Uniformity characterization method and equipment for special-shaped ceramic matrix composite component

Abstract

The invention provides a uniformity characterization method and device for a special-shaped ceramic matrix composite component, and the method comprises the steps: carrying out the image foreground segmentation of a CT (Computed Tomography) image sequence, and obtaining a foreground contour image sequence of the composite component; performing pore defect segmentation on the CT image sequence based on the foreground contour image sequence to obtain a pore defect image sequence; performing three-dimensional space partitioning based on the pore defect image sequence, and obtaining an average gray value of each block obtained after partitioning; and replacing the average gray value into a corresponding block in the foreground image sequence to carry out three-dimensional reconstruction so as to complete visual representation of the uniformity of the composite material component. The problems that a material uniformity characterization method in the prior art is large in error, and the uniformity distribution difference of the special-shaped component cannot be accurately and quantitatively characterized are solved.

Description

Technical Field

[0001] The invention belongs to the technical field of metamaterial design, and in particular relates to a method and a device for characterizing the uniformity of a special-shaped ceramic-based composite material component. Background Art

[0002] Ceramic-based composites are composite materials with new properties, which are composed of ceramic materials such as silicon carbide or silicon nitride as the matrix and other materials with different properties. They have a series of advantages such as high strength, high temperature resistance, oxidation resistance, strong fracture toughness and good thermal conductivity. They are widely used in many fields such as aerospace, automobile, electronics and electrical, as well as chemical and biomedical engineering industries. They serve in various environments and are a new generation of high-temperature structural materials with broad application prospects.

[0003] However, due to the complexity and diversity of the preparation process and the anisotropy of the material itself, defects such as pores, delamination, and cracks will inevitably occur during the preparation process, resulting in uneven local density distribution of the material and making the local area fragile, affecting the reliability of the structure during service. Its defect detection and structural uniformity characterization are of great significance. At present, most research work focuses on mechanical properties, tensile strength, thermal stability, oxidation resistance and other properties. There are relatively few studies on the impact of defects such as pores generated during the preparation process of special-shaped ceramic-based composites on structural uniformity. Therefore, it is necessary to establish a method to effectively characterize the density uniformity of ceramic-based composites without damaging the material, improve the preparation process through the characterization results, and reduce the impact of pore defects generated by the preparation on the material uniformity.

[0004] Most of the existing technical research focuses on the component as a whole, and most of the components studied are regular components, or the porosity is usually used to characterize the uniformity of the sample. However, the calculation of porosity usually introduces large errors and cannot accurately quantify the differences in uniformity distribution of special-shaped components. Summary of the invention

[0005] In view of the above analysis, the present invention aims to provide a method and device for characterizing the uniformity of special-shaped ceramic-based composite material components, so as to solve the problem that the material uniformity characterization method in the prior art has large errors and cannot accurately quantify the uniformity distribution differences of special-shaped components.

[0006] The purpose of the present invention is mainly achieved through the following technical solutions:

[0007] In one aspect, the present invention provides a method for characterizing the uniformity of a special-shaped ceramic-based composite component, the method comprising:

[0008] Performing image foreground segmentation on the CT tomographic image sequence to obtain a foreground contour image sequence of the composite material component;

[0009] Performing pore defect segmentation on the CT tomographic image sequence based on the foreground contour image sequence to obtain a pore defect image sequence;

[0010] Performing three-dimensional spatial block division based on the pore defect image sequence, and obtaining an average grayscale value of each block obtained after the block division;

[0011] The average grayscale value is replaced with the corresponding block in the foreground image sequence for three-dimensional reconstruction, thereby completing the visual representation of the uniformity of the composite material component.

[0012] Furthermore, the method further includes obtaining the number of voxels and the total grayscale value of each block; and obtaining the average grayscale value of each block based on the number of voxels and the total grayscale value.

[0013] Furthermore, three-dimensional space segmentation is performed based on the pore defect image sequence by a pixel labeling method, including:

[0014] Performing region division based on each image contour in the pore defect image sequence;

[0015] Based on the coordinate information of each pixel point in different regions corresponding to the pore defect image sequence, the pixel mark value of each pixel point is calculated, and the pixels with the same pixel mark value are divided into the same block.

