A 3D Reconstruction Modeling Method for Fiber Reinforced Composite Materials Based on CT Slice Images

Abstract

The present invention belongs to the technical field of three-dimensional reconstruction modeling of the mesoscopic structure of materials, and particularly relates to a three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images, which comprises the following steps: 1) Using CT technology to obtain mesoscopic slice images of fiber-reinforced composite materials and convert them into grayscale images; 2) Binarizing the mesoscopic CT slice grayscale images of the composite materials; 3) Identifying the fiber bundle contours of the binarized material grayscale images; 4) Smoothing the fiber bundle contours of the material grayscale images; 5) Three-dimensional reconstruction of the mesoscopic structure of the fiber-reinforced composite materials. The present invention provides a method for three-dimensional reconstruction modeling at the mesoscopic scale, which greatly reduces the cost and error of manual three-dimensional reconstruction modeling, can accurately and efficiently quantify and characterize the changes in the morphology and size of the mesoscopic structure of materials, and has good engineering popularization prospects.

Description

Technical Field

[0001] The present invention belongs to the technical field of three-dimensional reconstruction modeling of the mesoscopic structure of materials, and particularly relates to a three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images. Background Art

[0002] Due to their excellent mechanical properties such as high specific strength, high specific stiffness, and low density, fiber-reinforced composite materials have broad application prospects in engineering fields such as aerospace, vehicles, and construction. However, the mesoscopic structure of fiber-reinforced composite materials is complex and variable. During the preparation process, under the action of high temperature, stress, and time, the fiber bundles will inevitably be squeezed, twisted, and deformed with each other, forming an irregular mesoscopic braided structure, and the macroscopic properties of the material depend on its internal microscopic and mesoscopic structures. The macroscopic mechanical properties of composite materials have a strong correlation with parameters such as the morphology and shape of the mesoscopic braided structure. The distortion and deformation of the mesoscopic structure will inevitably affect the modulus and strength of the material, endangering the reliability and safety of engineering applications.

[0003] In order to ensure the safe application of fiber-reinforced composite materials in engineering fields, it is necessary to study the influence law of the mesoscopic braided characteristics of materials on the macroscopic mechanical properties of materials. However, existing mesoscopic modeling methods are all based on certain approximations and simplified assumptions, lacking a three-dimensional reconstruction modeling method that can characterize and reflect the true mesoscopic structure characteristics.

[0004] Chinese Patent CN 111063402 A discloses a mesoscopic scale geometric reconstruction method for fiber-reinforced composite materials based on the Monte Carlo method, which can obtain data such as the number, filling rate, average orientation, and length distribution of fibers in the geometric reconstruction body. However, this method cannot reflect the morphology and size change characteristics of fiber bundles at the mesoscopic scale. There is no other reconstruction modeling method in China that can truly restore the mesoscopic structure characteristics.

[0005] With the wide application of computer image processing methods in the analysis and extraction of the micro-mesoscopic structure characteristics of materials, various image processing algorithms have begun to be applied to the extraction and characterization of the micro-mesoscopic tissue and structure characteristics of materials. Therefore, developing a three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on image algorithms can significantly reduce the labor cost of three-dimensional reconstruction of the mesoscopic structure of fiber-reinforced composite materials while improving the accuracy of modeling. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention provides a three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images. The core image processing algorithms used include a pore filling algorithm and a contour extraction algorithm.

[0007] The present invention is implemented as follows: A three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images is provided, including the following steps:

[0008] 1) Use CT technology to obtain microscopic slice images of fiber-reinforced composites and convert them into grayscale images;

[0009] 2) Binarize the grayscale images of the microscopic CT slices of the composite material;

[0010] 3) Identify and extract the contours of the binarized microscopic grayscale images of the composite material;

[0011] 4) Smooth the contours of the microscopic grayscale images of the material;

[0012] 5) Three-dimensional reconstruction of the microscopic structure of fiber-reinforced composites.

[0013] Preferably, step 1) specifically includes the following steps:

[0014] Use X-ray three-dimensional CT scanning technology to obtain slice images at different positions inside the fiber-reinforced composite material. When scanning, the cross-section of the CT slice is perpendicular to the main weaving direction of the composite material.

