CN116429768A - Sealing nail welding quality detection method, system, equipment and storage medium - Google Patents

Abstract

The invention discloses a sealing nail welding quality inspection method, system, equipment and storage medium. The method includes: acquiring a two-dimensional image of a power battery, and performing image recognition on the two-dimensional image to determine whether the power battery exists Surface defects; when it is judged that there are surface defects in the power battery, order the tilted 3D camera to scan the surface of the power battery to obtain a three-dimensional image; locate the power battery multiple times according to the depth information of the three-dimensional image The weld area of the weld area is partitioned and detected to determine whether there is a depth defect in the weld area, and the result of the defect is displayed. The invention adopts the combination of 2D and 3D machine vision technologies to analyze the sealing nail welding condition on the power battery, improves the comprehensiveness of the sealing nail welding quality inspection, and reduces the occurrence of missed inspections.

Description

Method, system, equipment and storage medium for testing welding quality of sealing nails

technical field

The invention relates to the field of power battery welding quality detection, in particular to a sealing nail welding quality detection method, system, equipment and storage medium.

Background technique

At present, in the field of power lithium batteries, the existing quality inspection methods for sealing nail welds mainly include the following:

(1) Manually observe the real-time enlarged image of the weld on the screen of the microscope camera. When the product passes directly under the microscope, the camera captures the image of the weld and transmits it to the display screen. The worker observes the weld image on the display screen, thereby Judging whether there is a defect in the weld at this place, if there is a defect, put the product in the position of the NG product, if there is no defect, move the product, and continue to image the part that has not been imaged. This method is affected by factors such as workers' work experience, fatigue level, and workers' emotions, which can easily lead to missing and overkilling, the detection process is difficult to control, the cost is high, and it is difficult to realize the automatic monitoring of the production process;

(2) 2D machine vision automatic judgment, that is, the product moves directly under the camera, lights up the machine vision light source, and images the weld seam of the product, and then the computer displays, processes, calculates and judges whether there is a defect in the collected image. If there is a defect, the product will be discharged to the defect position, otherwise continue to the next station. This method is easily affected by the instability of the light source, and is limited by the characteristics of the height information that cannot be detected, such as poor flatness of the sealing nail, high weld seam and other defects. 2D machine vision cannot solve these problems, resulting in missed kill rate and overkill The rate is relatively high, and the quality cannot be guaranteed;

(3) 3D machine vision automatic judgment, the product moves below the laser line of the 3D camera, triggers the camera to collect the image of the weld seam of the product, and sends the image to the host computer, and the computer displays the collected point cloud data and converts it into a depth image. Image processing, calculation and analysis, and then judge whether there are defects in the weld. This method can detect defects with certain depth information, but it cannot judge defects that are not sensitive to depth information, such as cracks and welding black defects. In addition, when the weld has just been welded, the surface of the weld is relatively smooth, and the laser of the 3D sensor is irradiated on the surface of the weld, which is prone to diffuse reflection, thereby forming small pits on the surface of the weld, resulting in misjudgment; and The 3D sensor is susceptible to signal interference in the direction of motion, resulting in certain errors in weld detection, which cannot improve the detection accuracy.

Contents of the invention

In order to overcome the deficiencies of the prior art, one of the purposes of the present invention is to provide a sealing nail welding quality inspection method, which can improve the accuracy of sealing nail welding quality inspection.

The second object of the present invention is to provide a sealing nail welding quality inspection system.

The third object of the present invention is to provide an electronic device.

The fourth object of the present invention is to provide a computer-readable storage medium.

One of purpose of the present invention adopts following technical scheme to realize:

A method for testing the welding quality of sealing nails, comprising:

Acquiring a two-dimensional image of the power battery, and performing image recognition on the two-dimensional image to determine whether there is a surface defect in the power battery;

When it is judged that there is a surface defect in the power battery, order the tilted 3D camera to scan the surface of the power battery to obtain a three-dimensional image;

The weld area of the power battery is positioned multiple times according to the depth information of the three-dimensional image, and the weld area is subjected to partition detection to determine whether there is a depth defect in the weld area, and the result of the defect is displayed.

Further, the 3D camera is tilted downward at an angle of 5°-10°.

