tab layer interval is the tab layer interval where the calibration tab is located in the process of calibrating the real space and the image space to obtain the conversion relationship, so after determining the height difference between the target tab layer interval and the calibration tab layer interval and determining the correction value according to the field angle and the height difference of the vision sensor, the initial tab margin can be corrected according to the correction value to obtain the target tab margin, so as to reduce the error caused by inconsistent between the target tab layer interval and the calibration tab layer interval.

In one embodiment, determining the height difference between the target tab layer interval and the nominal tab layer interval includes:

Determining a first height of the target tab layer interval according to the heights of the tab layers included in the target tab layer interval;

and taking the difference value between the first height and the second height of the calibrated tab layer interval as a height difference value.

In the above embodiment, since the first height of the target tab layer section is determined according to the height of each tab layer included in the target tab layer section, and the difference between the first height and the second height of the calibration tab layer section is used as the height difference, the accuracy of the height difference is improved.

In one embodiment, correcting the initial tab edge distance according to the correction value to obtain the target tab edge distance includes:

determining the position relationship between the target tab layer interval and the calibrated tab layer interval and the visual sensor;

and correcting the initial tab edge distance according to the corrected value and the position relation to obtain the target tab edge distance.

In the above embodiment, since the position relationship between the target tab layer interval and the calibration tab layer interval and the vision sensor is determined, and the initial tab edge distance is corrected according to the correction value and the position relationship, the target tab edge distance is obtained. Therefore, after the initial tab edge distance is corrected, the target tab edge distance with good accuracy can be obtained.

In one embodiment, correcting the initial tab edge distance according to the correction value and the position relation to obtain the target tab edge distance includes:

if the position relation is that the target tab layer interval is close to the visual sensor, obtaining a target tab edge distance according to the difference between the initial tab edge distance and the correction value;

and if the position relation is that the calibrated tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the sum of the initial tab edge distance and the correction value.

In the above embodiment, the target tab edge distance is obtained according to the difference between the initial tab edge distance and the correction value when the position relationship is that the target tab layer interval is close to the vision sensor, and the target tab edge distance is obtained according to the sum of the initial tab edge distance and the correction value when the position relationship is that the calibration tab layer interval is close to the vision sensor. Therefore, after the initial tab edge distance is corrected by the correction value, the target tab edge distance with good accuracy can be obtained.

In one embodiment, the method further comprises:

acquiring a first tab image sample and a tab layer interval label corresponding to the first tab image sample;

processing the first lug image sample to obtain a second lug image sample;

And training the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain a detection model.

In the above embodiment, since the first tab image sample and the tab layer interval label corresponding to the first tab image sample are obtained, and the first tab image sample is processed to obtain the second tab image sample, the initial detection model can be trained to obtain the detection model according to the second tab image sample and the tab layer interval label corresponding to the second tab image sample. Therefore, the detection model has the capability of determining the target tab layer interval corresponding to the second exposure type image.

In one embodiment, training the initial detection model according to the second tab image sample and the tab layer interval label corresponding to the second tab image sample to obtain the detection model includes:

training the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain a plurality of candidate detection models;

determining quantized values of each candidate detection model;

and determining a detection model from the candidate detection models according to the quantized values of the candidate detection models.

In the above embodiment, since the initial detection model is trained according to the second tab image sample and the tab layer interval label corresponding to the second tab image sample, a plurality of candidate detection models are obtained, and the quantization value of each candidate detection model is determined, so that according to the quantization value of each candidate detection model, the detection model with better certainty can be determined from each candidate detection model, so as to improve the accuracy of the target tab layer interval.

In one embodiment, the first exposure type image and the second exposure type image are images photographed by a vision sensor under the condition that the position of the tab to be measured is unchanged.

The embodiment is beneficial to reducing the error between the first exposure type image and the second exposure type image, thereby improving the accuracy of the tab edge distance.

In a second aspect, the application further provides a tab dislocation detection system, which comprises a light source, a visual sensor and an upper computer, wherein the light source comprises a first light source and a second light source, and the light source is arranged in a first direction of the thickness of a battery cell to which a tab to be detected belongs;

each tab to be measured corresponds to one visual sensor, and the visual sensor is used for acquiring a first exposure type image and a second exposure type image of the tab to be measured;

The light of the first light source irradiates the battery cell to which the electrode lug to be detected belongs and is used for supplementing light to the battery cell, and the light of the second light source irradiates the electrode lug to be detected and is used for supplementing light to the electrode lug to be detected;

the upper computer is used for determining an initial tab edge distance of the tab to be detected according to a conversion relation among the first exposure type image, the real space and the image space corresponding to the first exposure type image of the tab to be detected, determining a target tab layer interval corresponding to the second exposure type image from a plurality of tab layer intervals according to the detection model and the second exposure type image of the tab to be detected, and correcting the initial tab edge distance according to the target tab layer interval to obtain a target tab edge distance of the tab to be detected; the first exposure type image and the second exposure type image are shot by a visual sensor along the thickness direction of the tab to be detected, the field of view range of the visual sensor covers the tab to be detected, and the target tab layer interval is the interval with the maximum offset of the edge in the thickness direction of the tab to be detected; the electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer; the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

In one embodiment, the first exposure type image is captured by the vision sensor with a first brightness of the light source, and the second exposure type image is captured by the vision sensor with a second brightness of the light source, wherein the first brightness is greater than the second brightness; or,

the method comprises the steps of shooting when a first exposure type image is the third brightness of a light source, shooting when the shutter speed of a visual sensor is the first speed, shooting when a second exposure type image is the third brightness of the light source, and shooting when the shutter speed of the visual sensor is the second speed, wherein the first speed is smaller than the second speed; or,

the light source further comprises a third light source, the third light source is arranged in a second direction of the thickness of the battery cell to which the tab to be measured belongs, and the second direction is opposite to the first direction; the first exposure type image is photographed by a visual sensor with a third light source at a fourth brightness, the second exposure type image is photographed by a visual sensor with a second brightness of the light source, or the second exposure type image is photographed with a shutter speed of the visual sensor at a second speed of the light source at the third brightness.

In one embodiment, two tabs to be tested correspond to the same visual sensor, or two tabs to be tested correspond to different visual sensors respectively.

In one embodiment, the system further comprises a conveyor assembly, a carrier assembly, a bar code detection device, and a bracket for supporting the vision sensor and the light source;

the object carrying assembly is used for carrying the battery cell to which the tab to be tested belongs;

the transmission assembly is used for transmitting the object carrying assembly carrying the battery cell to a preset image acquisition position corresponding to the vision sensor;

the bar code detection device is used for acquiring the identification of the battery cell and sending the identification to the upper computer under the condition of transmitting the battery cell;

and the upper computer is used for controlling the light source to supplement light through the light source controller under the condition that the cell corresponding to the mark is detected to reach the preset image acquisition position, and controlling the vision sensor to acquire the first exposure type image and the second exposure type image.

In a third aspect, the present application further provides a tab misalignment detection apparatus, including:

the first determining module is used for determining the initial tab edge distance of the tab to be detected according to the conversion relation among the first exposure type image, the real space and the image space corresponding to the first exposure type image of the tab to be detected; the electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer;

The second determining module is used for determining a target tab layer interval corresponding to the second exposure type image from the tab layer intervals according to the detection model and the second exposure type image of the tab to be detected; the first exposure type image and the second exposure type image are shot by a visual sensor along the thickness direction of the tab to be detected, the field of view range of the visual sensor covers the tab to be detected, and the target tab layer interval is the interval with the maximum offset of the edge in the thickness direction of the tab to be detected;

the correction module is used for correcting the initial tab edge distance according to the target tab layer interval to obtain the target tab edge distance of the tab to be detected; the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

In a fourth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of any of the methods described above.

In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of any of the methods described above.

The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort for a person having ordinary skill in the art.

