A method and system for defining defects in asphalt mixture

Abstract

A method and system for defining defects in asphalt mixtures, relating to the technical field of asphalt mixture evaluation. The present invention is intended to solve the problems of low accuracy and waste of manpower and material resources in existing methods for determining cracking in asphalt mixtures. The present invention comprises: obtaining a tomographic image of an asphalt mixture specimen, extracting asphalt mixture voids from the tomographic image of the asphalt mixture specimen; dividing the asphalt mixture voids into a plurality of voids, and obtaining the surface area of ​​each void; obtaining a defect definition threshold based on the void surface area, comparing each void surface area with the defect definition threshold, and if there is a void surface area greater than or equal to the defect definition threshold, then the current asphalt mixture specimen has defects. The present invention is used to determine whether an asphalt mixture has defects.

Description

Technical Field

[0001] The present invention relates to the technical field of asphalt mixture evaluation, and in particular to a method and system for defining defects in asphalt mixture. Background Art

[0002] Asphalt pavement is the primary type of high-grade pavement in my country, offering advantages such as comfortable driving, a short construction period, and easy maintenance and repair. However, asphalt mixtures, as heterogeneous, multiphase composite materials, contain numerous initial defects such as voids and microcracks. These initial defects, in turn, weaken the mechanical properties and safety of the asphalt mixture, resulting in an actual service life far below the designed level. Traditional research methods, primarily focusing on macroscopic approaches such as indoor testing and empirical formulas, treat asphalt mixtures as homogeneous, isotropic materials. These methods are unable to truly reveal the inherent mechanical mechanisms of asphalt mixture failure, let alone specifically study the influence of the material's microstructure on the occurrence and development of damage and fracture.

[0003] Current research on inherent voids in asphalt mixtures has largely focused on connectivity, emphasizing the morphological characterization of voids and analyzing their water permeability, while neglecting their impact on the mechanical bearing capacity of asphalt mixtures. Voids are typically defined using a volumetric approach, which contradicts the observed phenomenon that small, sharp voids are more destructive than larger, rounded, spherical pores. Consequently, current methods for determining cracking in asphalt mixtures are not highly accurate. Furthermore, the number of voids is enormous, and the majority are microscopic, with minimal impact on performance. Accounting for all voids would be a waste of significant manpower and resources. Summary of the Invention

[0004] The purpose of the present invention is to solve the problems of low judgment accuracy and waste of manpower and material resources in the existing asphalt mixture cracking judgment method, and to propose a defect definition method and system for asphalt mixture.

[0005] A method for defining defects in asphalt mixtures includes the following steps:

[0006] Step 1: obtaining a tomographic image of an asphalt mixture specimen, and then extracting asphalt mixture voids from the tomographic image of the asphalt mixture specimen;

[0007] Step 2: Divide the asphalt mixture voids obtained in step 1 into multiple voids, and obtain the surface area of ​​each void:

[0008] Step 3: Obtain a defect threshold value based on the void surface area, compare each void surface area with the defect threshold value, and if there is a void surface area greater than or equal to the defect threshold value, then the current asphalt mixture specimen has a defect;

[0009] The defect definition threshold is obtained by:

[0010] S1. Accumulate the gap surface areas in ascending order, and label the gap surface areas obtained by each accumulation in sequence;

[0011] S2. Obtain the ratio A of the void surface area obtained in each accumulation to the total surface area of ​​all voids, and draw a cumulative curve based on the relationship between A and the labels obtained in S1. The void area represented by the label corresponding to the turning point of the cumulative curve is the defect definition threshold;

[0012] Among them, if the cumulative curve has no turning point, then draw tangents on both sides of the cumulative curve, and the tangents intersect to form an angle bisector of the angle. The intersection of the angle bisector and the cumulative curve is taken as the turning point of the cumulative curve.

[0013] Furthermore, the asphalt mixture specimens in step one include: Marshall specimens, rotary compaction specimens, wheel rolling specimens, and road coring specimens.

[0014] Furthermore, the extraction of asphalt mixture voids from the tomographic image of the asphalt mixture specimen is specifically as follows:

[0015] First, the tomographic image of the asphalt mixture specimen is cut into the size of the asphalt mixture specimen cross section;

[0016] Then, the tomographic image of the cut asphalt mixture specimen is preprocessed to obtain a preprocessed image;

[0017] Finally, the threshold segmentation algorithm is used to identify and extract the asphalt mixture voids in the preprocessed image.

[0018] Furthermore, the tomographic scanning image of the cut asphalt mixture specimen is preprocessed, specifically: non-local mean filtering, grayscale equalization, background detection correction, island moving processing, and hole filling processing are sequentially performed on the tomographic scanning image of the cut asphalt mixture specimen.

