Adaptive double-threshold segmentation method for battery busbar ultrasonic phased array C-scan image

Abstract

This invention addresses the limitations of existing technologies by proposing an adaptive dual-threshold segmentation method for ultrasonic phased array C-scan images of battery busbars. It primarily targets the technical problem of blurred edges and difficulty in distinguishing foreground and background in ultrasonic phased array C-scan images of multi-station battery busbars. By enhancing blurred edges, redundant information in the ultrasonic phased array C-scan images is effectively eliminated, enhancing the detectability of useful information and simplifying the data volume to the greatest extent. This improves the reliability of image segmentation, feature detection, and feature recognition, avoiding the shortcomings of existing segmentation methods. Using this method to segment ultrasonic phased array C-scan images of power battery busbars yields significant results, enabling high-speed and automated segmentation. This facilitates subsequent image detection algorithms, improving the reliability and accuracy of detection, and is particularly suitable for segmenting ultrasonic phased array C-scan images with severe noise pollution.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic phased array C-scan image technology, and more specifically, to an adaptive dual-threshold segmentation method for ultrasonic phased array C-scan images of a battery bus. Background Technology

[0002] Phased array ultrasonic testing technology has been developed and applied as a novel technology. It can be combined with image processing technology for non-destructive testing of workpieces. In the new energy vehicle industry, ultrasonic phased array technology is used to perform C-scan imaging of power battery busbars. After processing with image processing algorithms, the welding quality of welded workpieces can be obtained. Integrating image inspection software into automated equipment allows for online inspection of large batches of workpieces, thus replacing manual destructive sampling inspection.

[0003] A common problem with this type of automated inspection equipment is that the true welding area in the ultrasonic phased array C-scan image may not be accurately extracted, which will affect the accuracy of the image processing results. Ultrasonic phased array C-scan images have low resolution, and due to the welding process, the image contains regularly arranged black and white dots near the welding area. These dots, unlike general noise, cannot be removed using traditional image denoising methods and have a significant impact on the extraction of the true welding area. Therefore, using generalized segmentation algorithms to extract the true welding area is not ideal.

[0004] Image segmentation refers to extracting targets from an image using certain features. Information analysis is performed on the image to be segmented, and important information is extracted as features. Image processing techniques such as thresholding, morphological processing, and image filtering can improve segmentation accuracy. For example, the Korean patent application published on April 10, 2020, entitled "IMAGE PROCESSINGMETHOD AND APPARATUS FOR SPOT WELDING QUALITY EVALUATION," provides an image processing method for welding quality assessment. This method includes the following steps: converting the echo signal reflected from the spot weld area into a grayscale signal at a specific level using C-scanning to generate a two-dimensional grayscale image; upsampling the grayscale image at a preset upsampling ratio; blurring the upsampled image according to a preset standard deviation; interpolating the blurred image using morphological techniques; and color mapping the interpolated image. However, for multi-station battery busbar workpieces, the edges of the ultrasonic phased array C-scan images are blurred and easily affected by the workpiece's levelness. There is a lot of redundant information in non-interest areas, and the results of multiple scans can differ, resulting in less than ideal segmentation performance. Therefore, existing technologies still have certain limitations. Summary of the Invention

[0005] To address the limitations of existing technologies, this invention proposes an adaptive dual-threshold segmentation method for C-scan images of battery busbar ultrasonic phased arrays. The technical solution adopted in this invention is as follows:

[0006] An adaptive dual-threshold segmentation method for C-scan images of battery bus ultrasonic phased arrays includes the following steps:

[0007] S1, acquire an ultrasonic phased array C-scan image containing several workpieces, and convert the ultrasonic phased array C-scan image into a grayscale image;

[0008] S2, obtain the average gray value of the grayscale image as the lower boundary threshold; set the upper boundary threshold with a fixed gray value;

[0009] S3, perform image segmentation on the grayscale image according to the lower boundary threshold and the upper boundary threshold to obtain a binary image;

[0010] S4, extract the contours of the grayscale image in the x and y directions respectively, and fuse the extracted contours into a two-dimensional edge image;

[0011] S5, perform morphological operations on the two-dimensional edge image to obtain an edge-enhanced contour image;

[0012] S6. After performing single-channel difference operation on the binary image and the edge-enhanced contour image, morphological operation is performed to obtain the foreground image of each workpiece in the ultrasonic phased array C-scan image.

