Titanium steel composite plate interface defect detection method based on X-ray imaging

Abstract

The invention belongs to the technical field of image processing, and particularly relates to a titanium steel composite plate interface defect detection method based on X-ray imaging, which comprises the following steps: acquiring a composite plate interface X-ray image, and normalizing the image direction through Fourier transform and affine transform; performing empirical mode decomposition on each row of pixel sequence of the regularized image to obtain a rough background image; processing the rough background image through the anisotropic Gaussian kernel to obtain a filtered background image; obtaining a residual image according to the filtered background image and the normalized image; and obtaining a defect index of the composite board through the residual image, and performing interface defect detection on the composite board according to the defect index. According to the method, the background model is constructed through empirical mode decomposition, and filtering is performed in combination with the anisotropic Gaussian kernel, so that the ripple background and the defect signal of the composite board interface are effectively separated, the problem of confusion of the ripple and the defect is relieved, and the accuracy of defect detection is improved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology. More specifically, this invention relates to a method for detecting interface defects in titanium-steel composite plates based on X-ray imaging. Background Technology

[0002] Titanium-steel composite plates are widely used in chemical, marine engineering, and other fields due to their excellent comprehensive performance. Because of the specific applications of composite plates, high quality is required, making quality inspection crucial. Composite plates are often manufactured using processes such as explosive welding. The core of this process lies in achieving the bonding of different metal materials at the interface; therefore, the quality of the interface bonding directly affects the safety and lifespan of the composite plate during use.

[0003] X-ray inspection is a commonly used non-destructive testing method for assessing the interface quality of composite panels. By observing the two-dimensional X-ray transmission image of the composite panel, inspectors can intuitively analyze the image to identify and locate minute defects such as unwelded areas and inclusions on the interface.

[0004] However, achieving automated detection of defects in composite plates faces significant challenges. Due to the inherent characteristics of explosive welding processes, the background of X-ray images of composite plates is not uniform and flat, but rather exhibits a distinct wavy texture. This wavy texture originates from the wavy plastic deformation of the metal interface under explosive impact; it is not a defect in itself, but appears as a strong background interference in the image with drastic grayscale changes and complex shapes. The actual defect signal is usually weak and also possesses high-frequency characteristics, making it easily masked or confused by this strong, non-stationary wavy background. Furthermore, the randomness of the workpiece's placement angle during imaging leads to uncertainty in the direction of the wavy texture, making it difficult for conventional image processing algorithms to effectively distinguish between the background and defects. This can easily result in misidentifying the wavy texture as a defect or missing the detection of a real defect, leading to low accuracy and poor reliability in existing automated defect detection technologies for composite plates. Summary of the Invention

[0005] To address the technical problems of non-stationary wavy background textures being confused with defect signals in X-ray images of composite plates, and inconsistent wavy directions leading to low defect detection accuracy and susceptibility to missed or misjudged defects, this invention provides a method for detecting interface defects in titanium-steel composite plates based on X-ray imaging. The method includes: acquiring an X-ray image of the composite plate interface; performing image orientation regularization on the X-ray image of the composite plate interface through Fourier transform and affine transform to obtain a corrected image; using any row in the corrected image as the target row; applying empirical mode decomposition (EMD) to the target sequence composed of pixel grayscale values ​​in the target row to decompose the target sequence into several IMF components and residual terms; removing a preset number of IMF components; and superimposing the remaining IMF components and residual terms to obtain the background model sequence of the target row. The background model sequence of each row in the corrected image is sequentially combined to form a rough background image. The rough background image is then processed using an anisotropic Gaussian kernel to obtain a filtered background image, where the standard deviation of the anisotropic Gaussian kernel in the ripple texture direction is less than its standard deviation perpendicular to the ripple texture direction. Based on the difference between the pixel value of each pixel in the filtered background image and the corresponding pixel value in the corrected image, the residual index of each pixel is obtained, and the residual indices of all pixels are combined to form a residual image. The residual image is then thresholded to obtain a defect mask. Connectivity component analysis is performed on the defect mask to determine the defect region. Based on the area of ​​the defect region and the residual index of the pixels in the defect region, the defect index of the composite board is obtained. Finally, interface defects of the composite board are detected based on the defect index.

