Method and system for extracting microcracks in mesoscopic void structure of asphalt mixture

Through three-dimensional image registration and void shape factor screening, combined with Boolean difference set operation, the problem of low accuracy in micro-fracture recognition in the existing technology is solved, and the quantitative extraction of micro-fractures in asphalt mixture is realized, and the accuracy of analysis is improved. .

CN119049036APending Publication Date: 2024-11-29CHINA AIRPORT CONSTR GRP +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411106710.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The accuracy of identification of microcracks in the prior art makes it difficult to quantitatively describe the structural failure and durability attenuation process of asphalt mixtures.

Method used

By obtaining the three-dimensional tomography images before and after the damage of the asphalt mixture sample, rigid registration and three-dimensional reconstruction are performed, the three-dimensional shape factor of the void structure is calculated, and the void structure with the three-dimensional shape factor greater than the threshold is screened using the Boolean difference set operation to quantify the void structure Extract microcracks.

Benefits of technology

The accuracy of micro-crack identification is improved, and the quantitative extraction of micro-crack structures after asphalt mixture is damaged is achieved, which improves the accuracy of analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049036A_ABST
    Figure CN119049036A_ABST
Patent Text Reader

Abstract

The invention discloses a method and a system for extracting micro-cracks in a mesoscopic void structure of an asphalt mixture, relates to the technical field of road engineering, and aims to solve the problem of low recognition accuracy of the micro-cracks in the prior art by means of three-dimensional image registration before and after damage, void shape factor screening, Boolean difference set operation and the like. Quantitative extraction of the micro-crack structure after the asphalt mixture is damaged is realized, and the micro-crack identification accuracy is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of road engineering, and in particular to a method and system for extracting micro cracks in a microscopic void structure of an asphalt mixture. Background Art

[0002] Asphalt pavement is subjected to repeated vehicle loads and complex and changing environmental influences. Its performance and durability continue to decline during service. The durability evolution of asphalt pavement is a key issue in the field of road research. Asphalt mixture is a multi-void composite pavement material with a large number of void defects. Under the influence of external factors such as vehicle loads and the environment, these void defects become the base for the generation and development of microcracks. The appearance and development of microcracks are an important cause of structural damage and reduced durability of asphalt mixtures. Therefore, clarifying the development of microcracks is the key to describing the damage and structural failure behavior of asphalt mixtures.

[0003] During the damage process, the original void structure in the asphalt mixture continues to develop and penetrate to form microcracks. Therefore, the cracks are part of the void structure and do not exist independently of the voids. Moreover, the void structure of damaged asphalt concrete presents a complex spatial network structure. How to extract microcracks from the complex void network structure is the premise for analyzing the development law of microcracks in asphalt concrete. However, in the existing methods, microcracks are often identified manually by qualitative analysis, which leads to the problem of low accuracy in identifying microcracks. Summary of the invention

[0004] The purpose of the present invention is to propose a method and system for extracting microcracks in the microscopic void structure of asphalt mixture in view of the problem of low accuracy in identifying microcracks in the prior art.

[0005] The technical solution adopted by the present invention to solve the above technical problems is:

[0006] A method for extracting micro cracks in the microscopic void structure of an asphalt mixture comprises the following steps:

[0007] Step 1: Obtain tomographic images of asphalt mixture samples before and after damage;

[0008] Step 2: Reconstruct the tomographic images before and after the injury in three dimensions respectively, and perform rigid registration on the reconstructed images to obtain the registered tomographic images before and after the injury;

[0009] Step 3: extract the void structures in the registered tomographic images before and after the injury respectively;

[0010] Step 4: Calculate the three-dimensional shape factor of each void structure in the registered post-damage tomographic image;

[0011] Step 5: Screen out the void structures whose three-dimensional shape factors are greater than the threshold, i.e., the void structures containing microcracks;

[0012] Step 6: Perform a Boolean difference operation on the void structure containing microcracks and the void structure in the registered pre-damage tomographic image to obtain the three-dimensional microcracks in the damaged asphalt mixture.

[0013] Furthermore, the specific steps of step one are:

[0014] Step 1: Perform CT scanning on the asphalt mixture sample to obtain a tomographic image before damage;

[0015] Step 1 and 2: Conduct a water-saturated freeze-thaw cycle test on the asphalt mixture sample;

[0016] Step 13: Perform CT scanning on the asphalt mixture sample after the saturated freeze-thaw cycle test to obtain a tomographic image after damage.