[0013] S7, the foreground image is filtered by position scale and area scale respectively; the filtering results are sorted based on the x-coordinate size to complete the accurate segmentation of each workpiece in the ultrasonic phased array C-scan image.

[0014] Compared to existing technologies, the present invention primarily addresses the technical problem of blurred edges and difficulty in distinguishing foreground and background in ultrasonic phased array C-scan images of multi-station battery busbars. By enhancing the blurred edges, redundant information in the ultrasonic phased array C-scan images is effectively eliminated, enhancing the detectability of useful information and simplifying the data volume to the greatest extent. This improves the reliability of image segmentation, feature detection, and feature recognition, avoiding the shortcomings of existing segmentation methods. Using this method to segment ultrasonic phased array C-scan images of power battery busbars yields significant results, enabling high-speed and automated segmentation. This facilitates subsequent image algorithm detection, improving the reliability and accuracy of detection, and is particularly suitable for segmenting ultrasonic phased array C-scan images with severe noise pollution.

[0015] As a preferred embodiment, in step S3, the grayscale image I is processed using the following formula. src Perform image segmentation to obtain a binary image I.bin :

[0016]

[0017] Where t1 represents the lower boundary threshold; t2 represents the upper boundary threshold.

[0018] As a preferred embodiment, step S4 includes the following process:

[0019] By analyzing the grayscale image I src Perform convolution operations to extract the grayscale images I respectively. src The contour m in the x and y directions x m y :

[0020]

[0021]

[0022] in, Indicates convolution operation; k x Denotes the convolution kernel in the x-direction, k y Represents the convolution kernel in the y-direction;

[0023] The extracted contour m x m y Merged into a two-dimensional edge image m xy Perform the following formula:

[0024] m xy =m y ∨m x ;

[0025] Here, ∨ represents the OR operation.

[0026] Furthermore, the convolution kernel k x k y The mathematical expression is as follows:

[0027]

[0028] As a preferred embodiment, the morphological operation performed in step S5 is a morphological dilation operation; and the morphological operation performed in step S6 is a morphological closing operation.

[0029] As a preferred embodiment, in step S7, the foreground image is filtered by position and scale in the following manner:

[0030] Find the smallest regular rectangle containing all points of each contour in the foreground image;

[0031] Calculate the center point of each smallest rectangle: if the center point is located at the top 1 / 3 of the image, mark the contour corresponding to the center point as the contour with the correct position and scale; otherwise, remove the contour.

[0032] Furthermore, in step S7, the foreground image is filtered by area scale in the following manner:

[0033] Calculate the average area of ​​each workpiece based on the number of workpieces n.

[0034] Area scale filtering is performed on contours with correct scale at each location: when the area s of the contour... a satisfy If the outline is marked as having the correct area scale, then the outline is removed.

[0035] This invention also includes the following:

[0036] An adaptive dual-threshold segmentation system for C-scan images of a battery busbar ultrasonic phased array includes an image conversion module, a threshold setting module, a binary image acquisition module, a contour fusion module, an edge enhancement module, a foreground image acquisition module, and a filtering and sorting module. The image conversion module is connected to both the threshold setting module and the contour fusion module. The threshold setting module is connected to the binary image acquisition module. The contour fusion module is connected to the edge enhancement module. The foreground image acquisition module is connected to the binary image acquisition module, the edge enhancement module, and the filtering and sorting module.

[0037] The image conversion module is used to acquire an ultrasonic phased array C-scan image containing several workpieces and convert the ultrasonic phased array C-scan image into a grayscale image.