[0016] Furthermore, the pixel mark value of each pixel is calculated by the following formula:

[0017]

[0018] Among them, k (i,j) Represents the pixel label value of pixel point (i, j), m , n are the number of blocks on the pore defect image and the number of blocks in the fault sequence direction, respectively; l is the length of each region in the block direction; N is the number of tomographic images of the composite component; i is the coordinate of the target pixel in the block direction; j represents the coordinate of the pore defect image sequence direction, i.e., the jth image, Indicates a rounding operation.

[0019] Furthermore, performing image foreground segmentation on the CT tomographic image sequence by using an improved image foreground segmentation algorithm includes:

[0020] Acquire an initial foreground contour of each image in the CT tomographic image sequence by edge detection;

[0021] Constructing an energy function, wherein the energy function includes internal energy for controlling the smoothness and elasticity of the contour line and external energy for attracting the initial foreground contour line to move toward the target edge;

[0022] The position of the initial foreground contour line in each image is iteratively adjusted to minimize the energy function and obtain the image foreground contour that conforms to the edge of the material component.

[0023] Furthermore, obtaining an initial foreground contour of the CT tomographic image sequence by edge detection includes:

[0024] Smoothing the CT tomographic image by using a Gaussian filter;

[0025] Use the Sobel operator to calculate the gradient intensity and direction of each pixel in the image and obtain the corresponding gradient image;

[0026] Based on a non-maximum suppression algorithm, the gradient image is scanned to retain points with the maximum local gradient so as to remove non-edge pixels;

[0027] Determine edge pixels based on two preset high and low gradient thresholds, including: determine points with gradient values ​​greater than the high threshold as strong edges and retain them; determine pixels with gradient values ​​between the two thresholds as weak edges; determine weak edges connected to strong edges as edge pixels and retain them, and determine weak edge pixels not connected to strong edges as noise together with pixels less than the low threshold and remove them;

[0028] Track the detected edges to identify and connect the broken edges to form a complete edge contour;

[0029] The CT tomographic image including the edge contour is processed by a mathematical morphological closing operation to fill the hole parts of the image and obtain the initial foreground contour.

[0030] Furthermore, replacing the average grayscale value to a corresponding block in the foreground image sequence to perform three-dimensional reconstruction includes:

[0031] Replacing the average grayscale value of each block into the foreground contour image sequence;

[0032] The average grayscale value corresponding to each block is projected to each block with corresponding color and transparency using a ray casting volume rendering algorithm, thereby completing the three-dimensional reconstruction of the CT tomographic image sequence.

[0033] Furthermore, it also includes:

[0034] The average gray value of each block of the composite material component is used to characterize the density distribution of the sample;

[0035] The average gray value of each block is statistically analyzed and a line graph is drawn to complete the digital characterization of the material uniformity of the composite material component based on density distribution.

[0036] Furthermore, an adaptive threshold segmentation method is used to segment the pore defects of the CT tomographic image sequence of the composite material component, including:

[0037] Within the corresponding foreground contour, smoothing each image in the CT tomographic image sequence to remove noise;

[0038] For each pixel in each image, the weighted average of the pixel values ​​of its neighboring pixels is calculated to obtain the local threshold;

[0039] Each pixel is binarized using the local threshold to complete the pore defect segmentation.

[0040] In another aspect, a computer device is disclosed, comprising at least one processor, and at least one memory communicatively connected to the processor;

[0041] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned method for characterizing the uniformity of a special-shaped ceramic-based composite material component.

[0042] Beneficial effects of this technical solution:

[0043] 1. The present invention automatically obtains accurate foreground contour images of composite material components through an improved image foreground segmentation algorithm, and divides blocks based on the foreground contour image through a pixel marking method, calculates the grayscale average of each block, and uses the average grayscale value to achieve visual and digital representation of the uniformity of the composite material component, thereby greatly improving the accuracy of the uniformity distribution representation of various parts of the material component.

