Brake disc welding defect detection method and system

Abstract

The present invention relates to the field of data processing technology, and discloses a brake disc welding defect detection method and system, the method comprising: collecting an original image of a brake disc welding area; performing denoising and edge enhancement processing on the original image; extracting features of cracks and unfused areas based on multi-scale wavelet transform, and preliminarily locating defect areas; introducing crack direction weights to optimize high-frequency energy response; generating a brake disc welding defect detection report; compared with the technical problem in the prior art that efficient and accurate detection of welding cracks and unfused area defects cannot be achieved under complex welding conditions, such as high reflection or complex welding texture conditions, the present invention achieves high-precision detection of welding cracks and unfused defects through multi-scale wavelet transform combined with directional weight optimization and grayscale smoothing error correction, thereby avoiding the problems of low detection accuracy and high misjudgment rate in traditional methods, and greatly improving the efficiency and reliability of brake disc welding quality detection.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to a brake disc welding defect detection method and system. Background Art

[0002] Currently, the detection of brake disc weld defects primarily relies on manual inspection or traditional automated inspection systems. However, existing technologies present numerous technical challenges when dealing with complex weld defects. First, manual inspection methods are not only inefficient but also easily affected by the operator's experience and fatigue, resulting in inconsistent inspection results and making it difficult to ensure consistent and reliable weld quality. Second, traditional automated inspection systems often fail to accurately identify cracks and lack of fusion defects when faced with complex weld surfaces, reflective materials, and varying textures. Existing image processing methods often rely on a single algorithm and lack adaptability to diverse welding environments. This presents significant technical bottlenecks, particularly in areas such as highly reflective materials, complex weld paths, and multi-scale weld defect detection, hindering high-precision and efficient defect detection. Consequently, existing technologies cannot fully meet the practical needs of brake disc weld quality inspection, particularly in highly reflective environments and complex weld textures, where the inability to accurately locate cracks and lack of fusion remains a challenge. Therefore, a method that can achieve high-precision defect identification under complex welding conditions is urgently needed to overcome the limitations of traditional technologies and improve the accuracy and efficiency of weld quality inspection. Summary of the Invention

[0003] In response to the above-mentioned technical deficiencies, the purpose of the present invention is to propose a brake disc welding defect detection method, which aims to solve the technical problem in the existing technology that it is impossible to efficiently and accurately detect welding cracks and unfused area defects under complex welding conditions, such as high reflectivity or complex welding texture conditions.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a method for detecting welding defects of brake discs.

[0005] The brake disc welding defect detection method comprises:

[0006] Step S10: Using a high-definition industrial camera to capture an image of the brake disc welding area, an original welding area image I(x, y) is obtained, where x and y are image coordinates, representing the horizontal and vertical coordinates of the pixels in the welding area, respectively. The industrial camera has a resolution of 3 million pixels or more and a frame rate of 30 frames per second.

[0007] Step S20: Based on the original welding area image I(x,y) obtained in step S10, denoising and edge enhancement are performed on it, including:

[0008] Use Gaussian filter to remove noise from the original welding area image I(x,y) to obtain the denoised welding area image I g (x,y);

[0009] Denoised welding area image I g (x, y) is edge enhanced to obtain the welding area enhanced image I e (x,y);

[0010] Step S30: Based on the welding area enhanced image I obtained in step S20 e (x,y), and extract the characteristics of cracks and unfused areas through multi-scale wavelet transform, including:

[0011] The weld area enhanced image is decomposed into low-frequency components L using three-layer wavelet decomposition. i and high frequency components H i , where L i is the low-frequency component of the welding area, which is used to reflect the background smoothness characteristics of the unfused area; H i is the high-frequency component of the welding area, which is used to reflect the linear characteristics of the crack edge; i is the scale of decomposition, which takes a value of 1, 2 or 3;