[0038] The threshold setting module is used to obtain the average gray value of the grayscale image as the lower boundary threshold; and to set the upper boundary threshold with a fixed gray value.

[0039] The binary image acquisition module is used to perform image segmentation on the grayscale image based on the lower boundary threshold and the upper boundary threshold to obtain a binary image.

[0040] The contour fusion module is used to extract the contours of the grayscale image in the x and y directions respectively, and fuse the extracted contours into a two-dimensional edge image.

[0041] The edge enhancement module is used to perform morphological operations on the two-dimensional edge image to obtain an edge-enhanced contour image;

[0042] The foreground image acquisition module is used to perform morphological operations on the binary image and the edge-enhanced contour image after performing single-channel difference operations to obtain the foreground image of each workpiece in the ultrasonic phased array C-scan image.

[0043] The filtering and sorting module is used to filter the foreground image at both the position and area scales; and sorts the filtering results based on the x-coordinate to complete the accurate segmentation of each workpiece in the ultrasonic phased array C-scan image.

[0044] A storage medium storing a computer program that, when executed by a processor, implements the steps of the aforementioned adaptive dual-threshold segmentation method for ultrasonic phased array C-scan images of battery busbars.

[0045] A computer device includes a storage medium, a processor, and a computer program stored in the storage medium and executable by the processor, wherein the computer program, when executed by the processor, implements the steps of the aforementioned adaptive dual-threshold segmentation method for ultrasonic phased array C-scan images of battery busbars. Attached Figure Description

[0046] Figure 1 This is a schematic diagram illustrating the steps of the adaptive dual-threshold segmentation method for C-scan images of battery bus ultrasonic phased array provided in an embodiment of the present invention;

[0047] Figure 2 An example of a C-scan image from an ultrasonic phased array;

[0048] Figure 3 For the embodiments of the present invention from Figure 2 The region of interest extracted from the data;

[0049] Figure 4 For the purposes of this invention, Figure 3 The contour m extracted in the x-direction x ;

[0050] Figure 5 For the purposes of this embodiment of the invention Figure 3 The contour m extracted in the y direction y ;

[0051] Figure 6 For the purposes of this embodiment of the invention Figure 4 , Figure 5 The resulting two-dimensional edge image m xy ;

[0052] Figure 7 The edge-enhanced contour image m′ obtained in step S5 of this embodiment of the invention. xy ;

[0053] Figure 8 The foreground image I obtained in step S6 of this embodiment of the invention. f ;

[0054] Figure 9 For the embodiments of the present invention Figure 2 The final image segmentation result obtained;

[0055] Figure 10 This is a schematic diagram of an adaptive dual-threshold segmentation system for C-scan images of an ultrasonic phased array of battery bus provided in an embodiment of the present invention. Detailed Implementation

[0056] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent.

[0057] It should be understood that the described embodiments are merely some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of the embodiments of this application.

[0058] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the embodiments of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0059] In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims. In the description of this application, it should be understood that the terms "first," "second," "third," etc., are used only to distinguish similar objects and are not necessarily used to describe a specific order or sequence, nor should they be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0060] Furthermore, in the description of this application, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. The invention will be further described below with reference to the accompanying drawings and embodiments.

[0061] To address the limitations of existing technologies, this embodiment provides a technical solution. The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0062] Example 1

[0063] An adaptive dual-threshold segmentation method for C-scan images of battery bus ultrasonic phased arrays; please refer to [link to relevant documentation]. Figure 1 This includes the following steps:

[0064] S1, acquire an ultrasonic phased array C-scan image containing several workpieces, and convert the ultrasonic phased array C-scan image into a grayscale image;

[0065] S2, obtain the average gray value of the grayscale image as the lower boundary threshold; set the upper boundary threshold with a fixed gray value;

[0066] S3, perform image segmentation on the grayscale image according to the lower boundary threshold and the upper boundary threshold to obtain a binary image;

[0067] S4, extract the contours of the grayscale image in the x and y directions respectively, and fuse the extracted contours into a two-dimensional edge image;

[0068] S5, perform morphological operations on the two-dimensional edge image to obtain an edge-enhanced contour image;

[0069] S6. After performing single-channel difference operation on the binary image and the edge-enhanced contour image, morphological operation is performed to obtain the foreground image of each workpiece in the ultrasonic phased array C-scan image.