[0044] 2. The present invention can quantitatively characterize the overall and local uniformity of the sample at the same time. By dividing the data into three-dimensional spatial blocks, it can characterize the uniformity differences of various parts of the component. It realizes the uniformity characterization of special-shaped components. The method of spatial block division can handle both regular components and irregular components, thereby realizing the uniformity characterization of special-shaped components. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the entire drawings, the same reference symbols represent the same components;

[0046] Figure 1 A flow chart of a method for characterizing the uniformity of a special-shaped ceramic-based composite material component according to an embodiment of the present invention;

[0047] Figure 2 A schematic diagram of the shape of a composite material component according to an embodiment of the present invention;

[0048] Figure 3Schematic diagram of foreground contour image segmentation according to an embodiment of the present invention, Figure 3 (a) is the original CT image. Figure 3 (b) is the foreground image obtained by edge detection. Figure 3 (c) is the initial foreground contour image after filling the holes. Figure 3 Middle (d) is the final foreground contour image after iterative adjustment of the initial foreground contour;

[0049] Figure 4 is a schematic diagram of adaptive threshold segmentation according to an embodiment of the present invention, Figure 4 (a) is the original image of part of the CT tomographic image. Figure 4 (b) is the pore defect image after adaptive threshold segmentation;

[0050] Figure 5 A schematic diagram of area division according to an embodiment of the present invention;

[0051] Figure 6 A schematic diagram of block division according to an embodiment of the present invention; Figure 6 (a) is a schematic diagram of the image segmentation along the image segmentation direction. Figure 6 (b) is a schematic diagram of the blocks along the image sequence direction;

[0052] Figure 7 A schematic diagram of an image after replacement using average grayscale value according to an embodiment of the present invention;

[0053] Figure 8 A schematic diagram of material uniformity visualization after three-dimensional reconstruction according to an embodiment of the present invention;

[0054] Fig. 9 A schematic diagram of the relationship between sample density and X-ray image grayscale according to an embodiment of the present invention;

[0055] Fig.10 This is a schematic diagram of digital characterization of material uniformity according to an embodiment of the present invention. Fig.10 (a) is a schematic diagram showing the average gray value of each block in region ①. Fig.10 (b) is a schematic diagram showing the average grayscale value of each block in regions ② and ③. Fig.10 (c) is a schematic diagram showing the average gray value of each block in area ④. Fig.10 (d) is a schematic diagram representing the average grayscale value of each block in area ⑤. DETAILED DESCRIPTION

[0056] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the implementation cases of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0057] One embodiment of the present invention provides a method for characterizing the uniformity of a special-shaped ceramic-based composite material component. A special-shaped ceramic-based composite material component refers to a composite material component with a complex and irregular structure, which usually has multiple deformation parts. Due to the influence of the processing technology, deformation parts such as bending are more likely to produce pore defects. Therefore, accurate characterization of the uniformity of special-shaped ceramic-based composite material components is a necessary process for improving the preparation process of ceramic-based composite materials.

[0058] like Figure 1 As shown, the uniformity characterization method includes the following steps:

[0059] Step S1: performing image foreground segmentation on the CT tomographic image sequence to obtain a foreground contour image sequence of the composite material component.

[0060] The CT tomographic image sequence can be obtained by scanning the sample to be tested using a cone-beam CT scanning system through a high-resolution micro-focus industrial CT device.

[0061] In particular, the present invention takes a carbon fiber reinforced ceramic matrix composite material special-shaped component as an example, and the specific structure is a "U" shape, such as Figure 2 Preferably, this embodiment performs image foreground segmentation on the CT tomographic image sequence by using an improved image foreground segmentation algorithm, specifically comprising:

[0062] Acquire an initial foreground contour of each image in the CT tomographic image sequence by edge detection;

[0063] Constructing an energy function, wherein the energy function includes internal energy for controlling the smoothness and elasticity of the contour line and external energy for attracting the initial foreground contour line to move toward the target edge;

[0064] The position of the initial foreground contour line in each image is iteratively adjusted to minimize the energy function and obtain the image foreground contour that conforms to the edge of the material component.