FIG. 1 is a schematic illustration of a detection process;

FIG. 2 is a schematic view of tab margins;

FIG. 3 is a schematic diagram of an error;

FIG. 4 is a schematic diagram of yet another error;

fig. 5 is an application environment diagram of a tab misalignment detection method in an embodiment of the present application;

fig. 6 is a schematic flow chart of a tab misalignment detection method in an embodiment of the present application;

fig. 7 is a schematic diagram of a tab layer interval tag in an embodiment of the present application;

FIG. 8 is a schematic flow chart of obtaining a target tab margin according to an embodiment of the present application;

FIG. 9 is a second flow chart for obtaining a target tab margin according to an embodiment of the present disclosure;

FIG. 10 is a third flow chart for obtaining a target tab margin according to an embodiment of the present disclosure;

FIG. 11 is a schematic flow chart of obtaining a detection model according to an embodiment of the present application;

FIG. 12 is a second flow chart of a method for obtaining a test model according to the embodiment of the present application;

FIG. 13 is a schematic diagram of a process for obtaining a detection model according to an embodiment of the present application;

fig. 14 is a schematic process diagram of a tab misalignment detection method in an embodiment of the present application;

fig. 15 is a schematic structural diagram of a tab misalignment detection system according to an embodiment of the present application;

FIG. 16 is a schematic diagram illustrating a tab misalignment detection system in accordance with an embodiment of the present disclosure;

FIG. 17 is a schematic diagram of a visual sensor according to an embodiment of the present application;

FIG. 18 is a second schematic diagram of a vision sensor according to the embodiment of the present disclosure;

FIG. 19 is a second schematic diagram of a tab misalignment detection system according to an embodiment of the present disclosure;

FIG. 20 is a block diagram of a tab misalignment detection apparatus according to an embodiment of the present disclosure;

FIG. 21 is a block diagram illustrating a configuration of a correction module according to an embodiment of the present application;

fig. 22 is a block diagram of the first determination unit in the embodiment of the present application;

FIG. 23 is a block diagram of a correction unit in an embodiment of the present application;

FIG. 24 is a second block diagram of a tab misalignment detection apparatus according to an embodiment of the present disclosure;

fig. 25 is a block diagram of a training module in an embodiment of the present application.

Detailed Description

Embodiments of the technical solutions of the present application will be described in detail below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical solutions of the present application, and thus are only examples, and are not intended to limit the scope of protection of the present application.

Unless defined otherwise, 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 application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description and claims of the present application and in the description of the figures above are intended to cover non-exclusive inclusions.

In the description of the embodiments of the present application, the technical terms "first," "second," etc. are used merely to distinguish between different objects and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated, a particular order or a primary or secondary relationship. In the description of the embodiments of the present application, the meaning of "plurality" is two or more unless explicitly defined otherwise.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

In the description of the embodiments of the present application, the term "plurality" refers to two or more (including two), and similarly, "plural sets" refers to two or more (including two), and "plural sheets" refers to two or more (including two).

The tab refers to a metal conductor which leads out the positive electrode and the negative electrode from the battery core, and is an important component of the battery. Since the misalignment of the tab affects the reliability and safety of the battery, it is a necessary technical means to detect the tab in order to prevent the outflow of the defective battery cell due to the misalignment of the tab.

The manual work detects the tab and needs to consume more manpower and time, and detection efficiency is lower. Thus, detection is currently performed by a machine vision based charge coupled device (Charge Coupled Device, CCD) appearance detection device.

Fig. 1 is a schematic diagram of a detection process, as shown in fig. 1, currently, a vision sensor 102 is disposed at an upper position between two tabs of a battery cell. In this way, after the battery cell 101 is transferred from the station 1 to the station 2 through the carrier component 103 such as a tray, the station 2 is a detection station, so that the upper computer triggers the light source and the vision sensor 102 to work, and the vision sensor 102 acquires the image of the tab to be detected. And then, the image of the tab to be measured is obtained through the visual sensor, so that the tab margin of the tab to be measured can be calculated. Afterwards, the battery cell 101 continues to be moved to the next station 3 through the carrier assembly 103, so that the tab dislocation detection of the next battery cell of the battery cell 101 is realized.

Fig. 2 is a schematic view of a tab edge distance, as shown in fig. 2, where the tab edge distance is used to indicate a distance between an edge position of a tab and an edge of a battery cell, and taking an example that a tab to be measured includes a left tab and a right tab, the tab edge distance includes a distance 1, a distance 2, a distance 3, and a distance 4 in fig. 1. The edge position is used for indicating the most protruding position of the side face of the tab, namely the position of the edge with the largest offset of the edge of the tab in the thickness direction of the tab. The edge positions of one tab may include a left edge position and a right edge position. Fig. 3 is an error diagram, as shown in fig. 3, the black dot in fig. 3 is the edge position of the left side of the tab 301.

Further, it can be determined that the tab to be measured has no tab dislocation under the condition that the distance 1, the distance 2, the distance 3 and the distance 4 are all located in the corresponding preset ranges, whereas it is determined that the tab to be measured has tab dislocation under the condition that the distance 1, the distance 2, the distance 3 and the distance 4 are not located in the corresponding preset ranges.

Therefore, in the prior art, the tab misalignment detection depends on the tab edge distance, and an image of the tab to be detected is needed to be used in the tab edge distance determination process. Because there is a coordinate difference between the real space where the tab to be measured is located and the image space corresponding to the vision sensor, before calculating the tab edge distance of the tab to be measured by using the image, coordinate system transformation is needed to determine the conversion relationship between the real space and the image space, and this step may also be referred to as calibration.

The tab may generally include a plurality of tab layers. With continued reference to fig. 3, fig. 3 illustrates a tab including a plurality of tab layers, which may be shown by the dashed lines in fig. 3, the outermost tab layer, i.e., the black region in fig. 3. In the calibration process, a calibration sheet is generally placed on the tab layer of the outermost layer of the tab, and an image of the calibration sheet is acquired by using a visual sensor, so that the conversion relationship between the real space and the image space is determined according to the actual size information of the calibration sheet and the image of the calibration sheet. And obtaining the tab edge distance of the tab to be measured according to the image of the tab to be measured and the conversion relation.

However, since the conversion relationship between the real space and the image space is calculated by using the calibration sheet in the calibration process, and the calibration sheet is placed at a fixed height, the conversion relationship obtained by the final calculation is only applicable to the plane in which the calibration sheet is located, that is, the obtained conversion relationship is the relationship of each coordinate point on the tab layer of the outermost layer of the tab. And, the differences in different planes cannot be measured. That is, the above-mentioned conversion relationship is more accurate on the tab layer of the outermost layer of the tab, but errors occur on other tab layers of the tab.

As shown in fig. 3, since the edge of the tab layer of the tab to be measured has poor uniformity, the edge position determined by the image of the tab to be measured may be located in any one tab layer of the tab, for example, in the middle tab layer of the tab 301 at the left edge position in fig. 3.

If the lug layer where the edge position of the lug is located is inconsistent with the lug layer where the calibration sheet is located, the information obtained by the imaging plane is distorted according to the optical principle of far, near and large, and the image of the lug to be measured has errors. Furthermore, the tab edge distance obtained based on the image and the conversion relation of the tab to be detected is inaccurate.

Fig. 4 is a schematic diagram of yet another error. Referring to fig. 3 and 4, assuming that the field angle of the visual sensor 102 at the edge of the tab is α, and the difference in height between the tab layer where the edge position is actually located and the tab layer where the calibration surface is located is l, there is an error of at least l×tan α in the imaging surface. By way of example, on the basis of fig. 1, assuming a tab edge field angle of 10 °, a tab height difference of 7.5 mm, the measurement error will reach 1.3 mm.

Therefore, in the actual use process, the method has the problem that the accuracy of the determined tab dislocation detection result is not high. In view of the above, it is necessary to provide a tab misalignment detection method that can improve accuracy. The tab misalignment detection method is described below.

The battery disclosed by the embodiment of the application can be used in electric devices such as vehicles, ships or aircrafts, but is not limited to the batteries. The power supply system with the battery and the like forming the power utilization device can be used, so that the situation of dislocation of the lugs in the battery is reduced, and the reliability and the safety of the battery are improved.

Fig. 5 is an application environment diagram of a tab misalignment detection method in an embodiment of the present application. In one exemplary embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 1. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing relevant data. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by a processor is used for realizing a tab misalignment detection method.

It will be appreciated by those skilled in the art that the structure shown in fig. 5 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

The embodiment is illustrated by applying the method to a server, and it is understood that the method may also be applied to a terminal, and may also be applied to a system including the terminal and the server, and implemented through interaction between the terminal and the server. The terminal may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers. The server may be implemented as a stand-alone server or as a server cluster composed of a plurality of servers.