[0017] Furthermore, the saturated water freeze-thaw cycle test is specifically as follows:

[0018] The asphalt mixture sample was placed under a negative pressure of 97.5 kPa for vacuum saturation for 15 minutes, and then immersed in water at normal pressure for 0.5 hours. After that, the sample was kept completely immersed in water and frozen at -20°C for 12 hours. The asphalt mixture sample was taken out and placed in a constant temperature water tank at 20°C for 12 hours, thus completing a freeze-thaw cycle test.

[0019] Furthermore, the three-dimensional reconstruction is performed using Avizo software.

[0020] Furthermore, the rigid registration is performed based on a rigid image registration algorithm.

[0021] Furthermore, in step three, the void structures in the registered tomographic images before and after the injury are extracted respectively by using a grayscale threshold segmentation method.

[0022] Furthermore, the specific steps of step 4 are:

[0023] Computer vision blob analysis is performed on the void structures in the registered post-injury tomographic images, each void in the void structure is marked, and then the three-dimensional shape factor of each void structure is calculated.

[0024] Further, the three-dimensional shape factor is expressed as:

[0025]

[0026] Among them, S_VA 3d is the three-dimensional shape factor of the void, A 3d is the surface area of ​​the void, and V is the volume of the void.

[0027] A system for extracting microcracks in a microscopic void structure of an asphalt mixture, the system comprising a tomographic image acquisition module, an image reconstruction module, a three-dimensional shape factor calculation module, and a three-dimensional microcrack acquisition module;

[0028] The tomographic image acquisition module is used to acquire tomographic images of the asphalt mixture sample before and after damage;

[0029] The image reconstruction module is used to perform three-dimensional reconstruction on the tomographic images before and after the injury, and to perform rigid registration on the reconstructed images to obtain the registered tomographic images before and after the injury;

[0030] The three-dimensional shape factor calculation module is used to respectively extract the void structures in the registered pre- and post-injury tomographic images, and calculate the three-dimensional shape factor of each void structure in the registered post-injury tomographic images;

[0031] The three-dimensional microcrack acquisition module is used to screen out void structures whose three-dimensional shape factors are greater than a threshold, that is, void structures containing microcracks, and then perform a Boolean difference operation on the void structures containing microcracks and the void structures in the aligned pre-damage tomographic image to obtain the three-dimensional microcracks in the asphalt mixture after damage.

[0032] Furthermore, the tomographic image acquisition module specifically performs the following steps:

[0033] Step 1: Perform CT scanning on the asphalt mixture sample to obtain a tomographic image before damage;

[0034] Step 1 and 2: Conduct a water-saturated freeze-thaw cycle test on the asphalt mixture sample;

[0035] Step 13: Perform CT scanning on the asphalt mixture sample after the saturated freeze-thaw cycle test to obtain a tomographic image after damage;

[0036] The saturated water freeze-thaw cycle test is specifically:

[0037] The asphalt mixture sample was placed under a negative pressure of 97.5 kPa for vacuum saturation for 15 minutes, and then immersed in water at normal pressure for 0.5 hours. After that, the sample was kept completely immersed in water and frozen at -20°C for 12 hours. The asphalt mixture sample was taken out and placed in a constant temperature water tank at 20°C for 12 hours, thus completing a freeze-thaw cycle test.

[0038] The three-dimensional reconstruction is performed using Avizo software;

[0039] The rigid registration is performed based on a rigid image registration algorithm;

[0040] The gray threshold segmentation method was used to extract the void structures in the registered tomographic images before and after the injury respectively;

[0041] The three-dimensional shape factor calculation module specifically performs the following steps:

[0042] Performing computer vision blob analysis on the void structure in the registered post-injury tomographic image, marking each void in the void structure, and then calculating the three-dimensional shape factor of each void structure;

[0043] The three-dimensional shape factor is expressed as:

[0044]

[0045] Among them, S_VA 3d is the three-dimensional shape factor of the void, A 3d is the surface area of ​​the void, and V is the volume of the void.