[0012] Calculate the high frequency energy response E d (x,y), preset high-frequency energy response threshold E t , when the high frequency energy response E d (x,y) is greater than the high-frequency energy response threshold E t When , the pixel at the image coordinate (x, y) is determined to be a crack and unfused defect pixel, and the crack and unfused defect pixels are aggregated to obtain the crack and unfused area;

[0013] Step S40: Introducing the crack direction weight P j , the calculation formula is:

[0014] P j =exp(-k·θ j )

[0015] Among them, θ j is the direction angle of the crack, ranging from 0° to 180°; k is the directional attenuation coefficient;

[0016] According to the crack direction weight P j Optimize high-frequency energy response:

[0017] E d ′(x,y)=P j ·E d (x,y)

[0018] Among them, E′ d(x, y) is the high-frequency energy response after directional weight optimization;

[0019] Set the high-frequency energy threshold T of the crack c , if E′ d (x,y)>T c , then the pixel at the image coordinate (x, y) is determined to be a crack area point pixel, and the image coordinates of the crack area point pixel are output;

[0020] Introducing the unfused grayscale smoothing error L u , set the unfused grayscale smoothing threshold T u , if L u <T u , then the pixel at the image coordinate (x, y) is determined to be a pixel in the unfused area, and the image coordinates of the pixel in the unfused area are output;

[0021] Step S50: Calculate the overall defect intensity parameter S f , combining the pixel image coordinates of the crack area point and the pixel image coordinates of the unfused area point as well as the overall defect intensity parameters to generate a brake disc welding inspection report.

[0022] Preferably, in step S20, the denoising formula is:

[0023]

[0024] Where σ is the standard deviation of the Gaussian filter, ranging from 1.2 to 1.5, and is used to control the denoising strength; exp is the exponential function.

[0025] Preferably, the formula for edge enhancement processing is:

[0026]

[0027] Among them, α is the edge enhancement weight parameter; is the second-order derivative of the denoised image, which is used to highlight the contrast between the crack edge and the unfused contour.

[0028] Preferably, the high frequency energy response E d The formula for (x,y) is:

[0029]

[0030] Among them, E d (x, y) is the high-frequency energy response at the image coordinate (x, y), reflecting the intensity of the crack edge and the contour intensity of the unfused area; N and M are the width and height of the crack and the unfused area.

[0031] Preferably, in step S50, the overall defect intensity parameter S f The calculation formula is:

[0032]

[0033] Among them, n, m are the image coordinates of any point in the crack and unfused area, f n,n is the pixel value in the crack and unfused area; is the background grayscale average of the crack and unfused area; A d is the area of ​​the crack and the unfused area, in square pixels; N and M are the width and height of the crack and the unfused area.

[0034] Preferably, in step S50, the content of the inspection report includes the defect type, defect coordinates, defect area and defect intensity parameter S f .

[0035] Preferably, in step S40, the unfused grayscale smoothing error L u The calculation formula is:

[0036]

[0037] Among them, n, m are the image coordinates of any point in the crack and unfused area, f n,m is the pixel value in the crack and unfused area; is the background grayscale mean of the crack and unfused area; λ is the smoothing weight parameter, ranging from 0.05 to 0.2; is the second-order derivative of the pixel value in the crack and unfused area.

[0038] The present invention also provides a brake disc welding defect detection system comprising:

[0039] The image acquisition module is used to acquire the original weld area image I(x, y) by using a high-definition industrial camera to capture the image of the brake disc weld area. x and y are image coordinates, representing the horizontal and vertical coordinates of the weld area pixels, respectively. The industrial camera has a resolution of 3 million pixels or more and a frame rate of 30 frames per second.