Fig. 6 is a flowchart of a tab misalignment detection method in an embodiment of the present application, and in an exemplary embodiment, as shown in fig. 6, a tab misalignment detection method is provided, and the method is applied to the computer device in fig. 1 for illustration, and includes the following S601 to S603.

S601, determining an initial tab edge distance of a tab to be detected according to a conversion relation among a first exposure type image, a real space and an image space corresponding to the first exposure type image of the tab to be detected; the tab to be measured comprises a plurality of tab layer intervals, and each tab layer interval comprises at least one tab layer.

In this embodiment, the tab to be measured includes at least one tab. The tab to be measured may be a left tab or a right tab of the cell, and in some embodiments, the tab to be measured may also include a left tab and a right tab. Continuing with the example above, taking the example where the tab to be measured includes the left tab in fig. 2, the initial tab margin may include a distance 1 and a distance 2. Taking the example that the tab to be measured includes the right tab in fig. 2, the initial tab margin may include a distance 3 and a distance 4. In the case where the tab to be measured includes the left tab and the right tab in fig. 2, the initial tab margin may include a distance 1, a distance 2, a distance 3, and a distance 4.

The electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer. For example, if a certain tab to be measured includes tab layers 1 to 10 sequentially from top to bottom in the thickness direction, each tab layer may correspond to one tab layer section, for example, tab layer 1 corresponds to tab layer section 1, tab layer 2 corresponds to tab layer section 2, and so on, and the tab to be measured includes 10 tab layer sections. Each two tab layers may correspond to one tab layer interval, for example, tab layer 1 and tab layer 2 should correspond to tab layer interval 1, tab layer 3 and tab layer 4 should correspond to tab layer interval 2, and so on, the tab to be measured may include 5 tab layer intervals. In some embodiments, the number of tab layers included in each tab layer interval may also be different.

The first exposure type image comprises a high exposure image, wherein the high exposure image refers to an image with a gray value larger than a preset gray value, and the contrast ratio is good, so that the first exposure type image can be used for measuring the distance. The high exposure image may be an image acquired with a vision sensor having a shutter speed not greater than a preset shutter speed, for example, a preset shutter speed of 1/200 seconds.

The conversion relationship between the real space and the image space corresponding to the first exposure type image may be a conversion relationship stored in the computer device in advance, or may be a conversion relationship obtained by calibration with a calibration piece before determining the initial polar edge distance.

Further, according to the first exposure type image and the conversion relation between the real space and the image space, the initial tab edge distance of the tab to be detected can be obtained.

Taking the example that the tab to be measured includes the tab a as an example, the initial tab margin a1 and the initial tab margin a2 of the tab a can be obtained according to the first exposure type image of the tab a and the conversion relationship between the real space and the image space. The initial tab margin a1 is used for indicating the distance between the edge position of the left side of the tab a and the edge of the battery cell, and the initial tab margin a2 is used for indicating the distance between the edge position of the right side of the tab a and the edge of the battery cell.

S602, determining a target tab layer interval corresponding to a second exposure type image from a plurality of tab layer intervals according to the detection model and the second exposure type image of the tab to be detected; the first exposure type image and the second exposure type image are shot by the vision sensor along the thickness direction of the electrode lug to be detected, the field of view range of the vision sensor covers the electrode lug to be detected, and the target electrode lug layer interval is the interval with the maximum offset of the edge in the thickness direction of the electrode lug to be detected.

In this embodiment, the computer device is capable of determining a detection model. The computer device may receive the trained detection model sent by the other device. In some embodiments, the detection model may be obtained by training the initial detection model based on the first tab image sample and a tab layer interval label corresponding to the first tab image sample.

It should be noted that, the first tab image sample is a second exposure type image. The second exposure type image comprises a low exposure image, wherein the low exposure image refers to an image with a gray value not larger than a preset gray value, and different layers of edges in the tab can be distinguished. The low exposure image may be an image acquired with a vision sensor having a shutter speed greater than a preset shutter speed. The first tab image sample is an image sample taken by the visual sensor along the thickness direction of the tab, and the field of view of the visual sensor needs to cover the tab.

The first tab image sample may include low exposure image samples corresponding to a plurality of tabs. Furthermore, the computer device may determine a tab layer interval tag corresponding to the first tab image sample. The tab layer interval label is used for indicating a tab layer interval where the edge position of the tab in the first tab image sample is located.

Fig. 7 is a schematic diagram of a tab layer interval tag in an embodiment of the present application. Taking one tab of each second exposure type image in the first tab image sample including the battery cell as an example, two tab layer interval labels corresponding to the left edge position and the right edge position of each tab can be determined. That is, the tab layer interval tag may include a first layer interval tag protruding toward the left side and a second layer interval tag protruding toward the right side. In other words, the first layer section label is a section in which the left side edge is most displaced in the thickness direction of the tab, and the second layer section label is a section in which the right side edge is most displaced in the thickness direction of the tab.

Each tab layer interval tag may include k layer interval tags of A1-Ak. With continued reference to fig. 3, A1 represents a first tab layer section closest to the vision sensor, A2 represents a second tab layer section below the first tab layer section, and Ak represents a kth tab layer section farthest from the vision sensor. k is an integer of 1 or more.

For example, in the lower right-hand low-exposure image sample in fig. 7, the tab layer section label on the left side of the tab is A5, which indicates that the edge position on the left side of the tab is in the 5 th tab layer section of the tab; in the lower right low exposure image sample, the tab layer section label on the right side of the tab is A1, which indicates that the edge position on the right side of the tab is in the 1 st tab layer section of the tab.

Optionally, the computer device may obtain a tab layer interval tag input by the user, so as to determine a tab layer interval tag corresponding to the first tab image sample. The computer equipment can also determine the tab layer interval label corresponding to the first tab image sample by using the inclination degree of the tab edge in a marking mode and the like. In some embodiments, the computer device may also determine, in combination with manual and automatic manners, a tab layer interval tag corresponding to the first tab image sample.

Further, the computer device may train the initial detection model based on the first tab image sample and the tab layer interval label corresponding to the first tab image sample to obtain the detection model. The initial detection model includes, but is not limited to, a supervised learning model, a semi-supervised learning model, an unsupervised learning model, and the like. For example, the initial detection model may include, but is not limited to, at least one of a convolutional neural network (Convolutional Neural Networks, CNN) model, a recurrent neural network (Recurrent Neural Network, RNN), a full convolutional neural network (Fully Convolutional Neural Network, FCN) model, a generative countermeasure network (Generative Adversarial Network, GAN) model, a Back-propagation (BP) machine learning model, a radial basis function (Radial Basis Function, RBF) model, a deep belief network (Deep Belief Networks, DBN) model, an Elman model, or a combination thereof.

The computer device may input the tab layer interval label based on the first tab image sample and the tab layer interval label corresponding to the first tab image sample to the initial detection model, so as to train the initial detection model, until the accuracy between the tab layer interval and the tab layer interval label predicted by the initial detection model reaches the preset accuracy, and further stop training to obtain the detection model.

Thus, the detection model has the capability of determining the tab layer interval corresponding to the second exposure type image according to the second exposure type image. Furthermore, the computer device can determine a target tab layer interval corresponding to the second exposure type image according to the detection model and the second exposure type image of the tab to be detected.

It should be noted that, the first exposure type image and the second exposure type image are taken by the vision sensor along the thickness direction of the tab to be measured. And, the field of view scope of vision sensor needs to cover the utmost point ear that awaits measuring. Thus, the first exposure type image and the second exposure type image can represent the edge position of the tab to be tested.

The computer device may directly input the second exposure type image of the tab to be detected to the detection model to obtain a target tab layer interval output by the detection model. In some embodiments, the computer device may also perform preprocessing on the second exposure type image of the tab to be detected, and then directly input the second exposure type image into the detection model, so as to obtain a target tab layer interval output by the detection model.

The target tab layer interval is an interval with the maximum offset of the edge in the thickness direction of the tab to be measured. That is, the target tab layer interval is used for indicating the tab layer area where the edge position of the tab to be measured is located.

Likewise, the target tab layer interval may also include a first tab layer interval and a second tab layer interval; the first tab layer section is a section with the largest offset of the left side edge in the thickness direction of the tab to be measured, and the second tab layer section is a section with the largest offset of the right side edge in the thickness direction of the tab to be measured. Referring to fig. 3, taking fig. 3 as an example, the first tab layer section is used to indicate the tab layer section where the edge position on the left side of the tab 301 is located, that is, the tab layer section where the black point is located, for example, as shown by the bold dashed line. Similarly, if the second tab layer section protrudes in the right direction, the second tab layer section is used to indicate the tab layer section where the edge position on the right side of the tab 301 is located, and the second tab layer section is, for example, a tab layer section corresponding to the black region.