Further, before obtaining the three-dimensional image, it also includes:

Collecting point cloud data scanned by the 3D camera, performing position compensation on the point cloud data according to the tilt angle of the 3D camera, and then splicing the position compensated point cloud data to generate the three-dimensional image.

Further, the method for locating the weld region includes:

Based on the depth information of the three-dimensional image, the average height values of the sealing nails and the casing on the power battery are respectively calculated, and the flatness of the sealing nails and the casing is evaluated to generate an evaluation result; and the average height of the casing is used As a reference, the height of the sealing nail is extracted to determine the area of the sealing nail, which is converted into a binary image and then subject to separation to obtain the preliminary weld area.

Further, the method for locating the weld area also includes:

A section contour line is generated in the preliminary weld area, and the inner and outer edge points of the weld seam are determined according to the height information of the section contour line to complete the welding seam positioning.

Further, the method for judging whether there is a depth defect in the weld region includes:

The feature information of the section contour line is identified, and the feature information of the section contour line is analyzed to determine whether there is a depth defect in the weld seam region.

Further, the method for judging whether there is a depth defect in the weld region also includes:

Based on the pre-built weld defect model, the defect judgment is performed on the specified position of the weld to identify whether there is a depth defect in the weld region.

Two of the purpose of the present invention adopts following technical scheme to realize:

A sealing nail welding quality inspection system, which implements the above-mentioned sealing nail welding quality inspection method; the system includes:

The two-dimensional vision module is used to use the 2D camera to shoot the power battery to obtain a two-dimensional image;

The three-dimensional vision module is used to control the inclined 3D camera to scan the surface of the power battery to obtain a three-dimensional image;

The defect analysis module is used to perform image recognition on the two-dimensional image to determine whether there is a surface defect in the power battery; at the same time, according to the depth information of the three-dimensional image, the weld area of the power battery is positioned multiple times, and the The above-mentioned weld area is inspected in partitions to determine whether there is a depth defect in the weld area, and the result of the defect is displayed.

Three of the purpose of the present invention adopts following technical scheme to realize:

An electronic device, comprising a processor, a memory, and a computer program stored on the memory and operable on the processor, when the processor executes the computer program, the above-mentioned seal nail welding quality inspection is realized method.

Four of the purpose of the present invention adopts following technical scheme to realize:

A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, the above method for detecting the welding quality of sealing nails is realized.

Compared with the prior art, the beneficial effects of the present invention are:

The invention adopts the combination of 2D and 3D machine vision technology to analyze the welding condition of the sealing nail on the power battery, improves the comprehensiveness of the welding quality inspection of the sealing nail, and reduces the occurrence of missed detection; at the same time, the receiving end of the 3D camera is installed in a downwardly inclined manner , can reduce the raised interference signal generated by the 3D camera in the motion mode, and improve the detection accuracy.

Description of drawings

1 is a schematic diagram of the overall flow of the battery quality detection method of the present invention;

Fig. 2 is the original image of the sealing nail of the present invention;

Fig. 3 is a schematic diagram of the binarization result of the sealing nail image of the present invention;

Fig. 4 is a schematic diagram of the contour extraction results of the present invention;

Fig. 5 is a schematic diagram of welding seam positioning in the present invention;

Fig. 6 is a schematic diagram of a straight line of an arc ring in the present invention;

Fig. 7 is a schematic diagram of a single weld outline in the present invention;

Fig. 8 is a schematic diagram of the profile distribution of the weld outline in Fig. 7;

Fig. 9 is a schematic flow chart of the method for detecting the welding quality of sealing nails according to the present invention.

Detailed ways

Below, the present invention will be further described in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of not conflicting, the various embodiments described below or the technical features can be combined arbitrarily to form new embodiments. .

Embodiment one

This embodiment provides a method for detecting the welding quality of sealing nails. This method combines 2D machine vision and 3D machine vision to jointly judge the welding quality of sealing nails on power batteries, reduce the occurrence of overkill and underkill, and improve Quality inspection accuracy.