[0040] The image preprocessing module is used to perform denoising and edge enhancement processing on the original welding area image I(x, y) obtained in step S10, including:

[0041] Use Gaussian filter to remove noise from the original welding area image I(x,y) to obtain the denoised welding area image I g (x,y);

[0042] Denoised welding area image I g (x, y) is edge enhanced to obtain the welding area enhanced image I e (x,y);

[0043] Feature extraction module, used to enhance the welding area image I based on the step S20 e (x,y), and extract the characteristics of cracks and unfused areas through multi-scale wavelet transform, including:

[0044] The weld area enhanced image is decomposed into low-frequency components L using three-layer wavelet decomposition. i and high frequency components H i , where L i is the low-frequency component of the welding area, which is used to reflect the background smoothness characteristics of the unfused area; H i is the high-frequency component of the welding area, which is used to reflect the linear characteristics of the crack edge; i is the scale of decomposition, which takes a value of 1, 2 or 3;

[0045] Calculate the high frequency energy response E d (x,y), preset high-frequency energy response threshold E t , when the high frequency energy response E d (x,y) is greater than the high-frequency energy response threshold E t When , the pixel at the image coordinate (x, y) is determined to be a crack and unfused defect pixel, and the crack and unfused defect pixels are aggregated to obtain the crack and unfused area;

[0046] Defect location and weight optimization module, used to introduce crack direction weight P j , the calculation formula is:

[0047] P j =exp(-k·θ j )

[0048] Among them, θ j is the direction angle of the crack, ranging from 0° to 180°; k is the directional attenuation coefficient;

[0049] According to the crack direction weight P j Optimize high-frequency energy response:

[0050] E d ′(x,y)=P j ·E d (x,y)

[0051] Among them, E′ d (x, y) is the high-frequency energy response after directional weight optimization;

[0052] Set the high-frequency energy threshold T of the crack c , if E′ d (x,y)>T c , then the pixel at the image coordinate (x, y) is determined to be a crack area point pixel, and the image coordinates of the crack area point pixel are output;

[0053] Introducing the unfused grayscale smoothing error L u , set the unfused grayscale smoothing threshold T u , if L u <T u , then the pixel at the image coordinate (x, y) is determined to be a pixel in the unfused area, and the image coordinates of the pixel in the unfused area are output;

[0054] Defect assessment and report generation module, used to calculate the overall defect intensity parameter S f , combining the pixel image coordinates of the crack area point and the pixel image coordinates of the unfused area point as well as the overall defect intensity parameters to generate a brake disc welding inspection report.

[0055] The present invention also provides a computer program product, including a brake disc welding defect detection program, which implements the brake disc welding defect detection method when executed by a processor.

[0056] The beneficial effect of the present invention is that compared with the technical problem in the prior art that under complex welding conditions, such as high reflectivity or complex welding textures, it is impossible to achieve efficient and accurate detection of welding cracks and unfused area defects, the present invention achieves high-precision detection of welding cracks and unfused defects through multi-scale wavelet transform combined with directional weight optimization and grayscale smoothing error correction, thereby avoiding the problems of low detection accuracy and high misjudgment rate in traditional methods, and greatly improving the efficiency and reliability of brake disc welding quality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 This is a flow chart of a first embodiment of a method for detecting welding defects in a brake disc according to the present invention.

[0059] Figure 2 This is a schematic diagram of equipment for a brake disc welding defect detection method according to the present invention. DETAILED DESCRIPTION

[0060] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0061] Example 1: Figure 1 2 is a flow chart of a first embodiment of a method for detecting welding defects of a brake disc according to the present invention, and provides a first embodiment of a method for detecting welding defects of a brake disc according to the present invention.

[0062] In a first embodiment, the brake disc welding defect detection method includes:

[0063] Step S10: Using a high-definition industrial camera to capture an image of the brake disc welding area, an original welding area image I(x, y) is obtained, where x and y are image coordinates, representing the horizontal and vertical coordinates of the pixels in the welding area, respectively. The industrial camera has a resolution of 3 million pixels or more and a frame rate of 30 frames per second.