[0006] This invention first performs Fourier and affine transforms on X-ray images to regularize the direction of the corrugated texture of composite plates, overcoming the problem of inconsistent corrugation direction caused by different acquisition angles. Furthermore, this invention processes each row of the regularized image using Empirical Mode Decomposition (EMD), and reconstructs the remaining components and residual terms by removing the initial IMF component representing high-frequency defect signals, thus constructing a coarse background image free of defect information. Simultaneously, to address the inter-row discontinuity artifacts caused by row-by-row decomposition, this invention uses an anisotropic Gaussian kernel for filtering. This kernel uses a larger standard deviation perpendicular to the corrugation direction for smoothing, eliminating artifacts, while using a smaller standard deviation in the corrugation direction to preserve waveform details, resulting in an accurate filtered background image. Finally, by subtracting the texture correction image from the filtered background image, the defect signal is effectively separated from the strong noise and non-stationary corrugated background, solving the technical problem of corrugation and defect confusion. Finally, by combining defect area and severity to construct a defect index, high-precision defect detection is achieved.

[0007] Preferably, the step of performing image orientation normalization on the composite plate interface X-ray image through Fourier transform and affine transform to obtain a corrected image includes: performing a Fourier transform on the composite plate interface X-ray image to obtain a spectrum; determining the rotation angle used for the affine transform based on the spectrum, and rotating the composite plate interface X-ray image to obtain the corrected image.

[0008] This invention transforms the directional wavy texture features in the composite plate interface X-ray image into the frequency domain by performing a Fourier transform. In the frequency domain, this periodic waveform structure transforms into high-energy bright lines with concentrated energy, and its directional features are clearer and easier to identify in the spectrogram than in the original image spatial domain. Therefore, by analyzing the spectrogram to determine the main direction of the wavy texture, the global directional characteristics of the image can be captured more accurately, providing a precise angular basis for subsequent affine transformation rotation. This ensures the accuracy of image orientation regularity and avoids interference from inconsistent wavy texture orientations in subsequent line-by-line background modeling.

[0009] Preferably, determining the rotation angle used for the affine transformation based on the spectrogram includes: applying a Radon transform to the spectrogram to obtain the angle that maximizes the peak value of the Radon transform result, and determining it as the bright line direction angle in the spectrogram; subtracting the bright line direction angle from 90 degrees to obtain the rotation angle used for the affine transformation.

[0010] Preferably, the background model sequence satisfies the following relation: In the formula, The background model sequence for the target row. For the target sequence One IMF component, For the residual term of the target sequence, This represents the number of IMF components removed from all IMF components of the target sequence. This represents the number of IMF components obtained from Empirical Mode Decomposition (EMD) of the target sequence.

[0011] This invention removes the m IMF components that represent the high-frequency defect signal first and retains only the subsequent IMF components and the residual term that represents the overall trend, and then superimposes them to achieve the separation of the high-frequency defect signal from the low-frequency waveform background trend. It can accurately reconstruct a background model that does not contain defect information but only contains waveform trend, providing a clean background reference for subsequent extraction of defect signals.

[0012] Preferably, removing a preset number of IMF components includes: removing the preset number of IMF components that were first decomposed.

[0013] Preferably, the residual exponent satisfies the following relationship: In the formula, To correct the coordinates in the graph The residual index of the pixel, To correct the coordinates in the graph The grayscale value of the pixel, The coordinates in the filtered background image are The grayscale value of the pixel, Parameters to prevent division by zero errors.

[0014] This invention uses the absolute difference between the grayscale value of the correction image and the grayscale value of the filtered background image as a measure of the defect signal. By adjusting the absolute difference through the grayscale value of the filtered background image, the defect signal is highlighted. This allows the location of the defect on the composite board to obtain a higher response value in the final residual image, making it more prominent and thus improving the detectability of the defect.

[0015] Preferably, the threshold segmentation of the residual map includes: calculating a segmentation threshold using the Otsu algorithm and segmenting the residual map.

[0016] Preferably, the defective region is the region whose area is greater than a preset area threshold in the results of connected component analysis.

[0017] Preferably, the defect index satisfies the following relationship: In the formula, This represents the defect index of the composite board. The sum of the areas of all defective regions. To correct the area of ​​the graph. This is the sum of the residual exponents of the corresponding pixels in all defective regions. As a regulating factor, It is an exponential function with the natural constant as the base.