[0046] The beneficial effects of the present invention are:

[0047] This application achieves quantitative extraction of microcrack structure of damaged asphalt mixture through three-dimensional image registration before and after damage, void shape factor screening, Boolean difference operation, etc., thereby improving the accuracy of microcrack identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 Reconstructing images for three-dimensional purposes;

[0049] Figure 2 Schematic diagram of the three-dimensional images of the sample before and after the damage before and after the registration;

[0050] Figure 3 Schematic diagram of the void structure Figure 1 ;

[0051] Figure 4 Schematic diagram of the void structure Figure 2 ;

[0052] Figure 5 Schematic diagram of the void containing microcracks;

[0053] Figure 6 Schematic diagram of three-dimensional microcracks in asphalt mixture after freeze-thaw cycles. DETAILED DESCRIPTION

[0054] It should be particularly noted that, in the absence of conflict, the various embodiments disclosed in this application can be combined with each other.

[0055] Specific implementation method 1: refer to Figure 1Specifically describing this embodiment, a method for extracting microcracks in the microscopic void structure of an asphalt mixture described in this embodiment comprises the following steps:

[0056] Step 1: Obtain tomographic images of asphalt mixture samples before and after damage;

[0057] Step 2: Reconstruct the tomographic images before and after the injury in three dimensions respectively, and perform rigid registration on the reconstructed images to obtain the registered tomographic images before and after the injury;

[0058] Step 3: extract the void structures in the registered tomographic images before and after the injury respectively;

[0059] Step 4: Calculate the three-dimensional shape factor of each void structure in the registered post-damage tomographic image;

[0060] Step 5: Screen out the void structures whose three-dimensional shape factor is greater than a threshold (the threshold is set to 10), that is, the void structures containing microcracks;

[0061] Step 6: Perform a Boolean difference operation on the void structure containing microcracks and the void structure in the registered pre-damage tomographic image to obtain the three-dimensional microcracks in the damaged asphalt mixture.

[0062] There are a large number of void defects distributed in asphalt mixtures, which are the base for the generation and development of microcracks. During the damage process, the original void structure continues to develop and penetrate to form microcracks, which expand and connect to form larger cracks, eventually causing the asphalt mixture to fall off, crack, and even structural damage. Therefore, the development of microcracks can reflect the development process of asphalt mixture damage, which is of great significance for the study of asphalt pavement durability decay. In existing studies, only the microcracks in asphalt mixtures are qualitatively described, and the evolution characteristics of microcracks cannot be quantitatively described. Therefore, it is necessary to propose a quantitative extraction method for microcracks.

[0063] In recent years, the widespread application of mesoscopic detection technology and the continuous development of three-dimensional image visualization and analysis methods have provided technical support for the quantitative extraction and analysis of the mesostructure of asphalt mixtures. Therefore, microcracks in the void structure of asphalt mixtures can be quantitatively obtained with the help of mesoscopic detection technologies such as X-ray CT and three-dimensional image visualization and analysis technologies.

[0064] Microcracks that appear during the damage process of asphalt mixtures generally present as complex curved surfaces or curves that follow the shape of the aggregate surface. The difference between voids containing cracks and ordinary voids is that voids containing microcracks have very complex three-dimensional shapes, and the three-dimensional shape factor is significantly higher than that of ordinary voids. Therefore, voids containing microcracks can be screened according to the three-dimensional shape of the voids. For asphalt mixtures before damage, it has been found that the number of voids with a three-dimensional shape factor greater than 10 accounts for less than 1.5%, and the volume of this part of the voids is less than 5% of the total volume of the voids before damage. After damage, the volume of voids with this feature has increased significantly, so voids with a three-dimensional shape factor greater than 10 after damage are defined as voids containing microcracks. Microcracks "grow" in the original void structure. Microcracks can be obtained by performing Boolean difference operations on voids containing microcracks after damage and void structures before damage.

[0065] The present application can be based on the tomographic images of the same asphalt mixture sample before and after damage, and can achieve spatial coordinate and angle matching of the asphalt concrete samples before and after damage through image rigid registration. The three-dimensional shape factor of each void in the asphalt concrete can be extracted by combining computer vision blob analysis. The voids containing cracks after damage are divided based on the condition that the shape factor is >10, and the Boolean difference operation is performed on the voids containing cracks after damage and the voids before damage to extract cracks.

[0066] Step 1: Obtaining the tomographic image of the sample before damage: With the help of microscopic detection equipment (such as X-ray CT equipment), obtain the tomographic image of the asphalt mixture sample in its initial state. The image size should be greater than 600 pixels × 600 pixels, and the resolution should be greater than 72ppi.