[0064] It is understandable that the application of high-resolution industrial cameras enables the captured images to have high precision and clarity, which can effectively reflect the subtle structure and defect characteristics of the welding area and avoid defect blur or missed detection due to insufficient resolution. In addition, the frame rate of 30 frames per second ensures the real-time and continuity of image acquisition, which is particularly suitable for dynamic detection scenarios and is conducive to the stability and accuracy of subsequent image processing. For example, in the case of tiny cracks or unfused defects in the welding area, the original images captured by the industrial camera can accurately present the edges, shapes and distribution characteristics of the defects, providing high-quality input data for subsequent denoising, feature extraction and defect identification.

[0065] Step S20: Based on the original welding area image I(x, y) obtained in step S10, denoising and edge enhancement are performed on it, including:

[0066] Use Gaussian filter to remove noise from the original welding area image I(x,y) to obtain the denoised welding area image I g (x,y);

[0067] Denoised welding area image I g (x, y) is edge enhanced to obtain the welding area enhanced image I e (x,y);

[0068] It should be noted that, in step S20, the denoising formula is:

[0069]

[0070] Where σ is the standard deviation of the Gaussian filter, ranging from 1.2 to 1.5, and is used to control the denoising strength; exp is the exponential function.

[0071] In step S20, the formula for edge enhancement processing is:

[0072]

[0073] Among them, α is the edge enhancement weight parameter; is the second-order derivative of the denoised image, which is used to highlight the contrast between the crack edge and the unfused contour.

[0074] As can be understood, the original weld area image is denoised using Gaussian filtering to eliminate noise interference and produce a denoised weld area image. The application of Gaussian filtering effectively smooths the image, removing the effects of random noise and ensuring a more uniform background in the weld area, facilitating the subsequent accurate extraction of defect edges. Based on the denoised image, edge enhancement is then performed to further highlight the defect contours in the weld area and enhance the edge contrast between the crack and the unfused area. This process amplifies the image's edge information, making the difference between the crack edge and the background more pronounced, thereby improving the recognizability of weld defects.

[0075] Through the denoising and edge enhancement processing in this step, the noise in the welding area image is effectively removed, and the edge details of defects such as cracks and lack of fusion are significantly enhanced, ensuring the clarity of the image and the accuracy of feature expression, laying the foundation for the subsequent precise positioning and identification of defects. Compared with the existing technology, this step not only improves the image quality, but also ensures the accuracy and reliability of welding defect detection in complex welding backgrounds.

[0076] Step S30: Based on the welding area enhanced image I obtained in step S20 e (x,y), and extract the characteristics of cracks and unfused areas through multi-scale wavelet transform, including:

[0077] The weld area enhanced image is decomposed into low-frequency components L using three-layer wavelet decomposition. i and high frequency components H i , where L i is the low-frequency component of the welding area, which is used to reflect the background smoothness characteristics of the unfused area; H i is the high-frequency component of the welding area, which is used to reflect the linear characteristics of the crack edge; i is the scale of decomposition, which takes a value of 1, 2 or 3;

[0078] Calculate the high frequency energy response E d (x,y), preset high-frequency energy response threshold E t , when the high frequency energy response E d(x,y) is greater than the high-frequency energy response threshold E t When , the pixel at the image coordinate (x, y) is determined to be a crack and unfused defect pixel, and the crack and unfused defect pixels are aggregated to obtain the crack and unfused area;

[0079] It should be noted that this step uses multi-scale wavelet transform to effectively separate the background information of the welding area from the defect characteristics, and accurately extracts the edge features of the cracks and unfused areas. Compared with the problem of insufficient defect recognition ability in highly reflective and complex backgrounds in the existing technology, this step can accurately identify and locate the welding defect area while ensuring image details, providing a reliable basis for subsequent defect intensity analysis and report generation.