[0065] Furthermore, obtaining an initial foreground contour of the CT tomographic image sequence by edge detection includes:

[0066] Smoothing the CT tomographic image by using a Gaussian filter;

[0067] Use the Sobel operator to calculate the gradient intensity and direction of each pixel in the image and obtain the corresponding gradient image;

[0068] Based on a non-maximum suppression algorithm, the gradient image is scanned to retain points with the maximum local gradient so as to remove non-edge pixels;

[0069] The edge pixels are determined based on the preset high and low gradient thresholds. The points with gradient values ​​greater than the high threshold are determined as strong edges and retained; the pixels with gradient values ​​between the two thresholds are determined as weak edges; the weak edges connected to the strong edges are determined as edge pixels and retained, and the weak edge pixels not connected to the strong edges are determined as noise together with the pixels less than the low threshold and removed; wherein the gradient threshold can be adaptively set according to the actual gradient of the image to be processed.

[0070] Track the detected edges to identify and connect the broken edges to form a complete edge contour;

[0071] The CT tomographic image including the edge contour is processed by a mathematical morphological closing operation to fill the hole parts of the image and obtain the initial foreground contour.

[0072] That is, this embodiment improves the active contour method by combining multiple algorithms to achieve adaptive segmentation of the foreground of the image of the special-shaped component. First, a method for adaptively obtaining the initial contour curve is established by combining Canny edge detection with mathematical morphology closing operation, which realizes the automatic setting of the initial contour curve according to the image foreground contour to be segmented.

[0073] An image in a CT tomographic image sequence is Figure 3 As shown in (a), the foreground contour of the tomographic image is initially obtained through Canny edge detection, as shown in Figure 3 Then, the tomographic image is processed using the mathematical morphology closing operation to fill the holes in the image and complete the foreground contour filling to obtain the initial contour curve. Figure 3 As shown in (c).

[0074] After obtaining the initial contour curve, the active contour method is used to segment the accurate foreground contour image. The basic principle of the active contour method is to construct an energy function, drive it by the minimum value of the energy function, make the contour curve close to the foreground edge of the image, and segment the required foreground.

[0075] More specifically, for a CT tomographic image I(x, y), C(S) = C(x(s), y(s)) is constructed as the evolution curve within the image, where s is used to uniformly represent the coordinates x and y of the curve;

[0076] The energy function is defined as:

[0077] J snake =J int +J ext +J cons ;

[0078] J intIt is the internal energy term related to the information inside the contour curve, which makes the contour curve maintain continuity and smoothness during the evolution process; J ext It is an external energy term related to image information, also called image force. Under the action of external energy term, the contour curve approaches the edge of the target; J cons It is a constraint term that provides constraints for the evolution of the curve and makes the segmentation result more accurate. It can be ignored.

[0079] The definition of the internal energy term related to the internal information of the contour curve is as follows:

[0080]

[0081] Among them, J coutin is the energy continuity term; α(s) is the elastic coefficient, which controls the contour curve to extend toward the target and maintain continuity; J smooth is the smooth energy term; β(s) is the rigidity coefficient, which controls the degree of concavity of the contour curve along with the shape of the target and maintains the smoothness of the contour curve.

[0082] The external energy term is determined by the information of the image, and a scalar representation that represents the global features of the image can be selected. Considering the gradient jump at the edge of the target, this embodiment defines the external energy term based on the gradient of the image, and its expression is shown in the following formula:

[0083]

[0084] External energy drives the contour curve to converge to the target. The coefficient γ(s) is used to adjust the convergence step size. ▽I is the gradient information of the image.

[0085] Then the energy function of this embodiment is expressed as:

[0086]

[0087] By solving the minimum value of the above formula, the contour curve can converge to the maximum gradient point of the image. The maximum gradient of the image is generally obtained at the edge of the target. Therefore, solving the minimum value of the above formula can detect the edge of the target.

[0088] That is, this embodiment uses the active contour method to deform the initial contour curve under the action of internal energy and external energy. The external energy attracts the initial contour curve to move toward the foreground edge of the object, while the internal energy maintains the smoothness of the initial contour curve. When the energy reaches the minimum, the initial contour curve moves to the foreground edge of the image to obtain the final foreground contour image. Figure 3 As shown in (d)

[0089] This embodiment uses an improved image foreground segmentation algorithm to accurately segment the foreground contour image of the tomographic image for subsequent three-dimensional space segmentation and average gray value calculation, thereby improving the accuracy of gray value calculation and the accuracy of composite material component uniformity characterization.