[0018] This invention not only considers the area proportion of the defect region by using the ratio of the defect area to the total image area, but also assesses the severity of defects through the average defect residual index of the defect region, thus achieving a comprehensive evaluation of the severity of defects in composite panels. Furthermore, this invention introduces an exponential function to handle the average defect severity, causing the defect index to increase at an accelerated rate with increasing severity. This makes the final defect index more sensitive to the presence of severe defects, providing a more reliable basis for defect judgment.

[0019] Preferably, the step of detecting interface defects of the composite board based on the defect index includes: determining that there are interface defects in the composite board in response to the defect index being greater than a preset defect threshold.

[0020] The beneficial effects of this invention are as follows: This invention uses Fourier transform and affine transform for global direction normalization, solving the interference problem of inconsistent ripple texture directions in X-ray images and providing a standardized image foundation for subsequent line-by-line signal analysis. This invention introduces empirical mode decomposition (IMF) technology to adaptively decompose each row of pixel sequences. By removing the initial IMF component representing the high-frequency defect signal and reconstructing the remaining components, it can accurately establish a waveform background model for each row, achieving effective separation of defect signals from background trends. Furthermore, this invention introduces anisotropic Gaussian kernels to smooth the direction perpendicular to the ripples while maintaining the sharpness of the ripple direction, thus obtaining a smooth and realistic global filtered background image without losing waveform details, solving the problem of inter-row transition artifacts caused by independent line-by-line decomposition. This invention not only locates defects through Otsu's algorithm and connected component analysis but also constructs a defect index that integrates defect area and defect severity, and uses an exponential function to amplify the weight of severe defects, making the evaluation results closer to the phenomenon that severe defects are more harmful to composite boards, thus effectively detecting interface defects in composite boards. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the X-ray imaging-based method for detecting interface defects in titanium-steel composite plates in this invention. Figure 2 This is a schematic diagram illustrating the correction in this invention; Figure 3 This is a schematic illustration of the residual diagram in this invention; Figure 4 This is a schematic illustration of a defect mask in the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] This invention discloses a method for detecting interface defects in titanium-steel composite plates based on X-ray imaging, referring to... Figure 1 This includes steps S1 to S5: S1. Acquire X-ray images of the composite plate interface and perform image orientation regularization through Fourier transform and affine transform.

[0025] It should be noted that the wavy texture in the X-ray image of the composite plate interface has obvious directionality. However, the wavy texture of the composite plate in the image may not extend in a horizontal direction. This phenomenon will affect the detection of defects in the composite plate interface in subsequent steps. Therefore, this invention uses Fourier transform and affine transform to perform image orientation regularization.

[0026] Specifically, a two-dimensional transmission image of the composite plate is acquired using an X-ray DR (Digital Radiography) device to obtain an X-ray image of the composite plate interface. A two-dimensional fast Fourier transform (FFT) was performed on the X-ray image of the composite plate interface to convert it into a spectrogram. A Radon transform was then applied to the spectrogram to obtain the angle that maximizes the peak value of the Radon transform result. This angle was used as the bright line direction angle in the spectrogram. Starting from 0 degrees to the right horizontally, the angles were increased sequentially in a counterclockwise direction. The ripple texture direction angle of the composite plate interface X-ray image was obtained by subtracting the bright line direction angle from 90 degrees. An affine transform was then used to rotate the composite plate interface X-ray image to obtain a corrected image, where the rotation angle is the ripple texture direction angle.

[0027] For example, if the wavy texture in the two-dimensional transmission image of the composite board is horizontal, then the bright line direction angle is 90 degrees, and the wavy texture direction angle is 0 degrees, and the rotation angle is 0 degrees; if the wavy texture in the two-dimensional transmission image of the composite board is vertical, then the bright line direction angle is 0 degrees, and the wavy texture direction angle is 90 degrees, and the two-dimensional transmission image of the composite board will be rotated 90 degrees counterclockwise to rotate the wavy texture in the image to a horizontal state; if the wavy texture in the two-dimensional transmission image of the composite board is tilted by 10 degrees, then the bright line direction angle is 100 degrees, and the wavy texture direction angle is -10 degrees, and the rotation angle is -10 degrees, and the two-dimensional transmission image of the composite board will be rotated 10 degrees clockwise to rotate the wavy texture to a horizontal state.