[0070] S7, the foreground image is filtered by position scale and area scale respectively; the filtering results are sorted based on the x-coordinate size to complete the accurate segmentation of each workpiece in the ultrasonic phased array C-scan image.

[0071] Compared to existing technologies, the present invention primarily addresses the technical problem of blurred edges and difficulty in distinguishing foreground and background in ultrasonic phased array C-scan images of multi-station battery busbars. By enhancing the blurred edges, redundant information in the ultrasonic phased array C-scan images is effectively eliminated, enhancing the detectability of useful information and simplifying the data volume to the greatest extent. This improves the reliability of image segmentation, feature detection, and feature recognition, avoiding the shortcomings of existing segmentation methods. Using this method to segment ultrasonic phased array C-scan images of power battery busbars yields significant results, enabling high-speed and automated segmentation. This facilitates subsequent image algorithm detection, improving the reliability and accuracy of detection, and is particularly suitable for segmenting ultrasonic phased array C-scan images with severe noise pollution.

[0072] Specifically, in step S1, an ultrasonic phased array probe can be used to scan the workpiece, and an ultrasonic phased array C-scan image can be obtained through ultrasonic board imaging. The image is then read into the integrated environment of Visual Studio, Qt, and OpenCV libraries.

[0073] The ultrasound phased array C-scan image I consists of three RGB channels, representing the red, green, and blue components respectively, with a ratio of 1:1:1. The ultrasound phased array C-scan image is separated by channel to obtain I. r I g I b Take one of the channels to form a grayscale image I src As an image for subsequent processing, it can effectively reduce computational complexity, such as Figure 2 As shown.

[0074] In a preferred embodiment, in step S3, the grayscale image I is processed using the following formula. src Perform image segmentation to obtain a binary image I. bin ,like Figure 3 As shown:

[0075]

[0076] Where t1 represents the lower boundary threshold; t2 represents the upper boundary threshold.

[0077] In a preferred embodiment, step S4 includes the following process:

[0078] By analyzing the grayscale image I src Perform convolution operations to extract the grayscale images I respectively. src The contour m in the x and y directions x m y :

[0079]

[0080]

[0081] in, Indicates convolution operation; k x Denotes the convolution kernel in the x-direction, k y The convolution kernel in the y-direction represents the contour m obtained. x m y like Figure 4 , Figure 5 As shown;

[0082] The extracted contour m x m y Merged into a two-dimensional edge image m xy Perform the following formula:

[0083] m xy =m y ∨m x ;

[0084] Where ∨ represents the OR operation, and the resulting two-dimensional edge image m xy like Figure 6 As shown.

[0085] Furthermore, the convolution kernel k x k y The mathematical expression is as follows:

[0086]

[0087] In a preferred embodiment, the morphological operation performed in step S5 is a morphological dilation operation; and the morphological operation performed in step S6 is a morphological closing operation.

[0088] Specifically, the edge-enhanced contour image m′ obtained in step S5 xy like Figure 7 As shown. The formula for performing single-channel difference operations on the binary image and the edge-enhanced contour image in step S6 is as follows:

[0089]

[0090] in, This represents the "image difference" operation; foreground image I f like Figure 8 As shown.

[0091] In a preferred embodiment, in step S7, the foreground image is filtered by position and scale in the following manner:

[0092] Find the smallest regular rectangle containing all points of each contour in the foreground image;

[0093] Calculate the center point of each smallest rectangle: if the center point is located at the top 1 / 3 of the image, mark the contour corresponding to the center point as the contour with the correct position and scale; otherwise, remove the contour.

[0094] Furthermore, in step S7, the foreground image is filtered by area scale in the following manner:

[0095] Calculate the average area of ​​each workpiece based on the number of workpieces n.