[0080] Step S40: Introducing the crack direction weight P j , the calculation formula is:

[0081] P j =exp(-k·θ j )

[0082] Among them, θ j is the direction angle of the crack, ranging from 0° to 180°; k is the directional attenuation coefficient;

[0083] According to the crack direction weight P j Optimize high-frequency energy response:

[0084] E d ′(x,y)=P j ·E d (x,y)

[0085] Among them, E′ d (x, y) is the high-frequency energy response after directional weight optimization;

[0086] Set the high-frequency energy threshold T of the crack c , if E′ d (x,y)>T c , then the pixel at the image coordinate (x, y) is determined to be a crack area point pixel, and the image coordinates of the crack area point pixel are output;

[0087] Introducing the unfused grayscale smoothing error L u , set the unfused grayscale smoothing threshold T u , if L u <T u , then the pixel at the image coordinate (x, y) is determined to be a pixel in the unfused area, and the image coordinates of the pixel in the unfused area are output;

[0088] It should be understood that by introducing the crack direction weight and the unfused grayscale smoothing error, this step can achieve high-precision identification of welding defect areas. The crack direction weight effectively enhances the energy response of the crack in a specific direction, significantly improving the accuracy of crack detection. The setting of the grayscale smoothing error accurately determines the smooth background features of the unfused area, avoiding the misjudgment of the unfused area in traditional methods. Compared with the existing technology, this step can accurately extract the edges and features of cracks and unfused areas under complex welding conditions, ensure the stability and accuracy of defect identification, and provide high-quality coordinate data and basis for subsequent defect assessment.

[0089] Step S50: Calculate the overall defect intensity parameter S f , combining the pixel image coordinates of the crack area point and the pixel image coordinates of the unfused area point as well as the overall defect intensity parameters to generate a brake disc welding inspection report.

[0090] It should be noted that the overall defect intensity parameter S f The calculation formula is:

[0091]

[0092] Among them, A d The area of ​​cracks and lack of fusion, in square pixels.

[0093] It should be understood that by calculating the overall defect intensity parameter, the defect distribution and intensity of the welding area can be quantified, reflecting the severity of defects in cracks and unfused areas. The overall defect intensity parameter makes the detection results more objective and accurate by statistically analyzing the pixel intensity deviation in the region and combining the regional area information. Compared with the existing technology, this step can not only quantitatively evaluate the distribution and intensity of defects, but also provide a complete welding quality assessment report, ensure the visualization and dataization of defect detection, and provide a reliable basis for welding quality control.

[0094] Furthermore, the present invention provides a brake disc weld defect detection system that employs the brake disc weld defect detection method described in the aforementioned embodiment, solving the technical problem of brake disc weld defect detection. Compared to the prior art, the present invention provides the same beneficial effects as the brake disc weld defect detection method described in the aforementioned embodiment. Other technical features of the present invention are the same as those disclosed in the aforementioned embodiment and are not further elaborated here.

[0095] The present invention provides a brake disc welding defect detection device, please refer to Figure 2A brake disc welding defect detection device includes: at least one processor; and a memory in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform a brake disc welding defect detection method in the above-mentioned embodiment 1. A brake disc welding defect detection device in an embodiment of the present invention may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. A brake disc welding defect detection device is merely an example and should not impose any limitations on the functions and scope of use of the embodiments of the present invention. A brake disc weld defect detection device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the brake disc weld defect detection device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input device 1007, such as a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output device 1008, such as a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003, such as a magnetic tape or hard disk; and communication device 1009. Communication device 1009 can allow a brake disc weld defect detection device to communicate wirelessly or wired with other devices to exchange data. While the figure shows a brake disc weld defect detection device with various systems, it should be understood that implementation or presence of all the illustrated systems is not required. More or fewer systems may alternatively be implemented or present.

[0096] The present invention also provides a computer program product comprising a computer program. When executed by a processor, the computer program implements the steps of the aforementioned method for detecting weld defects in brake discs. The computer program product provided by the present invention can solve the technical problem of detecting weld defects in brake discs. Compared to the prior art, the beneficial effects of the computer program product provided by the present invention are the same as those of the method for detecting weld defects in brake discs provided in the aforementioned embodiment, and are not further elaborated here.

[0097] In particular, according to the embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present invention are performed.