It should be noted that, in some embodiments, the first tab layer interval and the second tab layer interval may be the same. That is, the left edge position and the right edge position of the tab may be located in the same tab layer section.

In an exemplary embodiment, optionally, the first exposure type image and the second exposure type image are images captured by the vision sensor under the condition that the position of the tab to be measured is unchanged.

For example, after the battery cell reaches the detection station on the basis of fig. 1, the battery cell and the vision sensor remain relatively stationary, so that the vision sensor can acquire the first exposure type image and the second exposure type image at two moments respectively. Thus, the error between the first exposure type image and the second exposure type image is reduced, and the accuracy of the tab edge distance is improved.

S603, correcting the initial tab edge distance according to the target tab layer interval to obtain a target tab edge distance of the tab to be detected; the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

In this embodiment, after the target tab layer interval is obtained, the computer device may correct the initial tab edge distance by using the target tab layer interval to obtain the target tab edge distance of the tab to be measured.

Continuing with the above example, after the initial tab margin a1 and the initial tab margin a2 of the tab a are obtained, the initial tab margin a1 may be corrected by using the first tab layer section to obtain the target tab margin a1', and the initial tab margin a1 may be corrected by using the second tab layer section to obtain the target tab margin a2'.

Optionally, the computer device may determine a relationship between the target tab layer interval and the tab layer interval where the calibration sheet is located, and determine a corrected conversion relationship according to the relationship between the target tab layer interval and the tab layer interval where the calibration sheet is located, and further correct the initial tab edge distance by using the corrected conversion relationship, so as to obtain the target tab edge distance.

Optionally, the computer device may determine a distance between the target tab layer interval and the tab layer interval where the calibration sheet is located, and correct the conversion relationship between the real space and the image space according to the distance between the target tab layer interval and the tab layer interval where the calibration sheet is located, to obtain a corrected conversion relationship, and further correct the initial tab edge distance by using the corrected conversion relationship, so as to obtain the target tab edge distance.

Further, after the target tab edge distance is obtained, tab dislocation detection can be performed, that is, whether the tab to be detected is dislocated or not is confirmed according to the target tab edge distance. Optionally, the computer device may determine that the tab has no dislocation when the target tab edge distance is located in the preset range, and determine that the tab has a dislocation when the target tab edge distance is not located in the preset range. In some embodiments, the computer device may also determine that the battery cell has no tab misalignment according to a case that target tab pitches of two tabs on the same battery cell are both within a preset range, and otherwise, the battery cell has tab misalignment.

In the tab dislocation detection method, the initial tab edge distance of the tab to be detected is determined according to the conversion relation among the first exposure type image, the real space and the image space corresponding to the first exposure type image of the tab to be detected, and the target tab layer interval corresponding to the second exposure type image is determined according to the detection model and the second exposure type image of the tab to be detected. Because the first exposure type image and the second exposure type image are shot by the vision sensor along the thickness direction of the tab to be detected, the field of view range of the vision sensor covers the tab to be detected, the tab to be detected comprises a plurality of tab layer intervals, the tab layer interval comprises at least one tab layer, the target tab layer interval is the interval with the largest offset of the edge in the thickness direction of the tab to be detected, therefore, the determined target tab layer interval can reflect the tab layer interval in which the most convex edge position of the tab is actually located, the initial tab edge distance is corrected according to the target tab layer interval, and after the target tab edge distance of the tab to be detected is obtained, the influence caused by the inconsistency between the tab layer interval in which the most convex edge position is located and the tab layer interval in which the calibration piece is located can be reduced, so that the error of the initial tab edge distance is reduced, the accuracy of the target tab edge distance is improved, and the accuracy of tab dislocation detection is improved.

Fig. 8 is a schematic flow chart of obtaining a target tab edge distance in an embodiment of the present application, and in an exemplary embodiment, as shown in fig. 8, S603 includes S801 to S803.

S801, determining a height difference value between a target tab layer interval and a calibration tab layer interval; and the tab layer interval is the tab layer interval where the marked piece is in the process of calibrating the real space and the image space and obtaining the conversion relation.

In the present embodiment, in order to obtain a conversion relationship between a real space and an image space, the real space and the image space need to be calibrated by using a calibration piece. In the calibration process, the calibration sheet is placed on a certain tab layer of the tab, and an image of the calibration sheet is acquired by using a visual sensor, so that the conversion relationship between the real space and the image space is determined according to the actual size information of the calibration sheet and the image of the calibration sheet.

And the calibration tab layer interval is a tab layer interval where the calibration tab is located in the process of calibrating the real space and the image space to obtain the conversion relation. Continuing with the above example, assuming that the calibration sheet is placed on the outermost layer of the tab during the calibration process, the calibration tab layer interval may be A1.

Further, the computer device may determine a height difference between the target tab layer interval and the nominal tab layer interval. Optionally, the computer device may determine the height of the target tab layer interval and the height of the calibration tab layer interval, and further make a difference between the height of the target tab layer interval and the height of the calibration tab layer interval, so as to determine a height difference between the target tab layer interval and the calibration tab layer interval. For example, assuming a height of 5 mm for the target tab layer section and a height of 6 mm for the nominal tab layer section, the height difference is 1 mm.

S802, determining a correction value according to the field angle and the height difference value of the vision sensor.

In this embodiment, the field angle of the vision sensor may be stored in the computer device in advance, or may be determined by the computer device according to the type of the vision sensor. Illustratively, the field angle of the vision sensor is 10 °.

Further, the computer device may determine the correction value based on the field angle and the height difference of the vision sensor. Alternatively, the computer device may use the product of the tangent value of the angle of view of the vision sensor and the height difference value as the correction value, or may obtain the correction value after performing correction processing on the product of the tangent value of the angle of view and the height difference value.

S803, correcting the initial tab edge distance according to the correction value to obtain the target tab edge distance.

And then, the computer equipment can correct the initial tab edge distance by using the correction value so as to obtain the target tab edge distance. Optionally, since the calibration sheet is generally disposed on the tab layer of the outermost layer of the tab, the target tab layer interval is far away from the vision sensor compared with the calibration tab layer interval, and according to the principle of far-small near-large during imaging, the determined initial tab margin is smaller than the real tab margin, so that the computer device can take the sum of the initial tab margin and the correction value as the target tab margin.

In an exemplary embodiment, optionally, S801 may be implemented as follows: determining a first height of the target tab layer interval according to the heights of the tab layers included in the target tab layer interval; and taking the difference value between the first height and the second height of the calibrated tab layer interval as a height difference value.

In this embodiment, if the target tab layer section includes only one tab layer, the height corresponding to the tab layer may be used as the first height of the target tab layer section, or the height corresponding to the tab layer may be rounded and then used as the first height of the target tab layer section. For example, assuming that the target tab layer interval is A1 and the tab layer interval A1 includes only one tab layer 1, the computer device may regard the height of the tab layer 1 as the first height of the target tab layer interval A1.

If the target tab layer interval includes a plurality of tab layers, the average value, the median, and the weighted average value of heights corresponding to the target tab layer interval including the plurality of tab layers may be taken as the first height of the target tab layer interval. For example, assuming that the target tab layer interval is A1 and the tab layer interval A1 includes one tab layer 1 and one tab layer 2, the computer device may take an average of the height of the tab layer 1 and the height of the tab layer 2 as the first height of the target tab layer interval A1.

The height corresponding to each tab layer may be a parameter stored in the computer device in advance, or may be a parameter input by a user.

Further, the computer device knows the second height of the nominal tab layer interval, and therefore, after determining the first height of the target tab layer interval, the difference between the first height and the second height can be used as the height difference.

Fig. 9 is a second schematic flow chart of obtaining a target tab edge distance in the embodiment of the present application, and in an exemplary embodiment, as shown in fig. 9, S803 includes S901 to S902.

S901, determining the position relationship between the target tab layer interval and the calibrated tab layer interval and the visual sensor.