[0067] Step 2: Sample damage test: Perform damage tests on asphalt mixture samples (such as indirect tensile test, freeze-thaw cycle test, etc.)

[0068] Step 3: Obtaining the tomographic image of the damaged sample: With the help of the same microscopic detection equipment, obtain the tomographic image of the damaged asphalt mixture sample. The size and resolution of the tomographic image after damage should be consistent with those before damage.

[0069] Step 4: Reconstruction and registration of tomographic images before and after injury: Use the acquired tomographic images to perform three-dimensional reconstruction of the samples before and after injury; based on the rigid image registration algorithm, the three-dimensional reconstructed images before and after injury are registered to eliminate the translational and rotational displacement of the samples in the image when the two tomographic images are acquired, ensuring that the samples before and after injury completely overlap in the same coordinate system.

[0070] Step 5: Extraction of voids before and after damage: Use grayscale threshold segmentation and other methods to divide the void structure in the asphalt mixture sample before and after damage.

[0071] Step 6: Calculation of the void shape factor after damage: Perform computer vision blob analysis on the void structure after damage, mark each void in the void structure, and calculate the three-dimensional shape factor of each void. The shape factor calculation formula is as follows:

[0072]

[0073] In the formula, S_VA 3d is the three-dimensional shape factor of the void; A 3d is the surface area of ​​the void (mm 2 ); V is the volume of the void (mm 3 ).

[0074] Step 7: Screening of voids containing microcracks after damage: Screen out voids with a three-dimensional shape factor greater than 10 as voids containing microcracks.

[0075] Step 8: Extraction of microcracks after damage: Perform Boolean difference operation on the voids containing microcracks after damage and the voids before damage, that is, subtract the part of the voids containing microcracks after damage from the part that overlaps with the voids before damage, to obtain the three-dimensional microcracks in the asphalt mixture after damage. After that, the three-dimensional microcrack feature analysis can be performed according to the conventional mesoscopic structure analysis method.

[0076] The present application realizes the quantitative extraction of microcracks in asphalt mixture after damage by means of three-dimensional image registration before and after damage, void shape factor screening, Boolean difference operation and the like.

[0077] Example 1: Taking the SMA-13 ​​asphalt mixture sample in the freeze-thaw test as an example, the extraction of microcracks in the microscopic void structure of the asphalt mixture is carried out according to the following steps.

[0078] Step 1: Obtaining tomographic images of samples before damage

[0079] The Phoenix v|tome|x S240 industrial CT machine produced by Phoenix|x-ray of Germany was used to scan the SMA-13 ​​sample before damage. The scanning voltage was 190kV, the current was 100mA, the scanning time was 50min, and 700 tomographic images were output, with an image size of 800 pixels × 800 pixels and a resolution of 72ppi.

[0080] Step 2: Sample damage test

[0081] The samples were subjected to a water-saturated freeze-thaw cycle test. The samples were placed under a negative pressure of 97.5 kPa for vacuum saturation for 15 minutes, then immersed in water at normal pressure for 0.5 hours, keeping the samples completely immersed in water, and frozen at -20°C for 12 hours. After the specimens were taken out, they were placed in a constant temperature water tank at 20°C for 12 hours, thus completing a freeze-thaw cycle test.

[0082] Step 3: Obtaining tomographic images of damaged samples

[0083] The SMA-13 ​​samples after the freeze-thaw cycle test were scanned using a Phoenix v|tome|x S240 industrial CT machine produced by Phoenix|x-ray of Germany. The scanning voltage was 190 kV, the current was 100 mA, the scanning time was 50 min, and 700 tomographic images were output. The image size was 800 pixels × 800 pixels and the resolution was 72 ppi.

[0084] Step 4: Reconstruction and registration of tomographic images before and after injury

[0085] Avizo software was used to reconstruct the three-dimensional tomographic images of the samples before and after the freeze-thaw test. Figure 1 As shown. And output the three-dimensional matrix of the sample. Compile a Matlab program to perform rigid registration of the three-dimensional image matrix of the sample before and after the test, eliminate the translation displacement and rotation displacement of the sample in the image before and after the test, so that the coordinate position and angle of the sample in the three-dimensional space before and after the test are the same, as shown Figure 2 As shown, for ease of presentation, the sample before freezing and thawing is rendered as a green image, and the sample after freezing and thawing is rendered as an orange image.