[0098] It should be understood that the various parts disclosed in the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any appropriate manner in any one or more embodiments or examples.

[0099] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for detecting brake disc welding defects, characterized in that: Methods include: Step S10: Using a high-definition industrial camera to capture an image of the brake disc welding area, an original welding area image I(x, y) is obtained, where x and y are image coordinates, representing the horizontal and vertical coordinates of the pixels in the welding area, respectively. The industrial camera has a resolution of 3 million pixels or more and a frame rate of 30 frames per second. Step S20: Based on the original welding area image I(x, y) obtained in step S10, denoising and edge enhancement are performed on it, including: Use Gaussian filter to remove noise from the original welding area image I(x,y) to obtain the denoised welding area image I g (x,y); Denoised welding area image I g (x, y) is edge enhanced to obtain the welding area enhanced image I e (x,y); Step S30: Based on the welding area enhanced image I obtained in step S20 e (x,y), and extract the characteristics of cracks and unfused areas through multi-scale wavelet transform, including: The weld area enhanced image is decomposed into low-frequency components L using three-layer wavelet decomposition. i and high frequency components H i , where L i is the low-frequency component of the welding area, which is used to reflect the background smoothness characteristics of the unfused area; H i is the high-frequency component of the welding area, which is used to reflect the linear characteristics of the crack edge; i is the scale of decomposition, which takes a value of 1, 2 or 3; Calculate the high frequency energy response E d (x,y), preset high-frequency energy response threshold E t , when the high frequency energy response E d (x,y) is greater than the high-frequency energy response threshold E t When , the pixel at the image coordinate (x, y) is determined to be a crack and unfused defect pixel, and the crack and unfused defect pixels are aggregated to obtain the crack and unfused area; Step S40: Introducing the crack direction weight P j , the calculation formula is: P j =exp(-k·θ j ) Among them, θ j is the direction angle of the crack, ranging from 0° to 180°; k is the directional attenuation coefficient; According to the crack direction weight P j Optimize high-frequency energy response: AND d ′(x,y)=P j ·AND d (x,y) Among them, E′ d (x, y) is the high-frequency energy response after directional weight optimization; Set the high-frequency energy threshold T of the crack c , if E′ d (x,y)>T c , then the pixel at the image coordinate (x, y) is determined to be a crack area point pixel, and the image coordinates of the crack area point pixel are output; Introducing the unfused grayscale smoothing error L u , set the unfused grayscale smoothing threshold T u , if L u <T u , then the pixel at the image coordinate (x, y) is determined to be a pixel in the unfused area, and the image coordinates of the pixel in the unfused area are output; Step S50: Calculate the overall defect intensity parameter S f , combining the pixel image coordinates of the crack area point and the pixel image coordinates of the unfused area point as well as the overall defect intensity parameters to generate a brake disc welding inspection report.

2. A brake disc welding defect detection method according to claim 1, characterized in that: In step S20, the denoising formula is: Where σ is the standard deviation of the Gaussian filter, ranging from 1.2 to 1.5, and is used to control the denoising strength; exp is the exponential function.

3. A brake disc welding defect detection method according to claim 1, characterized in that: In step S20, the formula for edge enhancement processing is: Among them, α is the edge enhancement weight parameter; is the second-order derivative of the denoised image, which is used to highlight the contrast between the crack edge and the unfused contour.

4. A brake disc welding defect detection method according to claim 1, characterized in that: In step S30, the high frequency energy response E d The formula for (x,y) is: Among them, E d (x, y) is the high-frequency energy response at the image coordinate (x, y), which reflects the intensity of the crack edge and the contour intensity of the unfused area.

5. The method for detecting brake disc welding defects according to claim 1, wherein: In step S50, the overall defect intensity parameter S f The calculation formula is: Among them, n, m are the image coordinates of any point in the crack and unfused area, f n,m is the pixel value in the crack and unfused area; is the background grayscale average of the crack and unfused area; A d is the area of ​​the crack and the unfused area, in square pixels; N and M are the width and height of the crack and the unfused area.