Referring to Fig. 1 and Fig. 9, the sealing nail welding quality detection method of the present embodiment includes the following steps:

Step S1: Obtain a two-dimensional image of the power battery, and perform image recognition on the two-dimensional image to determine whether there is a surface defect in the power battery;

Step S2: When it is determined that the power battery has surface defects, order the tilted 3D camera to scan the surface of the power battery to obtain a three-dimensional image;

Step S3: Locate the weld area of the power battery multiple times according to the depth information of the three-dimensional image, perform partition detection on the weld area to determine whether there is a depth defect in the weld area, and display the result of the defect .

In this embodiment, the target object, that is, the power battery that has been welded with sealing nails, is sent to the directly below the imaging position of the 2D camera. After receiving the information of the target object being in place, the light source is ordered to be turned on, and the 2D camera is triggered to take pictures of the target object. to obtain a two-dimensional image.

Use the deep learning method to classify the collected two-dimensional images to judge whether there are surface defects. The surface defects refer to defects that can be directly identified through 2D vision, such as cracks, small pinholes, and blackening. And send the result to the control system; if there is a defect, directly transfer the target to the NG; if no surface defect is detected, continue to move the target and transfer the target to the 3D camera, and use the 3D camera to monitor the power battery. The sealing nails are photographed to obtain a three-dimensional image, and the three-dimensional image with depth information is used for further quality inspection of the weld area.

The 3D camera starts to shoot the position from the sealing nail. The 3D camera scans the point cloud data of the sealing nail at a speed of 20mm/s. The 3D camera collects the point cloud data of the sealing nail according to the pulse signal of the encoder. Each part of the data collected is sent to the industrial computer. 3D After the camera is scanned, the point cloud data collected each time are stitched together. Then convert the point cloud data into a depth image and display it.

In this embodiment, the 3D camera is installed at a downward slanting angle, which can reduce the raised interference signal generated by the 3D camera in the motion mode, and improve the accuracy of the subsequent weld seam quality inspection. At the same time, before processing the point cloud data, it is necessary to compensate the installation position of the collected point cloud data. The point cloud data compensation is based on the installation method by using a standard gauge block, that is, first place the 3D camera according to the Install parallel to the horizontal plane, then measure the value at the specified position; then install the 3D camera according to the actual tilt angle, measure the value at the specified position again, and calculate the installation angle compensation value by measuring the same position value twice before and after, and then write it into the camera The controller realizes the installation angle compensation; before the software processing, the installation angle of the camera is corrected to avoid the error caused by the installation. Wherein, the inclination angle of the 3D camera is 5°-10°, and in this embodiment, the receiving end of the 3D camera is installed in a manner of inclining downward by 7°.

This embodiment uses a 3D camera to collect complete weld point cloud data to avoid loss of point cloud data due to blind spots and high reflections on the weld surface; at the same time, the camera is installed at an angle to reflect part of the stray light and avoid the contour sensor from receiving these points. Stray light improves the quality of collected point cloud data and provides a strong guarantee for subsequent point cloud data processing.

After obtaining the three-dimensional image scanned by the 3D camera, select several areas on the surface of the sealing nail, calculate the average height value in each area, and take the maximum value Pin_Max and the minimum value Pin_Min of the height of these areas; Select several areas above, calculate the average height values in these areas, and take the maximum value Top_Max and minimum value Top_Min of the average height of each area. Afterwards, Abs(Pin_Max-Top_Min) and Abs(Pin_Min-Top_Max) are calculated respectively, and their values are used to evaluate the flatness between the sealing nail and the surface of the battery case, and the evaluation results are displayed.

As shown in Figure 2 and Figure 3, based on the average height of the upper surface of the shell, the height of the sealing nail is extracted, the extracted image is converted into an 8-bit grayscale image, and the converted grayscale image is thresholded , according to the gray value distribution of the gray image, the Otsu method is selected for threshold extraction, the purpose is to separate the main body of the object from the background. For the binarized image, the connection between the subjects is separated. This embodiment adopts the open operation morphology processing to realize the separation of some weakly connected objects, and then extracts the subject contour to obtain the subject contour, as shown in Figure 4 .

The center of the main body profile is used as the center of the sealing nail, and the maximum ring width of the main body profile is used as the radius of the weld seam, as shown in Figure 5, an inner circular line is generated according to the center of the sealing nail and the radius of the weld seam As well as the outer circular line, the area between the inner circular line and the outer circular line is marked as a preliminary weld area, which achieves the purpose of thick welding seam positioning.