In this embodiment, the timing sheet may be located in any tab layer section where the tab is located at the time of the calibration. Therefore, in order to improve the accuracy of the obtained tab margin, the computer device needs to determine the target tab layer interval and the positional relationship between the calibrated tab layer interval and the visual sensor. The position relationship is used for indicating the distance between the target tab layer interval and the calibration tab layer interval relative to the visual sensor.

Optionally, the computer device may determine the above positional relationship according to the target tab layer interval and the calibration tab layer interval. Continuing with the example above, the target tab layer interval is A2 and the nominal tab layer interval is A1, then the nominal tab layer interval is closer to the visual sensor than the target tab layer interval.

S902, correcting the initial tab edge distance according to the correction value and the position relation to obtain the target tab edge distance.

Further, after the above-mentioned position relation is determined, the initial tab margin is corrected by using the correction value and the position relation, so as to obtain the target tab margin.

Optionally, the computer device may use a difference between the initial tab margin and the correction value as the target tab margin when the target tab layer interval is closer to the visual sensor; and (3) the interval between the calibration lug layers is closer to the visual sensor, and the sum of the initial lug edge distance and the correction value is taken as the target lug edge distance.

Fig. 10 is a third schematic flow chart of obtaining the target tab edge distance in the embodiment of the present application, and in an exemplary embodiment, as shown in fig. 10, S902 includes S1001 to S1002.

S1001, if the position relation is that the target tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the difference between the initial tab edge distance and the correction value.

In this embodiment, if the positional relationship is that the target tab layer interval is close to the vision sensor, according to the principle of far, near and large, the determined initial tab edge distance is larger than the actual tab edge distance, so that the computer device can obtain the target tab edge distance according to the difference between the initial tab edge distance and the correction value.

For example, the computer device may take m× (initial tab margin-correction value) as the target tab margin. m may be an empirical value between 0 and 1.

S1002, if the position relation is that the calibrated tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the sum of the initial tab edge distance and the correction value.

Similar to S1001, if the positional relationship is that the calibrated tab layer interval is close to the vision sensor, according to the principle of far, near and big, the determined initial tab edge distance will be smaller than the actual tab edge distance, so the computer device can obtain the target tab edge distance according to the sum of the initial tab edge distance and the correction value.

For example, the computer device may take n× (initial tab margin+correction value) as the target tab margin. n may be an empirical value of 1 or more.

Since the detection model needs to be used when the target tab edge distance of the tab to be detected is obtained, and the detection model needs to be trained before the detection model is used, the process of training to obtain the detection model in the embodiment of the application is developed as follows.

Fig. 11 is a schematic flow chart of obtaining a detection model in the embodiment of the present application, and in an exemplary embodiment, as shown in fig. 11, the above-mentioned tab misalignment detection method further includes S1101 to S1103.

S1101, acquiring a first tab image sample and tab layer interval labels corresponding to the first tab image sample.

In this embodiment, the computer device first needs to obtain the first tab image sample and the tab layer interval tag corresponding to the first tab image sample.

The first tab image sample may include at least one tab. The computer equipment can collect a large number of high exposure images corresponding to actual lugs to serve as first lug image samples, and can simulate and generate the high exposure images corresponding to the lugs to serve as the first lug image samples. In some embodiments, the first tab image sample may also include an actual tab corresponding high exposure image and a simulated tab corresponding high exposure image.

Further optionally, the computer device may determine, manually and/or automatically, a tab layer interval tag corresponding to the first tab image sample. The process of determining the tab layer interval label may refer to S602, which is not described herein.

S1102, processing the first tab image sample to obtain a second tab image sample.

In this embodiment, the computer device may perform processing manners including, but not limited to, image enhancement processing, clipping processing, background blurring processing, noise reduction processing, filtering processing, and the like on the first tab image sample to obtain the second tab image sample.

S1103, training the initial detection model according to the second tab image sample and the tab layer interval label corresponding to the second tab image sample to obtain a detection model.

In this embodiment, since the second tab image sample is obtained by processing the first tab image sample, the second tab image sample corresponds to the first tab image sample, and further, the tab layer section label corresponding to the first tab image sample, that is, the tab layer section label corresponding to the second tab image sample is determined.

Furthermore, the computer equipment can train the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain the detection model. The computer device may input the tab layer interval label corresponding to the second tab image sample and the second tab image sample to the initial detection model to train the initial detection model until the accuracy between the tab layer interval corresponding to the second tab image sample and the tab layer interval label predicted by the initial detection model reaches the preset accuracy, and further stop training to obtain the detection model.

It will be appreciated that in some embodiments, the initial detection model is trained to obtain the detection model based on the second tab image sample and the tab layer interval label corresponding to the second tab image sample. Therefore, when the detection model is used, the computer device can also process the second exposure type image of the tab to be detected, and then input the processed second exposure type image into the detection model to obtain the target tab layer interval output by the detection model. The method for processing the second exposure type image of the tab to be tested may be the same as the method for processing the first tab image sample.

Fig. 12 is a second schematic flow chart of obtaining a detection model in the embodiment of the present application, and in an exemplary embodiment, as shown in fig. 12, S1103 includes S1201 to S1203.

And S1201, training the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain a plurality of candidate detection models.

In this embodiment, the number of initial detection models may be multiple, as an option. The computer equipment can train each initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample so as to obtain a candidate detection model corresponding to each initial detection model. Further alternatively, the type of each initial detection model may be different.

In some embodiments, the number of the initial detection models may be 1, and further, when the computer device trains the initial detection models according to the second tab image samples and tab layer interval labels corresponding to the second tab image samples, training may be stopped under different iteration times or different model parameters, so as to obtain a plurality of candidate detection models.

S1202, determining quantized values of each candidate detection model.

Further, after determining the plurality of candidate detection models, the computer device may also determine a quantization value for each candidate detection model. Wherein the quantized values are used to evaluate each candidate detection model. The evaluation dimension includes, but is not limited to, detection accuracy, detection false positive rate, detection speed, etc. of the candidate detection model.

Alternatively, the computer device may determine the quantized values of each candidate detection model by using an evaluation model, manual evaluation, or the like.

S1203, determining a detection model from the candidate detection models according to the quantized values of the candidate detection models.

Still further, the computer device may determine a detection model from among the candidate detection models based on the quantized values of the candidate detection models. For example, the computer device may take the candidate detection model with the highest quantization value as the end-use detection model.

In an exemplary embodiment, optionally, the computer device may further update the detection model according to the second exposure type images of the plurality of tabs to be detected. That is, the computer device may continuously update the detection model during use of the detection model to improve accuracy of the detection model.

Fig. 13 is a schematic diagram of a process of obtaining a detection model in the embodiment of the present application, and as shown in fig. 13, after the computer device sequentially performs data labeling, data preprocessing, model selection, model training, and model evaluation and optimization, practical application can be performed. And the data labeling is the process of acquiring the first tab image sample and the tab layer interval label corresponding to the first tab image sample. And (3) preprocessing the data, namely processing the first lug image sample to obtain a second lug image sample. Model selection, i.e., the process of determining an initial test model. And training the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain a detection model. And (3) evaluating and optimizing the model, namely determining the process of the detection model from the candidate detection models according to the quantized value of each candidate detection model. Further, practical applications can be understood as executing the processes of S601 to S603.

In order to more clearly describe the tab misalignment detection method in the present application, it is described with reference to fig. 14. Fig. 14 is a schematic process diagram of a tab misalignment detection method in an embodiment of the present application, and as shown in fig. 14, a computer device may execute the method according to the following flow.

S1401, acquiring a first tab image sample and a tab layer interval label corresponding to the first tab image sample.

S1402, processing the first tab image sample to obtain a second tab image sample.

S1403, training the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain a plurality of candidate detection models.

S1404, determining quantized values of each candidate detection model.

S1405, determining a detection model from the candidate detection models according to the quantized values of the candidate detection models.

S1406, determining the initial tab edge distance of the tab to be measured according to the conversion relation among the first exposure type image, the real space and the image space corresponding to the first exposure type image.

S1407, determining a target tab layer interval corresponding to the second exposure type image according to the detection model and the second exposure type image of the tab to be detected. The first exposure type image and the second exposure type image are images shot by the vision sensor under the condition that the position of the tab to be detected is unchanged.