6. A brake disc welding defect detection method according to claim 1, characterized in that: In step S50, the content of the inspection report includes the defect type, defect coordinates, defect area and defect intensity parameter S f .

7. The method for detecting brake disc welding defects according to claim 1, wherein: In step S40, the unfused grayscale smoothing error L u The calculation formula is: Among them, n, m are the image coordinates of any point in the crack and unfused area, f n,m is the pixel value in the crack and unfused area; is the background grayscale mean of the crack and unfused area; λ is the smoothing weight parameter, ranging from 0.05 to 0.2; is the second-order derivative of the pixel value in the crack and unfused area.

8. A brake disc welding defect detection system, characterized in that: The brake disc welding defect detection system includes: The image acquisition module is used to acquire the original weld area image I(x, y) by using a high-definition industrial camera to capture the image of the brake disc weld area. x and y are image coordinates, representing the horizontal and vertical coordinates of the weld area pixels, respectively. The industrial camera has a resolution of 3 million pixels or more and a frame rate of 30 frames per second. The image preprocessing module is used to perform denoising and edge enhancement processing on the original welding area image I(x, y) obtained in step S10, including: Use Gaussian filter to remove noise from the original welding area image I(x,y) to obtain the denoised welding area image I g (x,y); Denoised welding area image I g (x, y) is edge enhanced to obtain the welding area enhanced image I e (x,y); Feature extraction module, used to enhance the welding area image I based on the step S20 e (x,y), and extract the characteristics of cracks and unfused areas through multi-scale wavelet transform, including: The weld area enhanced image is decomposed into low-frequency components L using three-layer wavelet decomposition. i and high frequency components H i , where L i is the low-frequency component of the welding area, which is used to reflect the background smoothness characteristics of the unfused area; H i is the high-frequency component of the welding area, which is used to reflect the linear characteristics of the crack edge; i is the scale of decomposition, which takes a value of 1, 2 or 3; Calculate the high frequency energy response E d (x,y), preset high-frequency energy response threshold E t , when the high frequency energy response E d (x,y) is greater than the high-frequency energy response threshold E t When , the pixel at the image coordinate (x, y) is determined to be a crack and unfused defect pixel, and the crack and unfused defect pixels are aggregated to obtain the crack and unfused area; Defect location and weight optimization module, used to introduce crack direction weight P j , the calculation formula is: P j =exp(-k·θ j ) Among them, θ j is the direction angle of the crack, ranging from 0° to 180°; k is the directional attenuation coefficient; According to the crack direction weight P j Optimize high-frequency energy response: AND d ′(x,y)=P j ·AND d (x,y) Among them, E′ d (x, y) is the high-frequency energy response after directional weight optimization; Set the high-frequency energy threshold T of the crack c , if E′ d (x,y)>T c , then the pixel at the image coordinate (x, y) is determined to be a crack area point pixel, and the image coordinates of the crack area point pixel are output; Introducing the unfused grayscale smoothing error L u , set the unfused grayscale smoothing threshold T u , if L u <T u , then the pixel at the image coordinate (x, y) is determined to be a pixel in the unfused area, and the image coordinates of the pixel in the unfused area are output; Defect assessment and report generation module, used to calculate the overall defect intensity parameter S f , combining the pixel image coordinates of the crack area point and the pixel image coordinates of the unfused area point as well as the overall defect intensity parameters to generate a brake disc welding inspection report.

9. A brake disc welding defect detection device, characterized in that: The brake disc welding defect detection device includes: a memory, a processor, and a brake disc welding defect detection program stored in the memory and executable on the processor. When the brake disc welding defect detection program is executed by the processor, the brake disc welding defect detection method described in any one of claims 1 to 7 is implemented.

10. A computer program product, characterized in that The computer program product includes a brake disc welding defect detection program, which, when executed by a processor, implements the brake disc welding defect detection method described in any one of claims 1 to 7.

Images