[0090] Step S2: performing pore defect segmentation on the CT tomographic image sequence based on the foreground contour image sequence to obtain a pore defect image sequence;

[0091] Specifically, an adaptive threshold segmentation method may be used to segment the pore defects of the CT tomographic image sequence of the composite material component, including:

[0092] Within the corresponding foreground contour, smoothing each image in the CT tomographic image sequence to remove noise;

[0093] For each pixel in each image, the weighted average of the pixel values ​​of its neighboring pixels is calculated to obtain the local threshold;

[0094] Each pixel is binarized using the local threshold to complete the pore defect segmentation.

[0095] Pixels with values ​​greater than or equal to the threshold are marked as 1, indicating the image foreground, that is, pore defects; pixels with values ​​less than the threshold are marked as 0, indicating background pixels.

[0096] That is, in order to solve the problem of uneven pore density distribution in special-shaped composite components, this embodiment adopts an adaptive threshold segmentation method to segment pore defects. Adaptive threshold segmentation determines the threshold of each pixel based on the local characteristics of the image. The segmentation threshold at each pixel position is determined by the distribution of its surrounding pixels, and the threshold with the best segmentation effect can be calculated based on the grayscale distribution characteristics of the image. Figure 4 As shown in (a), the CT image of the curved part of the "U"-shaped structure is taken as an example, and the adaptive threshold segmentation result is shown in Figure 4 As shown in (b).

[0097] Step S3: performing three-dimensional space block division based on the pore defect image sequence, and obtaining the average grayscale value of each block obtained after the block division;

[0098] Specifically, the three-dimensional space can be divided into blocks based on the pore defect image sequence by a pixel labeling method, including:

[0099] Performing region division based on each image contour in the pore defect image sequence;

[0100] Based on the coordinate information of each pixel point in different regions corresponding to the pore defect image sequence, the pixel mark value of each pixel point is calculated, and the pixels with the same pixel mark value are divided into the same block.

[0101] More specifically, the CT tomographic image of the U-shaped sample used in this embodiment is as follows: Figure 4 Considering the structural shape characteristics, in order to analyze the uniformity of pore distribution in different parts, each image in the pore defect image sequence is divided into Figure 5 The five areas shown are two side parts, two curved parts and a horizontal part of the component, which are used for subsequent block division.

[0102] Furthermore, the pore defect image regions are divided into three-dimensional blocks, such as Figure 6 (a) and Figure 6 As shown in (b), the calculation of the pixel mark value of each pixel can be expressed as the following formula:

[0103]

[0104] Among them, k (i,j) Represents the pixel label value of pixel point (i, j), m , n are the number of blocks on the pore defect image and the number of blocks in the fault sequence direction, respectively; l is the length of each region in the block direction; N is the number of tomographic images of the composite component; i is the coordinate of the target pixel in the block direction; j represents the coordinate of the pore defect image sequence direction, i.e., the jth image, Indicates a rounding operation.

[0105] Furthermore, it is assumed that the number of voxels corresponding to the block marked as k in the pore defect image sequence is mk , the total gray value is gk , then the average gray value of the block corresponding to the corresponding pixel mark value is expressed as:

[0106] G k =g k / m k .

[0107] This embodiment calculates and obtains the precise average grayscale value of each block, which can be used for subsequent uniformity visualization and digital characterization of composite material components, thereby improving the accuracy and intuitiveness of material uniformity characterization.

[0108] Step S4: replacing the average grayscale value to the corresponding block in the foreground image sequence for three-dimensional reconstruction, thereby completing the visual representation of the uniformity of the composite material component.

[0109] Specifically, this embodiment performs 3D reconstruction on the tomographic image sequence and represents the average grayscale value of 2D in 3D space, thus realizing the visualization representation of material uniformity. In order to perform 3D reconstruction, the average grayscale value of each block is first replaced into the foreground contour image sequence according to the corresponding pixel label value. The replacement result is shown in FIG. Figure 7 Then, a volume rendering algorithm is used to perform three-dimensional reconstruction on the replaced image sequence. Volume rendering is based on the direct display of volume data, directly projecting pixels with a certain color and transparency onto the display plane. This embodiment uses a ray casting volume rendering algorithm, and the result is as follows: Figure 8 As shown in the figure, different gray values ​​are mapped to different colors after 3D reconstruction; in the CT tomographic image after 3D reconstruction, the part with large gray value is the high-density area, and the part with small gray value is the low-density area. Therefore, the density distribution of the composite material can be reflected by the distribution of colors, and the uniformity of the composite material can be intuitively characterized.