[0086] Step 5: Extraction of gaps before and after damage

[0087] A Matlab image processing program was compiled. According to the different gray values ​​of materials with different densities in CT images, the gray threshold segmentation method was used to divide the gaps of the samples before and after the test and binarize them. The gap images before and after the freeze-thaw test were as follows: Figure 3 , Figure 4 shown.

[0088] Step 6: Calculation of the void shape factor after damage

[0089] A Matlab program was compiled to perform computer vision blob analysis on the binarized voids after the freeze-thaw cycle test, and the three-dimensional shape factor of each void was calculated according to the three-dimensional shape factor calculation formula.

[0090] Step 7: Screening of voids containing microcracks after damage

[0091] The voids with a three-dimensional shape factor greater than 10 were screened as voids containing microcracks after the freeze-thaw cycle test, e.g. Figure 5 shown.

[0092] Step 8: Extraction of microcracks after damage

[0093] A Matlab Boolean difference operation program was compiled to perform a Boolean difference operation on the voids containing microcracks after the freeze-thaw cycle test and the voids before the test, that is, the voids containing microcracks after the freeze-thaw cycle test minus the part that overlaps with the voids before the test, and the three-dimensional microcracks in the asphalt mixture after the freeze-thaw cycle were obtained, as shown in Figure 2. Figure 6 Afterwards, the three-dimensional microcrack characteristics analysis can be performed according to the conventional mesostructure analysis method.

[0094] The innovation of this application lies in:

[0095] (1) Based on computer vision blob analysis, the three-dimensional shape factors of the voids in damaged asphalt concrete are extracted, and the voids containing crack characteristics are quantitatively divided according to the shape factors.

[0096] Microcracks generally appear as complex curved surfaces or curves that follow the shape of the aggregate surface, so voids with crack characteristics have a higher three-dimensional shape factor; for undamaged asphalt mixtures, the number of voids with a three-dimensional shape factor greater than 10 accounts for less than 1.5%, and the volume accounts for less than 5% of the total void volume; after damage, the volume of voids with a three-dimensional shape factor greater than 10 increases significantly; therefore, with the help of computer vision blob analysis, the shape factor of each void is calculated, and voids with crack characteristics are screened with a three-dimensional shape factor greater than 10.

[0097] (2) The spatial coordinates and angles of asphalt concrete samples before and after damage are matched by three-dimensional image rigid registration. The quantitative extraction of asphalt concrete cracks is achieved based on the Boolean difference operation between the voids containing cracks after damage and the voids before damage.

[0098] Microcracks "grow" in the original void structure. Cracks can be obtained by removing the voids before damage from the voids containing crack characteristics. Therefore, the asphalt concrete before and after damage is subjected to three-dimensional image rigid registration. After the spatial coordinates and angles of the asphalt concrete samples before and after damage are made consistent, a Boolean difference operation is performed on the voids containing cracks after damage and the voids before damage. The cracks generated during the damage process can be obtained, thereby realizing the quantitative extraction of cracks in asphalt concrete.

[0099] It should be noted that the specific implementation is only an explanation and description of the technical solution of the present invention, and cannot be used to limit the scope of protection of the rights. Any partial changes made according to the claims and description of the present invention should still fall within the scope of protection of the present invention.

Claims

1. A method for extracting microcracks in the microscopic void structure of asphalt mixture, characterized in that The following steps are involved: Step 1: Obtain tomographic images of asphalt mixture samples before and after damage; Step 2: Reconstruct the tomographic images before and after the injury in three dimensions respectively, and perform rigid registration on the reconstructed images to obtain the registered tomographic images before and after the injury; Step 3: extract the void structures in the registered tomographic images before and after the injury respectively; Step 4: Calculate the three-dimensional shape factor of each void structure in the registered post-damage tomographic image; Step 5: Screen out the void structures whose three-dimensional shape factors are greater than the threshold, i.e., the void structures containing microcracks; Step 6: Perform a Boolean difference operation on the void structure containing microcracks and the void structure in the registered pre-damage tomographic image to obtain the three-dimensional microcracks in the damaged asphalt mixture.

2. A method for extracting microcracks in the microscopic void structure of asphalt mixture according to claim 1, characterized in that The specific steps of step one are: Step 1: Perform CT scanning on the asphalt mixture sample to obtain a tomographic image before damage; Step 1 and 2: Conduct a water-saturated freeze-thaw cycle test on the asphalt mixture sample; Step 13: Perform CT scanning on the asphalt mixture sample after the saturated freeze-thaw cycle test to obtain a tomographic image after damage.