S1408, determining a height difference between the target tab layer interval and the calibration tab layer interval. The calibration tab layer interval is a tab layer interval where the calibration tab is located in the process of calibrating the real space and the image space to obtain the conversion relation. Optionally, the first height of the target tab layer interval may be determined according to the height of each tab layer included in the target tab layer interval, and a difference between the first height and the second height of the target tab layer interval is used as a height difference.

S1409, determining a correction value according to the angle of view and the height difference value of the vision sensor.

S1410, determining the position relationship between the target tab layer interval and the calibration tab layer interval and the visual sensor.

S1411, if the position relation is that the target tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the difference between the initial tab edge distance and the correction value.

And S1412, if the position relationship is that the calibrated tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the sum of the initial tab edge distance and the correction value.

Wherein S1401-S1405 are training processes. S1406-S1412 are the use processes. S1401 to S1412 may refer to the above embodiments, and are not described herein. Therefore, according to the tab dislocation detection method provided by the embodiment, the target tab layer interval can be identified through the modeling model, so that tab dislocation is dynamically compensated by using the target tab layer interval, the height error is reduced to 1/2k, and the determination accuracy of tab edge distance is improved.

It should be noted that, the application describes with detecting tab dislocation as application scene, and the tab dislocation detection method provided by the application can also be suitable for other tab bad detection scenes.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides a tab misalignment detection system for realizing the tab misalignment detection method. Fig. 15 is a schematic structural diagram of a tab misalignment detection system according to an embodiment of the present application, and as shown in fig. 15, the tab misalignment detection system 1500 includes a vision sensor 1501, an upper computer 1502 and a light source 1503.

The light source 1503 includes a first light source 1503a and a second light source 1503b, and the light source 1503 is disposed in a first direction of a thickness of the electrical core to which the tab to be measured belongs. The first direction is, for example, above the thickness of the cell to which the tab to be measured belongs.

The light of the first light source 1503a irradiates the cell to which the tab to be measured belongs, and is used for supplementing light to the cell. Optionally, the first light source 1503a may be disposed between the vision sensor 1501 and the electrical core to which the tab to be measured belongs. Further alternatively, the number of the first light sources 1503a may be plural. For example, four first light sources 1503a may be disposed around the electric core of the tab to be measured and between the visual sensor 1501 and the electric core of the tab to be measured.

The light of the second light source 1503b irradiates the tab to be measured, and is used for supplementing light to the tab to be measured. Optionally, a second light source 1503b may also be disposed between the vision sensor 1501 and the cell of the tab to be measured. It can be appreciated that, in order to supplement the tab to be measured, the second light source 1503b needs to be close to the tab to be measured. Optionally, the distance between the second light source 1503b and the tab to be measured is smaller than the preset distance. Further alternatively, the number of the second light sources 1503b may be plural.

In this embodiment, since the second exposure type image has a phenomenon of insufficient brightness caused by the stepping down of the tab during the actual use, and affects the measurement of the tab margin, the light source can be used for supplementing light to the tab to be measured, so as to improve the accuracy of the second exposure type image.

Each tab to be measured corresponds to one visual sensor 1501, and the visual sensor 1501 is used for acquiring a first exposure type image and a second exposure type image of the tab to be measured. Each tab to be measured may correspond to the same visual sensor 1501, or each tab to be measured may correspond to a different visual sensor 1501.

The upper computer 1502 is configured to determine an initial tab edge distance of a tab to be measured according to a conversion relationship between a first exposure type image, a real space and an image space corresponding to the first exposure type image of the tab to be measured, and determine a target tab layer interval corresponding to the second exposure type image from a plurality of tab layer intervals according to a detection model and a second exposure type image of the tab to be measured, so as to correct the initial tab edge distance according to the target tab layer interval, and obtain a target tab edge distance of the tab to be measured; the first exposure type image and the second exposure type image are shot by a visual sensor along the thickness direction of the tab to be detected, the field of view range of the visual sensor covers the tab to be detected, and the target tab layer interval is the interval with the maximum offset of the edge in the thickness direction of the tab to be detected; the electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer; the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

Fig. 16 is a schematic structural diagram of a tab misalignment detection system in the embodiment of the present application, where fig. 16 (a) illustrates a side view of the tab misalignment detection system, fig. 16 (b) illustrates a side view of the tab misalignment detection system, and as shown in fig. 16, four light sources 1503 may be respectively disposed around a cell of a tab to be detected and between a visual sensor 1501 and the cell. The visual sensor 1501 can capture images along the thickness direction of the tab to be measured, and the field of view range of the visual sensor 1501 covers the tab to be measured.

The principles of the tab misalignment detection system 1500 may refer to the above embodiments, and are not described herein.

In an exemplary embodiment, optionally, the first exposure type image is captured by the vision sensor with a first brightness of the light source and the second exposure type image is captured by the vision sensor with a second brightness of the light source, wherein the first brightness is greater than the second brightness.

That is, the first exposure type image may be acquired with the vision sensor 1501 in the case where the luminance of the light source 1503 is higher luminance, and the second exposure type image may be acquired with the vision sensor 1501 in the case where the luminance of the light source 1503 is lower luminance.

In one exemplary embodiment, optionally, the first exposure type image is photographed with the brightness of the light source being the third brightness, the shutter speed of the vision sensor being the first speed, and the second exposure type image is photographed with the brightness of the light source being the third brightness, the shutter speed of the vision sensor being the second speed, wherein the first speed is smaller than the second speed.

That is, the shutter speed of the visual sensor 1501 may be decreased to acquire the first exposure type image and the shutter speed of the visual sensor 1501 may be increased to acquire the second exposure type image with the luminance of the light source 1503 unchanged.

In an exemplary embodiment, optionally, the light source further includes a third light source, where the third light source is disposed in a second direction of a thickness of the electrical core to which the tab to be measured belongs, and the second direction is opposite to the first direction. For example, if the first direction is above the thickness of the cell to which the tab to be measured belongs, the second direction is above the thickness of the cell to which the tab to be measured belongs. That is, the tab to be measured may be located between the vision sensor and the third light source.

The first exposure type image is photographed by a visual sensor with a third light source at a fourth brightness, the second exposure type image is photographed by a visual sensor with a second brightness of the light source, or the second exposure type image is photographed with a shutter speed of the visual sensor at a second speed of the light source at the third brightness.

That is, in the case where the third light source is of the fourth brightness, the tab to be measured is in the backlight, and at this time, a high exposure image can be obtained by the vision sensor. Further, the low exposure image may be acquired by the vision sensor in the case where the light source is at the second brightness, or in the case where the second exposure type image is the light source and the brightness is the third brightness.

It is understood that the first brightness, the second brightness, the third brightness, the fourth brightness, and the first speed and the second speed may be set according to the requirements.

In an exemplary embodiment, optionally, two tabs to be tested correspond to the same visual sensor. That is, for the left and right tabs of the same cell, the same vision sensor may be used to acquire the first and second exposure type images. Because two pole lugs to be measured correspond to the same visual sensor, the hardware cost is reduced.

In an exemplary embodiment, optionally, the height between the visual sensor 1501 and the tab to be measured is greater than a preset height.

In this embodiment, as can be seen from fig. 1 and 2, in the case that the height difference between the target tab layer section and the calibration tab layer is 7.5mm, the 5 ° field angle error is reduced by 0.67mm compared to the 10 ° field angle error. Therefore, in order to improve accuracy of the target tab margin, the angle of view of the vision sensor 1501 may be reduced.

Fig. 17 is a schematic diagram illustrating the use of a visual sensor according to an embodiment of the present application, and in an exemplary embodiment, as shown in fig. 17, the height between the visual sensor 1501 and the tab to be measured may be increased, so that the height between the visual sensor 1501 and the tab to be measured is greater than the preset height. The preset height can be set according to requirements and is a number larger than 0. Illustratively, the height between the vision sensor 1501 and the tab to be measured may be increased from 400 millimeters to 600 millimeters to reduce the field angle of 3 °.

In an exemplary embodiment, optionally, the two tabs to be tested correspond to different visual sensors, respectively.

Fig. 18 is a second schematic diagram of the use of a visual sensor in an embodiment of the present application, and in an exemplary embodiment, as shown in fig. 18, the number of visual sensors may be increased, so that each tab to be tested corresponds to at least one visual sensor 1501. For example, changing the visual sensor from facing the middle of two lugs to facing the lug can reduce the angle of view by half.

In some embodiments, the height between the visual sensor 1501 and the tab to be measured may be increased, and the number of visual sensors may be increased, so that the two tabs to be measured correspond to different visual sensors 1501, respectively.