[0110] Furthermore, this embodiment also uses the average grayscale value of each block of the composite material component to characterize the density distribution of the sample; the average grayscale value of each block is statistically analyzed and a line graph is drawn to complete the digital characterization of the material uniformity of the composite material component based on the density distribution.

[0111] Specifically, CT tomographic images reflect the degree of absorption of X-rays by samples through different grayscales. When the thickness of samples is not much different, the difference in sample density is reflected in the X-ray image as a difference in brightness. Therefore, CT tomographic images can represent the density of materials based on the grayscale value of the pixels in the tomographic images. The relationship between samples of different densities and the grayscale of X-ray images is as follows: Fig. 9 As shown in the figure, for the high-density part, more X-rays are absorbed and less X-rays are transmitted, which is white on the CT tomographic image; for the medium-density part, less X-rays are absorbed and more X-rays are transmitted, which is gray on the image; for the low-density part, less X-rays are absorbed and more X-rays are transmitted, which is black on the image. By calculating the average gray value of each block of the composite material component, the density distribution of the sample is characterized, and the distribution of gray values ​​is used as the basis for evaluating the uniformity of the material. By statistically analyzing the average gray value of each block and drawing a line graph, the digital representation of the uniformity of the material can be achieved.

[0112] This embodiment performs statistical analysis on the block average grayscale values ​​calculated for each region. Each part is divided into 16 blocks along the Y-axis direction (the direction of the CT tomographic image sequence), region ① is divided into 5 blocks along the X-axis direction, and regions ④ and ⑤ are divided into 8 blocks along the Z-axis direction. The block average grayscale values ​​of each part are as follows: Fig.10 As shown in (a), (b), (c) and (d), the horizontal axis represents the number of blocks along the Y axis, and the vertical axis represents the average grayscale value of each block. Fig.10It can be seen that the uniformity of the five regions in the Y direction (the length direction of the component) shows the difference in the grayscale value distribution of each part. The high and low grayscale values ​​in the figure can represent the high and low density of each part, and then the uniformity distribution difference of each part can be characterized. In addition, the area with abnormal density can be located according to the data, which can be used to generate process analysis and optimization, and improve the uniformity and reliability of composite components.

[0113] In another aspect, a computer device is provided, comprising at least one processor and at least one memory communicatively connected to the processor;

[0114] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned method for characterizing the uniformity of a special-shaped ceramic-based composite material component.

[0115] In summary, the uniformity characterization method of the special-shaped ceramic-based composite material component proposed in the present invention automatically obtains an accurate foreground contour image of the composite material component through an improved image foreground segmentation algorithm, and divides the blocks based on the foreground contour image through the pixel marking method, calculates the grayscale average value of each block, and uses the average grayscale value to realize the visual characterization and digital characterization of the uniformity of the composite material component, which greatly improves the accuracy of the uniformity distribution characterization of various parts of the material component. In addition, the present invention can quantitatively characterize the overall and local uniformity of the sample at the same time, and can characterize the uniformity differences of various parts of the component through three-dimensional spatial block division data; it realizes the uniformity characterization of special-shaped components, and the spatial block division method can process both regular components and irregular components, thereby realizing the uniformity characterization of special-shaped components.

[0116] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0117] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for characterizing the uniformity of a special-shaped ceramic matrix composite component, characterized in that: include: Performing image foreground segmentation on the CT tomographic image sequence to obtain a foreground contour image sequence of the composite material component; Performing pore defect segmentation on the CT tomographic image sequence based on the foreground contour image sequence to obtain a pore defect image sequence; Performing three-dimensional spatial block division based on the pore defect image sequence, and obtaining an average grayscale value of each block obtained after the block division; The average grayscale value is replaced with the corresponding block in the foreground image sequence for three-dimensional reconstruction, thereby completing the visual representation of the uniformity of the composite material component.

2. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to claim 1, characterized in that: The method also includes obtaining the number of voxels and the total grayscale value of each block; and obtaining the average grayscale value of each block based on the number of voxels and the total grayscale value.

3. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to claim 2, characterized in that: The three-dimensional space is divided into blocks based on the pore defect image sequence by a pixel labeling method, including: Performing region division based on each image contour in the pore defect image sequence; Based on the coordinate information of each pixel point in different regions corresponding to the pore defect image sequence, the pixel mark value of each pixel point is calculated, and the pixels with the same pixel mark value are divided into the same block.

4. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to claim 3, characterized in that: The pixel mark value of each pixel is calculated by the following formula: Among them, k (i,j) Represents the pixel label value of pixel point (i, j), m , n are the number of blocks on the pore defect image and the number of blocks in the fault sequence direction, respectively; l is the length of each region in the block direction; N is the number of tomographic images of the composite component; i is the coordinate of the target pixel in the block direction; j represents the coordinate of the pore defect image sequence direction, i.e., the jth image, Indicates a rounding operation.

5. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to claim 1, characterized in that: Performing image foreground segmentation on the CT tomographic image sequence by using an improved image foreground segmentation algorithm includes: Acquire an initial foreground contour of each image in the CT tomographic image sequence by edge detection; Constructing an energy function, wherein the energy function includes internal energy for controlling the smoothness and elasticity of the contour line and external energy for attracting the initial foreground contour line to move toward the target edge; The position of the initial foreground contour line in each image is iteratively adjusted to minimize the energy function and obtain the image foreground contour that conforms to the edge of the material component.

6. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to claim 5, characterized in that: The step of obtaining the initial foreground contour of the CT tomographic image sequence by edge detection comprises: Smoothing the CT tomographic image by using a Gaussian filter; Use the Sobel operator to calculate the gradient intensity and direction of each pixel in the image and obtain the corresponding gradient image; Based on a non-maximum suppression algorithm, the gradient image is scanned to retain points with the maximum local gradient so as to remove non-edge pixels; Based on the preset high and low gradient thresholds, edge pixels are determined. Points with gradient values ​​greater than the high threshold are determined as strong edges and retained; pixels with gradient values ​​between the two thresholds are determined as weak edges; weak edges connected to strong edges are determined as edge pixels and retained, and weak edge pixels not connected to strong edges are determined as noise together with pixels less than the low threshold and removed; Track the detected edges to identify and connect the broken edges to form a complete edge contour; The CT tomographic image including the edge contour is processed by a mathematical morphological closing operation to fill the hole parts of the image and obtain the initial foreground contour.

7. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to claim 1, characterized in that: The step of replacing the average grayscale value with a corresponding block in the foreground image sequence to perform three-dimensional reconstruction includes: Replacing the average grayscale value of each block into the foreground contour image sequence; The average grayscale value corresponding to each block is projected to each block with corresponding color and transparency using a ray casting volume rendering algorithm, thereby completing the three-dimensional reconstruction of the CT tomographic image sequence.

8. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to claim 2, characterized in that: Also includes: The average gray value of each block of the composite material component is used to characterize the density distribution of the sample; The average gray value of each block is statistically analyzed and a line graph is drawn to complete the digital characterization of the material uniformity of the composite material component based on density distribution.

9. The method for characterizing the uniformity of a special-shaped ceramic matrix composite component according to any one of claims 1 to 8, characterized in that: Using an adaptive threshold segmentation method, the CT tomographic image sequence of the composite material component is segmented for pore defects, including: Within the corresponding foreground contour, smoothing each image in the CT tomographic image sequence to remove noise; For each pixel in each image, the weighted average of the pixel values ​​of its neighboring pixels is calculated to obtain the local threshold; The local threshold is used to perform binarization processing on each pixel to complete the pore defect segmentation.

10. A computer device, characterized in that: comprising at least one processor, and at least one memory communicatively connected to the processor; The memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the method for characterizing the uniformity of a special-shaped ceramic-based composite material component according to any one of claims 1 to 9.

Images