[0028] It's important to note that in the spectrogram, the periodically extending, directional waveform structure in the composite plate interface X-ray image transforms into a series of concentrated high-energy frequency components. These components are typically arranged along a high-energy bright line passing through the center of the spectrum. The direction of this high-energy bright line in the frequency domain is perpendicular to the main extension direction of the waveform structure in the composite plate interface X-ray image. Therefore, the direction of the high-energy bright line in the spectrogram is determined using the Radon transform. The Radon transform operation involves calculating the integral of the spectrogram along different angles (from 0° to 179°). Since the waveform energy is concentrated on the high-energy bright line, the integral value reaches its maximum peak when the projection angle coincides with the angle of the high-energy bright line. Therefore, by finding the angle that maximizes the Radon transform result, the direction of the high-energy bright line in the spectrogram can be determined, thereby determining the ripple texture direction angle in the composite plate interface X-ray image and correcting the ripple texture direction to be horizontal.

[0029] For example, Figure 2 This is the corrected diagram in the present invention. As can be seen from the diagram, the ripple texture has become close to a horizontal state, thus providing a data basis for subsequent empirical mode decomposition.

[0030] S2. Apply empirical mode decomposition to the sequence of gray values ​​of each row of pixels in the corrected image, and obtain a rough background image based on the decomposition results.

[0031] It should be noted that after image orientation normalization, each row of the image can be regarded as a composite signal consisting of low-frequency waveform trend, high-frequency defect signal, and noise superposition. Due to the characteristics of the explosive welding process, this waveform trend is usually not stationary; therefore, this invention uses empirical mode decomposition to obtain a rough background image based on the sequence of gray values ​​of each row of pixels in the corrected image.

[0032] Specifically, any row in the corrected image is taken as the target row, and the sequence of grayscale values ​​of pixels in the target row is taken as the target sequence. Empirical Mode Decomposition (EMD) is applied to the target sequence to decompose it into several IMF components and residual terms. The first component decomposed is selected as the target sequence. One IMF component is removed, the remaining IMF components are superimposed, and then superimposed with the residual term to obtain the background model sequence of the target row; the background model sequence of each row in the corrected image is combined into a rough background image according to the order in which they appear in the image.

[0033] For example, the target sequence is decomposed into using empirical mode decomposition. The first three IMF components are removed, and the remaining IMF components are superimposed and then superimposed with the residual term to obtain the background model sequence of the target sequence.

[0034] In one embodiment, the background model sequence satisfies the following relation: ; In the formula, The background model sequence for the target row. For the target sequence One IMF component, For the residual term of the target sequence, This represents the number of IMF components removed from all IMF components of the target sequence. This represents the number of IMF components obtained from Empirical Mode Decomposition (EMD) of the target sequence.

[0035] Among them, the first to be decomposed Each IMF component contains high-frequency defect signals. The combination of the remaining IMF components after removing these IMF components and the residual terms represents the waveform trend of the target line signal.

[0036] It should be added that, , and Both are sequences. Adding sequences together results in another sequence. The operation of adding sequences together involves adding the values ​​in the sequences one by one.

[0037] S3. The coarse background image is processed by anisotropic Gaussian kernel to obtain a filtered background image.

[0038] It should be noted that during the acquisition of the coarse background image, the empirical mode decomposition (EMD) of the correction image is performed independently row by row, which leads to a lack of correlation between the decomposition results of the target sequence and the adjacent rows of the target row. This independence may cause the output coarse background image to exhibit physically unrealistic stripe artifacts or slight grayscale jumps in the direction perpendicular to the ripples. Therefore, this invention processes the coarse background image using an anisotropic Gaussian kernel.

[0039] Specifically, an anisotropic Gaussian kernel is set such that its standard deviation in the horizontal direction is smaller than its standard deviation in the vertical direction. The anisotropic Gaussian kernel is then used to process the rough background image to obtain a filtered background image.

[0040] For example, the horizontal standard deviation of the anisotropic Gaussian kernel is 0.5, and the vertical standard deviation is 5.

[0041] The standard deviation in the horizontal direction is relatively small to maintain the sharpness of the waveform structure along the ripple direction; the standard deviation in the vertical direction is relatively large to perform cross-line smoothing and eliminate inter-line jump artifacts in the rough background image.

[0042] S4. Obtain the residual image based on the filtered background image and the correction image.