As shown in Figure 6, several arc ring straight lines are generated in the width direction of the preliminary weld area, and the gap between the straight lines takes the detection accuracy as a reference value to determine the interval angle; therefore, before generating the arc ring straight lines , the detection accuracy needs to be customized by the user, and the detection accuracy parameter of the user-defined setting is obtained to determine the straight line gap, and several arc ring straight lines are evenly generated in the preliminary weld area according to the straight line gap.

The depth image of the weld area is divided into four areas, and the weld area analysis is carried out separately. In each area, based on the height information of the weld, the depth information of the welding image is used to assign the depth data of the position of each arc ring line to form a weld outline; as shown in Figure 7 and Figure 8 As shown, according to the profile distribution of the weld outline, the inner and outer sides of the weld can be found, thereby achieving precise positioning of the weld.

In the weld area, based on the mathematical model of the defect, search for defects along the direction of the weld, record the key feature points of the weld found in the search process, and judge whether there is a depth defect in the weld based on the feature information of the key feature points, That is, weld defects that cannot be identified from the two-dimensional image, such as pits, protrusions, surface flatness of sealing nails, broken welds, missing welds, etc.; and display the judgment results. Wherein, the key feature points include the outermost point, the innermost point, the highest point and the lowest point of the weld contour line; further, the feature information of the key feature points can also be obtained, including the maximum height value of the weld, weld Information such as the minimum height of the seam and the width of the weld.

Among them, the weld defect model can collect the point cloud data of sample defects in advance according to the customer's inspection requirements, and carry out mathematical modeling according to the specific shape of the defect, so as to obtain the mathematical model of each sample defect, which provides the necessary information for the search of weld defects. Conditions, the closer the mathematical model is to the actual defect shape, the more accurate the subsequent defect search will be.

In the area where defects may occur, according to the key characteristic values of the weld, process, calculate and judge according to the weld defect model, calculate the length of the weld defect, judge whether it is a defect, and then make a second judgment according to the characteristics of the weld , to determine whether there is a defect.

Each divided area of the weld area is processed according to the above method, and the defect situation in each area is counted by area, and then it is determined whether there is a defect in the entire weld, and the judgment result is displayed on the display screen. If there is a defect, Make an alarm and send the product to the NG box; if it is a non-defective product, send the product to the next station.

In this embodiment, 2D machine vision and 3D machine vision are combined to jointly judge weld defects; since the weld has just been welded, the surface of the weld is relatively bright, and when the laser is irradiated on the surface of the weld, the laser will produce diffuse reflection. The collected point cloud data will cause loss, and pits will be formed in the image, resulting in misjudgment; and 2D machine vision can just make up for this defect. 2D imaging does not have such small pit defects, thereby improving the detection accuracy. The accuracy of defect judgment is high, and the misjudgment rate is low. In addition, this embodiment directly searches for defects through the weld defect model, and the entire process of finding defects uses 3D point cloud information, which avoids misjudgment caused by changing 3D to 2D.

In this embodiment, 2D imaging is used for small pinholes, cracks, and blackened defects, which improves the quality and resolution of image imaging; the method of deep learning is adopted, and the more samples are detected, the higher the accuracy of the judgment result.

Embodiment two

This embodiment provides a sealing nail welding quality inspection system, which implements the sealing nail welding quality inspection method as described in Embodiment 1; the system includes:

The two-dimensional vision module is used to use the 2D camera to shoot the power battery to obtain a two-dimensional image;

The three-dimensional vision module is used to control the inclined 3D camera to scan the surface of the power battery to obtain a three-dimensional image;

The defect analysis module is used to perform image recognition on the two-dimensional image to determine whether there is a surface defect in the power battery; at the same time, according to the depth information of the three-dimensional image, the weld area of the power battery is positioned multiple times, and the The above-mentioned weld area is inspected in partitions to determine whether there is a depth defect in the weld area, and the result of the defect is displayed.