3. A method for extracting microcracks in the microscopic void structure of asphalt mixture according to claim 2, characterized in that The saturated water freeze-thaw cycle test is specifically: The asphalt mixture sample was placed under a negative pressure of 97.5 kPa for vacuum saturation for 15 minutes, and then immersed in water at normal pressure for 0.5 hours. After that, the sample was kept completely immersed in water and frozen at -20°C for 12 hours. The asphalt mixture sample was taken out and placed in a constant temperature water tank at 20°C for 12 hours, thus completing a freeze-thaw cycle test.

4. A method for extracting microcracks in the microscopic void structure of asphalt mixture according to claim 1, characterized in that The three-dimensional reconstruction was performed using Avizo software.

5. A method for extracting microcracks in the microscopic void structure of asphalt mixture according to claim 1, characterized in that The rigid registration is performed based on a rigid image registration algorithm.

6. A method for extracting microcracks in the microscopic void structure of asphalt mixture according to claim 1, characterized in that In the step three, the void structures in the registered tomographic images before and after the injury are respectively extracted using a grayscale threshold segmentation method.

7. The method for extracting microcracks in the microscopic void structure of asphalt mixture according to claim 1, characterized in that The specific steps of step 4 are: Computer vision blob analysis is performed on the void structures in the registered post-injury tomographic images, each void in the void structure is marked, and then the three-dimensional shape factor of each void structure is calculated.

8. A method for extracting microcracks in the microscopic void structure of asphalt mixture according to claim 1 or 7, characterized in that The three-dimensional shape factor is expressed as: Among them, S_VA 3d is the three-dimensional shape factor of the void, A 3d is the surface area of ​​the void, and V is the volume of the void.

9. A system for extracting micro cracks in the microscopic void structure of asphalt mixture, characterized in that The system includes a tomographic image acquisition module, an image reconstruction module, a three-dimensional shape factor calculation module and a three-dimensional microcrack acquisition module; The tomographic image acquisition module is used to acquire tomographic images of the asphalt mixture sample before and after damage; The image reconstruction module is used to perform three-dimensional reconstruction on the tomographic images before and after the injury, and to perform rigid registration on the reconstructed images to obtain the registered tomographic images before and after the injury; The three-dimensional shape factor calculation module is used to respectively extract the void structures in the registered pre- and post-injury tomographic images, and calculate the three-dimensional shape factor of each void structure in the registered post-injury tomographic images; The three-dimensional microcrack acquisition module is used to screen out void structures whose three-dimensional shape factors are greater than a threshold, that is, void structures containing microcracks, and then perform a Boolean difference operation on the void structures containing microcracks and the void structures in the aligned pre-damage tomographic image to obtain the three-dimensional microcracks in the asphalt mixture after damage.

10. A system for extracting micro cracks in the microscopic void structure of asphalt mixture according to claim 9, characterized in that The tomographic image acquisition module specifically performs the following steps: Step 1: Perform CT scanning on the asphalt mixture sample to obtain a tomographic image before damage; Step 1 and 2: Conduct a water-saturated freeze-thaw cycle test on the asphalt mixture sample; Step 13: Perform CT scanning on the asphalt mixture sample after the saturated freeze-thaw cycle test to obtain a tomographic image after damage; The saturated water freeze-thaw cycle test is specifically: The asphalt mixture sample was placed under a negative pressure of 97.5 kPa for vacuum saturation for 15 minutes, and then immersed in water at normal pressure for 0.5 hours. After that, the sample was kept completely immersed in water and frozen at -20°C for 12 hours. The asphalt mixture sample was taken out and placed in a constant temperature water tank at 20°C for 12 hours, thus completing a freeze-thaw cycle test. The three-dimensional reconstruction is performed using Avizo software; The rigid registration is performed based on a rigid image registration algorithm; The gray threshold segmentation method was used to extract the void structures in the registered tomographic images before and after the injury respectively; The three-dimensional shape factor calculation module specifically performs the following steps: Performing computer vision blob analysis on the void structure in the registered post-injury tomographic image, marking each void in the void structure, and then calculating the three-dimensional shape factor of each void structure; The three-dimensional shape factor is expressed as: Among them, S_VA 3d is the three-dimensional shape factor of the void, A 3d is the surface area of ​​the void, and V is the volume of the void.