In the above-described embodiment, since the angle of view of the vision sensor is reduced, the accuracy of the first exposure type image and the second exposure type image is improved.

Fig. 19 is a second schematic structural view of a tab misalignment detection system according to an embodiment of the present application, and in an exemplary embodiment, as shown in fig. 19, the tab misalignment detection system 1500 further includes a conveying assembly 1901, a carrying assembly 1902, a bar code detection device 1903, and a bracket 1904.

Wherein a support 1904 is used to support the vision sensor 1501 and the light source 1503. The carrying component 1902 is used for carrying the battery cell to which the tab to be tested belongs. Carrier assembly 1902 includes, but is not limited to, a tray. The transmission component 1901 is configured to transmit the carrier component 1902 carrying the electrical core to a preset image acquisition position corresponding to the vision sensor. The conveyor assembly 1901 includes, but is not limited to, a conveyor belt. The preset image acquisition locations corresponding to the vision sensors may include, but are not limited to, detection stations on the conveyor assembly 1901.

Referring to fig. 1, after the carrier assembly 1902 with the battery cells is placed on the transfer assembly 1901, the transfer assembly 1901 transfers the carrier assembly 1902 to a preset image acquisition position corresponding to the vision sensor during the movement process. Alternatively, the distance between stations on the transfer module 1901 may be fixed or preset so that the transfer module 1901 can stop moving and send an arrival signal to the host computer 1502 with each previous station.

The bar code detection device 1903 is configured to obtain the identification of the battery cell and send the identification to the upper computer 1502 when the battery cell is transmitted. The bar code detection device 1903 may be disposed at the detection station, may be disposed at a station previous to the detection station, and may be disposed at other positions of the conveying assembly 1901.

The battery cell can be attached with a unique identifier of the battery cell, and the identifier of the battery cell can comprise at least one of characters, numbers, symbols or letters. Alternatively, the bar code detection device may be provided with a camera, a bar code detector, or the like for detecting the identification of the battery cell. Further, the bar code detection device 1903 may acquire the identification of the battery cell in the case where the transmission assembly 1901 transmits the battery cell.

Further, the bar code detection device 1903 sends the identification of the battery cell to the host computer 1502 after acquiring the identification of the battery cell.

The upper computer 1502 is configured to control the light source 1503 to perform light filling and control the vision sensor 1501 to acquire a first exposure type image and a second exposure type image when detecting that the cell corresponding to the identifier reaches a preset image acquisition position corresponding to the vision sensor.

Optionally, the upper computer 1502 may determine that the cell corresponding to the identifier reaches the preset image acquisition position corresponding to the visual sensor when receiving the identifier of the cell and the reaching signal sent by the transmission component 1901. Further, when the battery cell reaches the preset image acquisition position corresponding to the vision sensor, the upper computer 1502 can control the light source 1503 to perform light filling, and control the vision sensor 1501 to acquire the first exposure type image and the second exposure type image. For example, the upper computer 1501 acquires a first exposure type image and a second exposure type image when the first light source 1503a supplements light to the battery cell and the second light source 1503b supplements light to the tab to be measured.

Further, after acquiring the first exposure type image and the second exposure type image, the upper computer 1502 may control the transmission assembly 1901 to move continuously to drive the next cell to reach the detection station.

Based on the same inventive concept, the embodiment of the application also provides a tab misalignment detection device for realizing the tab misalignment detection method. The implementation scheme of the device for solving the problem is similar to that described in the above method, so the specific limitation of the embodiment of one or more tab misalignment detection devices provided below may be referred to the limitation of the tab misalignment detection method hereinabove, and will not be repeated herein.

Fig. 20 is a block diagram of a tab misalignment detection apparatus according to an embodiment of the present application, and in an exemplary embodiment, as shown in fig. 20, a tab misalignment detection apparatus 2000 is provided, including: a first determination module 2001, a second determination module 2002, and a correction module 2003, wherein:

the first determining module 2001 is configured to determine an initial tab edge distance of the tab to be measured according to a conversion relationship among the first exposure type image, the real space, and the image space corresponding to the first exposure type image of the tab to be measured.

The second determining module 2002 is configured to determine, according to the detection model and the second exposure type image of the tab to be detected, a target tab layer interval corresponding to the second exposure type image from the plurality of tab layer intervals; the first exposure type image and the second exposure type image are shot by the vision sensor along the thickness direction of the electrode lug to be detected, the field of view range of the vision sensor covers the electrode lug to be detected, and the target electrode lug layer interval is the interval with the maximum offset of the edge in the thickness direction of the electrode lug to be detected.

The correction module 2003 is used for correcting the initial tab edge distance according to the target tab layer interval to obtain the target tab edge distance of the tab to be detected; the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

In the tab dislocation detection device, the initial tab edge distance of the tab to be detected is determined according to the conversion relation among the first exposure type image, the real space and the image space corresponding to the first exposure type image of the tab to be detected, and the target tab layer interval corresponding to the second exposure type image is determined according to the detection model and the second exposure type image of the tab to be detected. Because the first exposure type image and the second exposure type image are shot by the vision sensor along the thickness direction of the tab to be detected, the field of view range of the vision sensor covers the tab to be detected, the tab to be detected comprises a plurality of tab layer intervals, the tab layer interval comprises at least one tab layer, the target tab layer interval is the interval with the largest offset of the edge in the thickness direction of the tab to be detected, therefore, the determined target tab layer interval can reflect the tab layer interval in which the most convex edge position of the tab is actually located, the initial tab edge distance is corrected according to the target tab layer interval, and after the target tab edge distance of the tab to be detected is obtained, the influence caused by the inconsistency between the tab layer interval in which the most convex edge position is located and the tab layer interval in which the calibration piece is located can be reduced, so that the error of the initial tab edge distance is reduced, the accuracy of the target tab edge distance is improved, and the accuracy of tab dislocation detection is improved.

Fig. 21 is a block diagram of the configuration of the correction module in the embodiment of the present application, and in an exemplary embodiment, as shown in fig. 21, the correction module 2003 includes:

a first determining unit 2101 for determining a height difference between the target tab layer section and the calibration tab layer section; and the tab layer interval is the tab layer interval where the marked piece is in the process of calibrating the real space and the image space and obtaining the conversion relation.

A second determining unit 2102 for determining a correction value based on the field angle and the height difference of the vision sensor.

And the correction unit 2103 is used for correcting the initial tab edge distance according to the correction value to obtain the target tab edge distance.

Fig. 22 is a block diagram of the first determining unit in the embodiment of the present application, and in an exemplary embodiment, as shown in fig. 22, the first determining unit 2101 includes:

a first determining subunit 2201, configured to determine a first height of the target tab layer interval according to the heights of the tab layers included in the target tab layer interval;

a second determining subunit 2202, configured to take a difference between the first height and the second height of the nominal tab layer interval as a height difference.

Fig. 23 is a block diagram of the configuration of the correction unit in the embodiment of the present application, and in an exemplary embodiment, as shown in fig. 23, the correction unit 2103 includes:

And a third determining subunit 2301, configured to determine a target tab layer interval and a positional relationship between the calibration tab layer interval and the vision sensor.

And a correction subunit 2302, configured to correct the initial tab edge distance according to the correction value and the position relationship, so as to obtain the target tab edge distance.

In an exemplary embodiment, the correction subunit 2302 is further configured to obtain the target tab margin according to the difference between the initial tab margin and the correction value if the position relationship is that the target tab layer interval is close to the vision sensor; and if the position relation is that the calibrated tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the sum of the initial tab edge distance and the correction value.

Fig. 24 is a second block diagram of a tab misalignment detection apparatus according to an embodiment of the present application, and in an exemplary embodiment, as shown in fig. 24, the tab misalignment detection apparatus 2000 further includes:

the acquiring module 2401 is configured to acquire a first tab image sample and a tab layer interval tag corresponding to the first tab image sample.

And the processing module 2402 is configured to process the first tab image sample to obtain a second tab image sample.

The training module 2403 is configured to train the initial detection model to obtain a detection model according to the second tab image sample and the tab layer interval label corresponding to the second tab image sample.