[0043] It should be noted that the filtered background image represents the normal waveform background, while the corrected image contains all the feature information of the composite board interface. By subtracting the corrected image from the filtered background image, the defects on the composite board interface can be extracted, thereby realizing the detection of defects on the composite board interface. Therefore, this invention obtains the residual image based on the filtered background image and the corrected image.

[0044] Specifically, the residual index of each pixel is obtained based on the difference between the pixel value of each pixel in the filtered background image and the pixel value of the corresponding pixel in the correction image, and the residual indices of all pixels are combined to form a residual image.

[0045] Specifically, the residual index satisfies the following relationship: ; In the formula, To correct the coordinates in the graph The residual index of the pixel, To correct the coordinates in the graph The grayscale value of the pixel, The coordinates in the filtered background image are The grayscale value of the pixel, To prevent division by zero errors in parameters, this embodiment... The value is 0.001, and the implementers can adjust it according to the actual situation. .

[0046] in, This represents the absolute grayscale difference between the corrected image and the filtered background image. The larger the value, the more significant the difference between the corrected image and the filtered background image in terms of coordinates. The greater the difference at the interface, the better the coordinates of the composite board. The more likely the corresponding location is to have interface defects, the larger the residual index will be; The smaller the value, the more accurate the comparison between the corrected image and the filtered background image on the coordinate system. The smaller the difference at the interface, the better the coordinates of the composite board. The more likely the corresponding location is to be free of defects, the smaller the residual index; to further highlight the defect area in the correction plot, by... This method makes the pixels corresponding to the defective locations on the composite board more prominent in the residual map.

[0047] For example, Figure 3 The residual image in this invention shows that the strong periodic ripple background in the original image is effectively suppressed, while the areas in the image that differ significantly from the background model are highlighted, thus effectively separating the defect signal from the complex ripple interference.

[0048] S5. Obtain the defect index of the composite board through the residual map, and perform interface defect detection of the composite board based on the defect index.

[0049] It should be noted that the normal, high-contrast wavy background in the residual image has been stripped away, while the defect signal is highlighted. This effectively alleviates the problem of confusion between wavy patterns and defects in the image. The residual image can effectively detect defects present in the interface of the composite board. Therefore, this invention obtains the defect mask image through the residual image.

[0050] Specifically, the Otsu algorithm is used on the residual map to automatically calculate the optimal segmentation threshold. Pixels in the residual map that are greater than the optimal segmentation threshold are marked as 1, and pixels that are less than or equal to the optimal segmentation threshold are marked as 0, thus obtaining the defect mask. Connectivity component analysis is performed on the defect mask to obtain each independent region. The area of ​​each region is obtained, and regions with areas greater than the area threshold are identified as defect regions. The defect index is obtained based on the area of ​​each defect region and the residual index of the corresponding pixel in the defect region.

[0051] Furthermore, when the defect index is greater than the defect threshold, the composite board has interface defects.

[0052] For example, the area threshold is 5 pixels and the defect threshold is 0.6.

[0053] Specifically, the defect index satisfies the following relationship: ; In the formula, This represents the defect index of the composite board. The sum of the areas of all defective regions. To correct the area of ​​the graph. This is the sum of the residual exponents of the corresponding pixels in all defective regions. As a regulating factor, In this embodiment, it is an exponential function with the natural constant as the base. The value is 2, and the implementers can adjust it according to the actual situation. The value of .

[0054] in, This represents the area ratio of the defective region in the residual map. The larger the value, the larger the defective area of ​​the composite board, and the larger the defect index of the composite board; the smaller the value, the smaller the defective area of ​​the composite board, and the smaller the defect index of the composite board. This represents the average defect severity of the defect area, which is the average residual index of each pixel in the defect area. The larger the value, the more likely there are serious welding defects in the composite board manufacturing process, and the larger the defect index of the composite board. The smaller the value, the smaller the defect index of the composite board.

[0055] It should be added that severe welding defects within the composite panel pose a greater risk to its use. The function's output value increases at an accelerated rate as the input value increases, therefore... right Further adjustments were made to give the composite plate with severe welding defects a larger defect index.

[0056] For example, Figure 4 As shown in the defect mask of this invention, it can be seen that only a few discrete white areas are retained in the image. These areas precisely correspond to the positions where the composite plate is not tightly fused in the correction image. Therefore, the interface defects of the composite plate can be detected by the area of ​​these white areas.