[0096] Area scale filtering is performed on contours with correct scale at each location: when the area s of the contour... a satisfy If the outline is marked as having the correct area scale, then the outline is removed.

[0097] Finally, the workpiece images that are correct in both scales are sorted based on their x-coordinate size, as shown below. Figure 9 As shown.

[0098] Example 2

[0099] An adaptive dual-threshold segmentation system for C-scan images of battery bus ultrasonic phased arrays, please refer to [link to relevant documentation]. Figure 10 The system includes an image conversion module 1, a threshold setting module 2, a binary image acquisition module 3, a contour fusion module 4, an edge enhancement module 5, a foreground image acquisition module 6, and a filtering and sorting module 7. The image conversion module 1 is connected to the threshold setting module 2 and the contour fusion module 4. The threshold setting module 2 is connected to the binary image acquisition module 3. The contour fusion module 4 is connected to the edge enhancement module 5. The foreground image acquisition module 6 is connected to the binary image acquisition module 3, the edge enhancement module 5, and the filtering and sorting module 7.

[0100] The image conversion module 1 is used to acquire an ultrasonic phased array C-scan image containing several workpieces and convert the ultrasonic phased array C-scan image into a grayscale image.

[0101] The threshold setting module 2 is used to obtain the average gray value of the grayscale image as the lower boundary threshold; and to set the upper boundary threshold with a fixed gray value.

[0102] The binary image acquisition module 3 is used to perform image segmentation on the grayscale image according to the lower boundary threshold and the upper boundary threshold to obtain a binary image;

[0103] The contour fusion module 4 is used to extract the contours of the grayscale image in the x and y directions respectively, and fuse the extracted contours into a two-dimensional edge image.

[0104] The edge enhancement module 5 is used to perform morphological operations on the two-dimensional edge image to obtain an edge-enhanced contour image;

[0105] The foreground image acquisition module 6 is used to perform morphological operations on the binary image and the edge-enhanced contour image after performing single-channel difference operations to obtain the foreground image of each workpiece in the ultrasonic phased array C-scan image.

[0106] The filtering and sorting module 7 is used to filter the foreground image by position and area scale respectively; and sort the filtering results based on the x-coordinate to complete the accurate segmentation of each workpiece in the ultrasonic phased array C-scan image.

[0107] Example 3

[0108] A storage medium storing a computer program that, when executed by a processor, implements the steps of the adaptive dual-threshold segmentation method for battery bus ultrasonic phased array C-scan images as described in Example 1.

[0109] Example 4

[0110] A computer device includes a storage medium, a processor, and a computer program stored in the storage medium and executable by the processor, wherein the computer program, when executed by the processor, implements the steps of the adaptive dual-threshold segmentation method for battery bus ultrasonic phased array C-scan images as described in Example 1.

[0111] Specifically, as a preferred embodiment, C++ can be used as the main programming language to write the relevant image processing functions.