In some embodiments, there is also provided an electronic device, which includes a processor, a memory, and a computer program stored on the memory and operable on the processor, when the processor executes the computer program, the The sealing nail welding quality detection method in the first embodiment.

In some embodiments, there is also provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, the method for detecting the sealing nail welding quality in the first embodiment is implemented.

The system, equipment, and storage medium in this embodiment are based on multiple aspects under the same inventive concept as the methods in the foregoing embodiments. The implementation process of the method has been described in detail above, so those skilled in the art can The description clearly understands the structure and implementation process of the system, device, and storage medium in this embodiment, and for the sake of brevity, details are not repeated here.

The above-mentioned embodiment is only a preferred embodiment of the present invention, and cannot be used to limit the protection scope of the present invention. Any insubstantial changes and substitutions made by those skilled in the art on the basis of the present invention belong to the scope of the present invention. Scope of protection claimed.

Claims (10)

1. A sealing nail welding quality detection method, characterized in that, comprising: Acquiring a two-dimensional image of the power battery, and performing image recognition on the two-dimensional image to determine whether there is a surface defect in the power battery; When it is judged that there is a surface defect in the power battery, order the tilted 3D camera to scan the surface of the power battery to obtain a three-dimensional image; The weld area of the power battery is positioned multiple times according to the depth information of the three-dimensional image, and the weld area is subjected to partition detection to determine whether there is a depth defect in the weld area, and the result of the defect is displayed. 2. The sealing nail welding quality inspection method according to claim 1, wherein the 3D camera is inclined downward at an angle of 5° to 10°. 3. The sealing nail welding quality detection method according to claim 1, characterized in that, before obtaining the three-dimensional image, further comprising: Collecting point cloud data scanned by the 3D camera, performing position compensation on the point cloud data according to the tilt angle of the 3D camera, and then splicing the position compensated point cloud data to generate the three-dimensional image. 4. The sealing nail welding quality inspection method according to claim 1, wherein the method for locating the weld seam region comprises: Based on the depth information of the three-dimensional image, the average height values of the sealing nails and the casing on the power battery are respectively calculated, and the flatness of the sealing nails and the casing is evaluated to generate an evaluation result; and the average height of the casing is used As a reference, height extraction is performed on the height of the sealing nail to determine the area of the sealing nail, which is converted into a binary image and then subjected to subject separation to obtain a preliminary weld seam area. 5. The sealing nail welding quality detection method according to claim 4, characterized in that, the method for locating the weld seam region further comprises: A section contour line is generated in the preliminary weld area, and the inner and outer edge points of the weld seam are determined according to the height information of the section contour line to complete the welding seam positioning. 6. The sealing nail welding quality inspection method according to claim 5, wherein the method for judging whether there is a depth defect in the weld region comprises: The feature information of the section contour line is identified, and the feature information of the section contour line is analyzed to determine whether there is a depth defect in the weld seam region. 7. The sealing nail welding quality inspection method according to claim 5, wherein the method for judging whether there is a depth defect in the weld region further comprises: Based on the pre-built weld defect model, the defect judgment is performed on the specified position of the weld to identify whether there is a depth defect in the weld region. 8. A sealing nail welding quality detection system, characterized in that, the method for detecting the sealing nail welding quality as described in any one of claims 1 to 7 is performed; the system comprises: The two-dimensional vision module is used to use the 2D camera to shoot the power battery to obtain a two-dimensional image; The three-dimensional vision module is used to control the inclined 3D camera to scan the surface of the power battery to obtain a three-dimensional image; The defect analysis module is used to perform image recognition on the two-dimensional image to determine whether there is a surface defect in the power battery; at the same time, according to the depth information of the three-dimensional image, the weld area of the power battery is positioned multiple times, and the The above-mentioned weld area is inspected in partitions to determine whether there is a depth defect in the weld area, and the result of the defect is displayed. 9. An electronic device, characterized in that it comprises a processor, a memory, and a computer program stored on the memory and operable on the processor, and the processor implements the claims when executing the computer program The sealing nail welding quality inspection method described in any one of 1 to 7. 10. A computer-readable storage medium, characterized in that a computer program is stored thereon, and when the computer program is executed, the sealing nail welding quality detection method according to any one of claims 1 to 7 is implemented.

Images