Claims

1. A method for detecting interface defects of a titanium-steel clad plate based on X-ray imaging, characterized by, The method comprises the following steps: An interface X-ray image of the composite board is acquired, and the interface X-ray image of the composite board is subjected to image direction normalization through Fourier transform and affine transformation to obtain a corrected image; Taking any row in the corrected image as a target row, an empirical mode decomposition is used on a target sequence composed of pixel gray values in the target row to decompose the target sequence into a plurality of IMF components and a residual term, after removing a preset number of IMF components, the residual term and the remaining IMF components are superimposed to obtain a background model sequence of the target row; the background model sequences of each row in the corrected image are sequentially combined into a rough background image; the rough background image is processed through an anisotropic Gaussian kernel to obtain a filtered background image, wherein the standard deviation of the anisotropic Gaussian kernel in the ripple texture direction is less than the standard deviation perpendicular to the ripple texture direction; A residual index of each pixel point is acquired according to the difference between the pixel value of each pixel point in the filtered background image and the pixel value of the corresponding pixel point in the corrected image, and the residual indices of all pixel points are combined into a residual image; the residual image is subjected to threshold segmentation to obtain a defect mask; The defect mask is subjected to connected component analysis to determine a defect region; A defect index of the composite board is acquired according to the area of the defect region and the residual indices of the pixel points in the defect region; The interface defect of the composite board is detected according to the defect index.

2. The X-ray imaging-based titanium steel clad plate interface defect detection method of claim 1, wherein, The method for obtaining the corrected image through Fourier transform and affine transformation on the interface X-ray image of the composite board comprises the following steps: Fourier transform is performed on the interface X-ray image of the composite board to obtain a frequency spectrum image; a rotation angle used by the affine transformation is determined according to the frequency spectrum image, and the interface X-ray image of the composite board is rotated to obtain the corrected image. 3.The X-ray imaging-based titanium steel clad plate interface defect detection method according to claim 2, characterized in that, The method for determining the rotation angle used by the affine transformation according to the frequency spectrum image comprises the following steps: a Radon transform is used on the frequency spectrum image to obtain an angle at which a maximum peak value of the Radon transform result is obtained, and the angle is determined as a bright line direction angle in the frequency spectrum image; a rotation angle used by the affine transformation is obtained by subtracting the bright line direction angle from 90 degrees. 4.The X-ray imaging-based titanium steel clad plate interface defect detection method according to claim 1, wherein, The background model sequence satisfies the following relationship: ； In the formula, The background model sequence for the target row. For the target sequence One IMF component, For the residual term of the target sequence, This represents the number of IMF components removed from all IMF components of the target sequence. This represents the number of IMF components obtained from Empirical Mode Decomposition (EMD) of the target sequence. 5.The X-ray imaging-based titanium steel clad plate interface defect detection method according to claim 1, wherein, The method for removing the preset number of IMF components comprises the following steps: a preset number of IMF components that are decomposed first are removed. 6.The X-ray imaging-based titanium steel clad plate interface defect detection method according to claim 1, wherein, The residual index satisfies the following relationship: ； wherein, is the residual index of the pixel point with coordinates in the correction map, is the gray value of the pixel point with coordinates in the correction map, is the gray value of the pixel point with coordinates in the filter background map, is a parameter to prevent division by zero error. 7.The X-ray imaging-based titanium steel clad plate interface defect detection method according to claim 1, wherein, The method for performing threshold segmentation on the residual image comprises the following steps: a segmentation threshold is calculated through an Otsu algorithm, and the residual image is segmented. 8.The X-ray imaging-based titanium steel clad plate interface defect detection method of claim 1, wherein, The defect region is a region with an area greater than a preset area threshold in the result of the connected component analysis. 9.The X-ray imaging-based titanium steel clad plate interface defect detection method of claim 1, wherein, The defect index satisfies the following relationship: ； In the formula, is a defect index of the composite board, is the sum of all defect area, is the area of the corrected map, is the sum of all defect area corresponding to the pixel residual index, is an adjustment factor, is an exponential function with a natural constant as the base. 10.The X-ray imaging-based titanium steel clad plate interface defect detection method according to claim 1, wherein, The method for detecting the interface defect of the composite board according to the defect index comprises the following steps: in response to the defect index being greater than a preset defect threshold, it is determined that the composite board has an interface defect.

Images