FIG. 25 is a block diagram of a training module in an embodiment of the present application, in an exemplary embodiment, as shown in FIG. 25, the training module 2403 includes:

the training unit 2501 is configured to train the initial detection model according to the second tab image sample and the tab layer interval label corresponding to the second tab image sample, so as to obtain a plurality of candidate detection models.

A third determination unit 2502 for determining a quantized value of each candidate detection model.

A fourth determination unit 2503 for determining a detection model from among the candidate detection models according to the quantized values of the candidate detection models.

Optionally, the first exposure type image and the second exposure type image are images shot by a vision sensor under the condition that the position of the tab to be detected is unchanged.

All or part of each module in the tab dislocation detection device can be realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

In one embodiment, the processor when executing the computer program further performs the steps of:

determining a height difference between the target tab layer interval and the calibration tab layer interval; calibrating the tab layer interval, namely calibrating the real space and the image space to obtain the tab layer interval where the calibration sheet is positioned in the process of converting the relationship; determining a correction value according to the field angle and the height difference value of the vision sensor; and correcting the initial tab edge distance according to the corrected value to obtain the target tab edge distance.

determining a first height of the target tab layer interval according to the heights of the tab layers included in the target tab layer interval; and taking the difference value between the first height and the second height of the calibrated tab layer interval as a height difference value.

determining the position relationship between the target tab layer interval and the calibrated tab layer interval and the visual sensor; and correcting the initial tab edge distance according to the corrected value and the position relation to obtain the target tab edge distance.

if the position relation is that the target tab layer interval is close to the visual sensor, obtaining a target tab edge distance according to the difference between the initial tab edge distance and the correction value; and if the position relation is that the calibrated tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the sum of the initial tab edge distance and the correction value.

acquiring a first tab image sample and a tab layer interval label corresponding to the first tab image sample; processing the first lug image sample to obtain a second lug image sample; and training the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain a detection model.

training the initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain a plurality of candidate detection models; determining quantized values of each candidate detection model; and determining a detection model from the candidate detection models according to the quantized values of the candidate detection models.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims

1. The tab dislocation detection method is characterized by comprising the following steps of:

determining an initial tab edge distance of a tab to be tested according to a conversion relation among a first exposure type image, a real space and an image space corresponding to the first exposure type image of the tab to be tested; the electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer;

Determining a target tab layer interval corresponding to the second exposure type image from the tab layer intervals according to the detection model and the second exposure type image of the tab to be detected; the first exposure type image and the second exposure type image are shot by a visual sensor along the thickness direction of the tab to be detected, the field of view range of the visual sensor covers the tab to be detected, and the target tab layer interval is an interval with the maximum offset of the edge in the thickness direction of the tab to be detected;

correcting the initial tab edge distance according to the target tab layer interval to obtain a target tab edge distance of the tab to be measured; and the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

2. The method of claim 1, wherein the correcting the initial tab edge distance according to the target tab layer interval to obtain the target tab edge distance of the tab to be measured comprises:

determining a height difference between the target tab layer interval and the calibration tab layer interval; the tab layer interval is a tab layer interval where a calibration sheet is located in the process of calibrating the real space and the image space to obtain the conversion relation;

Determining a correction value based on the field angle of the vision sensor and the height difference value;

and correcting the initial tab edge distance according to the correction value to obtain the target tab edge distance.

3. The method of claim 2, wherein the determining a height difference between the target tab layer interval and a nominal tab layer interval comprises:

and taking the difference value between the first height and the second height of the calibrated tab layer interval as the height difference value.

4. A method according to claim 2 or 3, wherein said correcting said initial tab margin according to said correction value to obtain said target tab margin comprises:

determining the position relationship between the target tab layer interval and the visual sensor as well as the calibration tab layer interval;

and correcting the initial tab edge distance according to the correction value and the position relation to obtain the target tab edge distance.

5. The method of claim 4, wherein correcting the initial tab margin according to the correction value and the positional relationship to obtain the target tab margin comprises:

If the position relation is that the target tab layer interval is close to the visual sensor, obtaining the target tab edge distance according to the difference between the initial tab edge distance and the correction value;

6. A method according to any one of claims 1-3, wherein the method further comprises:

processing the first lug image sample to obtain a second lug image sample;

and training an initial detection model according to the second lug image sample and the lug layer interval label corresponding to the second lug image sample to obtain the detection model.

7. The method of claim 6, wherein training the initial detection model according to the second tab image sample and the tab layer interval label corresponding to the second tab image sample to obtain the detection model comprises:

Training the initial detection model according to the second lug image sample and a lug layer interval label corresponding to the second lug image sample to obtain a plurality of candidate detection models;

determining a quantized value of each candidate detection model;

and determining the detection model from the candidate detection models according to the quantized value of each candidate detection model.

8. A method according to any one of claims 1-3, wherein the first exposure type image and the second exposure type image are images captured by the vision sensor with the position of the tab to be measured unchanged.

9. The tab dislocation detection system is characterized by comprising a light source, a visual sensor and an upper computer, wherein the light source comprises a first light source and a second light source, and the light source is arranged in a first direction of the thickness of a battery cell to which a tab to be detected belongs;

each lug to be detected corresponds to one visual sensor, and the visual sensor is used for acquiring a first exposure type image and a second exposure type image of the lug to be detected;

the light of the first light source irradiates the battery cell to which the tab to be detected belongs and is used for supplementing light to the battery cell, and the light of the second light source irradiates the tab to be detected and is used for supplementing light to the tab to be detected;

The upper computer is used for determining an initial tab edge distance of the tab to be detected according to a conversion relation among a first exposure type image, a real space and an image space corresponding to the first exposure type image of the tab to be detected, determining a target tab layer interval corresponding to the second exposure type image from a plurality of tab layer intervals according to a detection model and a second exposure type image of the tab to be detected, and correcting the initial tab edge distance according to the target tab layer interval to obtain a target tab edge distance of the tab to be detected; the first exposure type image and the second exposure type image are shot by a visual sensor along the thickness direction of the tab to be detected, the field of view range of the visual sensor covers the tab to be detected, and the target tab layer interval is an interval with the maximum offset of the edge in the thickness direction of the tab to be detected; the electrode lug to be tested comprises a plurality of electrode lug layer intervals, and each electrode lug layer interval comprises at least one electrode lug layer; and the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

10. The system of claim 9, wherein the first exposure type image is captured by the vision sensor with a first brightness of the light source and the second exposure type image is captured by the vision sensor with a second brightness of the light source, wherein the first brightness is greater than the second brightness; or,

The first exposure type image is photographed when the brightness of the light source is a third brightness, the shutter speed of the vision sensor is photographed when the first speed is smaller than the second speed, and the second exposure type image is photographed when the brightness of the light source is the third brightness, the shutter speed of the vision sensor is photographed when the second speed is smaller than the first speed; or,

the light source further comprises a third light source, the third light source is arranged in a second direction of the thickness of the battery cell to which the tab to be measured belongs, and the second direction is opposite to the first direction; the first exposure type image is shot by the vision sensor under the condition of fourth brightness of the third light source, the second exposure type image is shot by the vision sensor under the condition of second brightness of the light source, or the second exposure type image is shot under the condition of second speed of the shutter speed of the vision sensor under the condition of third brightness of the light source.

11. The system according to claim 9 or 10, wherein two of the tabs to be measured correspond to one and the same visual sensor, or two of the tabs to be measured correspond to different visual sensors, respectively.

12. The system of claim 9 or 10, further comprising a conveyor assembly, a carrier assembly, a bar code detection device, and a rack for supporting the vision sensor and the light source;

the upper computer is used for controlling the light source to supplement light through the light source controller under the condition that the cell corresponding to the mark is detected to reach the preset image acquisition position, and controlling the vision sensor to acquire the first exposure type image and the second exposure type image.

13. A tab misalignment detection apparatus, the apparatus comprising:

The second determining module is used for determining a target tab layer interval corresponding to the second exposure type image from the tab layer intervals according to the detection model and the second exposure type image of the tab to be detected; the first exposure type image and the second exposure type image are shot by a visual sensor along the thickness direction of the tab to be detected, the field of view range of the visual sensor covers the tab to be detected, and the target tab layer interval is an interval with the maximum offset of the edge in the thickness direction of the tab to be detected;

the correction module is used for correcting the initial tab edge distance according to the target tab layer interval to obtain the target tab edge distance of the tab to be detected; and the target tab edge distance is used for confirming whether the tab to be tested is dislocated or not.

14. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 8.