[0015] Further preferably, step 2) specifically includes the following steps:

[0016] Use the threshold segmentation method to convert the grayscale image of the CT slice obtained in step 1) into a black-and-white binary image. In the black-and-white binary image, black represents the fiber bundle and the pixel value is 0; white represents the matrix phase and the pixel value is 1.

[0017] Further preferably, step 3) specifically includes the following steps:

[0018] 301) Adopt a pore filling algorithm to fill the tiny pores inside the fiber bundle / matrix. The filling algorithm traverses the binary image with a window of a certain size (such as a square or rectangular window). If the pixel values of the points on the window boundary are the same, then make the pixel values of the points inside the window equal to the pixel values of the boundary points. The shape and size of the window and the number of times the window traverses should be determined according to the actual situation.

[0019] 302) Extract the contour points of the fiber bundle. The specific operation method is: traverse the binary image with a window of a certain size. If the pixel values of the points on the window boundary are all 0 (a pixel value of 0 indicates a fiber bundle), then make the pixel values of the points inside the window equal to 1. The shape and size of the window and the number of times the window traverses should be determined according to the actual situation.

[0020] (303) Group the contour points by curves and sort the contour points of each group. The methods for grouping and sorting are as follows: First, arbitrarily select a contour point A, and based on point A, calculate the distances between other contour points and point A. If the distance between contour point B and point A is the smallest and the distance is less than x, then point A and point B are adjacent; then, based on contour point B, and so on, until the minimum distance is greater than x, then the above contour points belong to the same contour line. Then, select any one of the remaining contour points as the reference point and repeat the above process until all contour points are grouped. Among them, the value of the critical distance x is determined according to the actual situation. During the process of grouping the contour points using this method, the sorting of the contour points is also carried out and completed simultaneously.

[0021] Further preferably, step 4) specifically includes the following steps:

[0022] (401) For each group of contour curves, within a certain distance range, only take one contour point to reduce the number of contour points of the fiber bundle contour line and improve the smoothness of the fiber bundle contour line.

[0023] (402) Group the reduced fiber bundle contour points by curves and output the three-dimensional coordinates of the contour points.

[0024] Further preferably, step 5) specifically includes the following steps:

[0025] (501) In the UG software, use the "Spline - Through Points" function to import the contour points successively according to each independent curve.

[0026] (502) Stack the fiber bundle contours of each slice cross-section along the height direction, and then use the "Mesh Surface - Ruled" function to connect the corresponding fiber bundle contour lines in different slices successively to reconstruct the mesoscopic fiber bundle model of the fiber-reinforced composite material.

[0027] Further preferably, steps 2) to 4) are implemented by relying on commercial computational programming software.

[0028] Compared with the prior art, the advantages of the present invention are as follows:

[0029] 1. The present invention combines the gray-scale features of the image, and can conveniently and efficiently extract the fiber bundle contour through the window algorithm function, and can be programmed through computer software to realize the quick operation of batch CT slice images, greatly reducing the cost and error of the manual operation method;

[0030] 2. The present invention can not only be used for the three-dimensional reconstruction and modeling of the mesoscopic scale of fiber-reinforced composite materials, but also be applicable to the three-dimensional reconstruction and modeling of multi-phase composite materials / structures, and has good engineering popularization. Description of the Drawings

[0031] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments:

[0032] Figure 1 This is a simple flowchart of the three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images of the present invention;

[0033] Figure 2 This is a CT slice image of a certain fiber-reinforced composite material;

[0034] Figure 3 This is a binary image of the CT slice of the fiber-reinforced composite material;

[0035] Figure 4 This is a flowchart of the binary image processing algorithm;

[0036] Figure 5 This is the binary image of the CT slice after being processed by the pore filling algorithm;

[0037] Figure 6 This is a scatter plot of the fiber bundle contour extraction results of the slice image;

[0038] Figure 7 This is the contour curve of the fiber bundle contour points grouped and sorted by curve in the slice image;

[0039] Figure 8 This is the fiber bundle contour curve graph after smoothing processing;