[0112] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. An adaptive dual-threshold segmentation method for C-scan images of battery bus ultrasonic phased arrays, characterized in that, Includes the following steps: S1, acquire an ultrasonic phased array C-scan image containing several workpieces, and convert the ultrasonic phased array C-scan image into a grayscale image; S2, obtain the average gray value of the grayscale image as the lower boundary threshold; set the upper boundary threshold with a fixed gray value; S3, perform image segmentation on the grayscale image according to the lower boundary threshold and the upper boundary threshold to obtain a binary image; S4, extract the contours of the grayscale image in the x and y directions respectively, and fuse the extracted contours into a two-dimensional edge image, including the following process: By analyzing the grayscale image Perform convolution operations to extract the grayscale images respectively. The outline in the x and y directions , : ； ； in, This represents the convolution operation; This represents the convolution kernel in the x-direction. Represents the convolution kernel in the y-direction; Extracted contours , Fused into a two-dimensional edge image Perform the following formula: ； in, OR operation; The convolution kernel , The mathematical expression is as follows: S5, perform morphological operations on the two-dimensional edge image to obtain an edge-enhanced contour image; S6. After performing single-channel difference operation on the binary image and the edge-enhanced contour image, morphological operation is performed to obtain the foreground image of each workpiece in the ultrasonic phased array C-scan image. S7, the foreground image is filtered by both position and area scales; the filtering results are sorted based on the x-coordinate to complete the accurate segmentation of each workpiece in the ultrasonic phased array C-scan image. Specifically, the position scale of the foreground image is filtered in the following way: Find the smallest regular rectangle containing all points of each contour in the foreground image; Calculate the center point of each smallest rectangle: if the center point is located at the top 1 / 3 of the image, mark the contour corresponding to the center point as the contour with the correct position and scale; otherwise, remove the contour. The foreground image is filtered by area scale using the following method: Based on the number of workpieces Calculate the average area of ​​each workpiece ; Area scale filtering is performed on contours with correct dimensions at each location: when the area of ​​the contour... satisfy If the area scale is correct, then mark the contour as the correct contour; otherwise, remove the contour.

2. The adaptive dual-threshold segmentation method for C-scan images of battery bus ultrasonic phased arrays according to claim 1, characterized in that, In step S3, the grayscale image is processed using the following formula. Perform image segmentation to obtain a binary image. : ； in, This represents the lower boundary threshold; This represents the upper boundary threshold.

3. The adaptive dual-threshold segmentation method for C-scan images of battery bus ultrasonic phased arrays according to claim 1, characterized in that, The morphological operation performed in step S5 is a morphological dilation operation; the morphological operation performed in step S6 is a morphological closing operation.

4. An adaptive dual-threshold segmentation system for C-scan images of battery busbar ultrasonic phased array, characterized in that, The method for implementing the method as described in any one of claims 1 to 3 includes an image conversion module (1), a threshold setting module (2), a binary image acquisition module (3), a contour fusion module (4), an edge enhancement module (5), a foreground image acquisition module (6), and a filtering and sorting module (7); the image conversion module (1) is connected to the threshold setting module (2) and the contour fusion module (4); the threshold setting module (2) is connected to the binary image acquisition module (3); the contour fusion module (4) is connected to the edge enhancement module (5); the foreground image acquisition module (6) is connected to the binary image acquisition module (3), the edge enhancement module (5), and the filtering and sorting module (7); wherein: The image conversion module (1) is used to acquire an ultrasonic phased array C-scan image containing several workpieces and convert the ultrasonic phased array C-scan image into a grayscale image. The threshold setting module (2) is used to obtain the average gray value of the grayscale image as the lower boundary threshold; and to set the upper boundary threshold with a fixed gray value. The binary image acquisition module (3) is used to perform image segmentation on the grayscale image according to the lower boundary threshold and the upper boundary threshold to obtain a binary image; The contour fusion module (4) is used to extract the contours of the grayscale image in the x and y directions respectively, and fuse the extracted contours into a two-dimensional edge image; The edge enhancement module (5) is used to perform morphological operations on the two-dimensional edge image to obtain an edge-enhanced contour image; The foreground image acquisition module (6) is used to perform morphological operations on the binary image and the edge-enhanced contour image after performing single-channel difference operation to obtain the foreground image of each workpiece in the ultrasonic phased array C-scan image. The filtering and sorting module (7) is used to filter the foreground image by position scale and area scale respectively; and sort the filtering results based on the x-coordinate size to complete the accurate segmentation of each workpiece in the ultrasonic phased array C-scan image.

5. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the adaptive dual-threshold segmentation method for C-scan images of battery bus ultrasonic phased arrays as described in any one of claims 1 to 3.

6. A computer device, characterized in that: The method includes a storage medium, a processor, and a computer program stored in the storage medium and executable by the processor, wherein the computer program, when executed by the processor, implements the steps of the adaptive dual-threshold segmentation method for battery bus ultrasonic phased array C-scan images as described in any one of claims 1 to 3.

Images