[0040] Figure 9 This is the fiber bundle contour curve after processing the CT slice image group;

[0041] Figure 10 This is the stacked graph of the fiber bundle contour curve along the height direction;

[0042] Figure 11 This is the three-dimensional model of the meso-cell of the fiber-reinforced composite material obtained by reconstruction;

[0043] Figure 12 This is a comparison graph between the meso-scale three-dimensional reconstruction model of the composite material and the original CT slice image. Specific Embodiments

[0044] To make the advantages of the technical method of the present invention clearer and make it easier for professionals in the field to implement, the following further describes the specific operations of the present invention in conjunction with the accompanying drawings and embodiments of the invention. It should be noted that the embodiments are only for making it easier for those in the field to understand and master the advantages and operation methods of the present invention, and cannot limit the application of the present invention in the three-dimensional reconstruction modeling of fiber-reinforced composite materials and other related fields.

[0045] The present invention provides a three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images. Refer to Figure 1, which is a simplified flowchart of a three-dimensional reconstruction modeling method for a certain fiber-reinforced composite material provided by an embodiment of the present invention. The specific steps are as follows:

[0046] S101: Acquisition of CT slice images of fiber-reinforced composite materials. Using X-ray three-dimensional CT scanning technology, slice images at different positions inside the fiber-reinforced composite material are obtained, as Figure 2 shown. During scanning, the CT slice cross-section should be perpendicular to the main weaving direction of the composite material.

[0047] S102: Binarization of CT slice images. For the CT slice images of the composite material obtained in S101, taking Figure 2 as an example, the CT image of the composite material is converted into a black-and-white binary image by using the threshold segmentation method, as Figure 3 shown. At this time, Figure 3 the black pixel area in represents the reinforcing fiber bundle, and the pixel value is 0; the white area represents the matrix phase, and the pixel value is 1.

[0048] S103: Pore filling, contour recognition and grouping of binary images. It is divided into the following three steps:

[0049] (1) Adopt a pore filling algorithm to fill the tiny pores inside the fiber bundle / matrix. The filling algorithm traverses the binary image with a window of a certain size (such as a square or rectangular window). If the pixel values of the points on the window boundary are the same, the pixel values of the points inside the window are set equal to the pixel values of the points on the curve. The shape and size of the window and the number of times the window traverses should be determined according to the actual situation. After processing with the pore filling algorithm, the obtained CT slice binary image is as Figure 5 shown.

[0050] (2) Extract the contour points of the fiber bundle. Traverse the Figure 5 binary image with a window of a certain size. If the pixel values of the points on the window boundary are all 0 (a pixel value of 0 indicates a fiber bundle), the pixel values of the points inside the window are set equal to 1. The shape and size of the window and the number of traversals should be determined according to the actual situation. The extracted fiber bundle contour points are as Figure 6 shown.

[0051] (3) Group and sort the contour points according to the curve. First, select Figure 6Any one of the contour points A of the fiber bundle, and taking point A as a reference, calculate the distances between other contour points and point A. If the distance between contour point B and point A is the smallest and less than 2, then point A and point B are adjacent; then taking contour point B as a reference, and so on, until the minimum distance is greater than 2, then the above contour points belong to the same contour line. Then select any one of the remaining contour points as a reference point, repeat the above process until all contour points are grouped. During the process of grouping the contour points by this method, the sorting of the contour points is also carried out and completed at the same time. After grouping and sorting, the contour points can be connected according to the grouping and order to obtain Figure 7 the fiber bundle contour shown in

[0052] S104: Contour smoothing. For each group of contour curves, within a certain distance range, only take one contour point to reduce the number of contour points of the fiber bundle contour line and improve the smoothness of the fiber bundle contour line( Figure 8 ). Group the reduced fiber bundle contour points according to the curves and output the three-dimensional coordinates of the contour points.

[0053] S105: Three-dimensional reconstruction of the mesoscopic structure of fiber-reinforced composites. It is divided into the following two steps:

[0054] (1) Import the contour points into the modeling software. In the UG software, use the "Spline - Through Points" function to import the contour points in sequence according to each independent curve. The fiber bundle contours of each imported slice cross-section are as shown in Figure 9 shown.

[0055] (2) Three-dimensional reconstruction modeling. Stack the fiber bundle contours of each slice cross-section along the height direction (see Figure 10 ), and then use the "Mesh Surface - Ruled" function to connect the corresponding fiber bundle contour lines in different slices in sequence to reconstruct the mesoscopic fiber bundle model of the fiber-reinforced composite material, as shown in Figure 11 . The comparison between the finally reconstructed three-dimensional model of the fiber bundle and the original CT slice image is as shown in Figure 12 shown.

[0056] Those skilled in the art can implement S102 to S104 by programming with computer language according to the said flowchart. When a large number of composite material CT slice images need to be processed, batch processing will greatly reduce the time cost of extracting the mesoscopic fiber bundle contours of the composite material.

[0057] The above are only the preferred embodiments of the present invention, and do not limit the implementation manners and protection scope of the present invention accordingly. For those skilled in the art, it should be realized that all equivalent replacements and obvious changes made by using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images, characterized in that, It includes the following steps: 1) Using CT technology, obtain the mesoscopic slice images of the fiber-reinforced composite material and convert them into grayscale images; 2) Binarize the mesoscopic CT slice grayscale images of the composite material; 3) Identify the contours of the binarized mesoscopic grayscale images of the material, which specifically includes the following steps: 301) Use the pore filling algorithm to fill the tiny pores inside the fiber bundle / matrix. The filling algorithm traverses the binary image with a window of a certain size. If the pixel values of the points on the window boundary are the same, then set the pixel values of the points inside the window equal to the pixel values of the boundary points; 302) Extract the contour points of the fiber bundle. The specific operation method is: Traverse the binary image with a window of a certain size. If the pixel values of the points on the window boundary are all 0, then set the pixel values of the points inside the window equal to 1; 303) Group the contour points by curves and sort the contour points of each group. The grouping and sorting operation method is: First, arbitrarily select a contour point A, and based on point A, calculate the distances between other contour points and point A. If the distance between contour point B and point A is the smallest and the distance is less than x, then point A and point B are adjacent; then, based on contour point B, and so on, until the minimum distance is greater than x, then the above contour points belong to the same contour line; then, select any one of the remaining contour points as the reference point and repeat the above process until all contour points are grouped; During the process of grouping the contour points using this grouping and sorting operation method, the sorting of the contour points is also carried out and completed simultaneously; 4) Smooth the contours of the mesoscopic grayscale images of the material; 5) Three-dimensional reconstruction of the mesoscopic structure of the fiber-reinforced composite material, which specifically includes the following steps: 501) In the UG software, use the "Spline - Through Points" function to import the contour points in sequence according to each independent curve; 502) Stack the fiber bundle contours of each slice cross-section along the height direction, and then use the "Mesh Surface - Ruled" function to connect the corresponding fiber bundle contour lines in different slices in sequence to reconstruct the mesoscopic fiber bundle model of the fiber-reinforced composite material.

2. The three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images according to claim 1, characterized in that, Step 1) specifically includes the following steps: Using X-ray three-dimensional CT scanning technology, obtain the slice images at different positions inside the fiber-reinforced composite material. When scanning, the CT slice cross-section is perpendicular to the main weaving direction of the composite material.

3. The three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images according to claim 1, characterized in that, Step 2) specifically includes the following steps: Use the threshold segmentation method to convert the CT slice grayscale image obtained in step 1) into a black-and-white binary image. In the black-and-white binary image, black represents the fiber bundle with a pixel value of 0; white represents the matrix phase with a pixel value of 1.

4. The three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images according to claim 1, characterized in that, Step 4) specifically includes the following steps: 401) For each group of contour curves, within a certain distance range, only take one contour point to reduce the number of contour points of the fiber bundle contour line and improve the smoothness of the fiber bundle contour line; 402) Group the reduced fiber bundle contour points by curves and output the three-dimensional coordinates of the contour points.

5. The three-dimensional reconstruction modeling method for fiber-reinforced composite materials based on CT slice images according to any one of claims 1-4, characterized in that, Steps 2) to 4) are implemented by programming with commercial computer software.

Images