[0019] Furthermore, the threshold segmentation algorithm adopts a watershed segmentation algorithm.

[0020] Furthermore, the asphalt mixture void obtained in step 1 is divided into a plurality of voids, specifically by using a cutting algorithm to perform division at the throat of the asphalt mixture void to obtain a plurality of independent voids.

[0021] Furthermore, the surface area of ​​each gap is obtained using Area3d.

[0022] An asphalt mixture defect definition system includes: an asphalt mixture void extraction module, a void surface area acquisition module, and an asphalt mixture defect determination module;

[0023] The asphalt mixture extraction module is configured to obtain a tomographic image of an asphalt mixture specimen and then extract asphalt mixture voids from the tomographic image of the asphalt mixture specimen;

[0024] The void surface area acquisition module is configured to divide the asphalt mixture void into a plurality of voids and obtain the surface area of ​​each void;

[0025] The asphalt mixture void is divided into a plurality of voids, specifically: a cutting algorithm is used to divide the asphalt mixture void at the throat to obtain a plurality of independent voids;

[0026] The asphalt mixture defect determination module obtains a defect threshold value based on the void surface area, compares each void surface area with the defect threshold value, and if there is a void surface area greater than or equal to the defect threshold value, the current asphalt mixture specimen has a defect;

[0027] The defect definition threshold is obtained by:

[0028] S1. Accumulate the gap surface areas in ascending order, and label the gap surface areas obtained by each accumulation in sequence;

[0029] S2. Obtain the ratio A of the void surface area obtained in each accumulation to the total surface area of ​​all voids, and draw a cumulative curve based on the relationship between A and the labels obtained in S1. The void area represented by the label corresponding to the turning point of the cumulative curve is the defect definition threshold;

[0030] Among them, if the cumulative curve has no turning point, then draw tangents on both sides of the cumulative curve, and the tangents intersect to form an angle bisector of the angle. The intersection of the angle bisector and the cumulative curve is taken as the turning point of the cumulative curve.

[0031] Furthermore, the asphalt mixture specimens include: Marshall specimens, rotary compaction specimens, wheel rolling specimens, and road coring specimens.

[0032] Furthermore, the extraction of asphalt mixture voids from the tomographic image of the asphalt mixture specimen is specifically as follows:

[0033] First, the tomographic image of the asphalt mixture specimen is cut into the size of the asphalt mixture specimen cross section;

[0034] Then, the tomographic image of the cut asphalt mixture specimen is preprocessed to obtain a preprocessed image;

[0035] The preprocessing of the tomographic scanning image of the cut asphalt mixture specimen is specifically as follows: non-local mean filtering, grayscale equalization, background detection correction, island moving processing, and hole filling processing are sequentially performed on the tomographic scanning image of the cut asphalt mixture specimen.

[0036] Finally, the threshold segmentation algorithm is used to identify and extract the asphalt mixture voids in the preprocessed image.

[0037] The beneficial effects of the present invention are:

[0038] This invention innovatively proposes the concept of asphalt mixture defects, providing a theoretical basis for evaluating the cracking performance of asphalt mixtures. This theoretical analysis of the nature of cracking performance reveals that, from an energy perspective, surface cracking energy represents the "effective energy" required for actual cracking. The presence of void surface area reduces the energy required for cracking in asphalt mixtures. Therefore, using surface area as a void definition metric aligns with the nature of cracking and effectively avoids the phenomenon that occurs when volume is used as an evaluation metric, where "small, sharp voids are more destructive than large, rounded spherical pores." This also avoids the need to combine volumetric and morphological parameters, reducing processing complexity and thus improving the accuracy of asphalt mixture cracking determination while conserving human and material resources. Furthermore, cracking performance is significantly affected by specimen size, material properties, and other factors. Using a turning point approach to define defects aligns with cracking test phenomena and avoids the "one-size-fits-all" numerical definition that is difficult to adapt to the evaluation of diverse asphalt mixture types. This defect definition improves the reliability of evaluation results, effectively resolving the current difficulty in standardizing the cracking performance of asphalt mixtures and providing theoretical support for evaluating the practical application of asphalt mixtures. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is the process of extracting voids from asphalt mixture;

[0040] Figure 2 3D reconstruction of voids in asphalt mixture;

[0041] Figure 3 To segment the "connected voids" of asphalt mixture;

[0042] Figure 4 This is the surface area cumulative density curve of AC type asphalt mixture;

[0043] Figure 5 This is the surface area cumulative density curve of SMA type asphalt mixture;

[0044] Figure 6 This is the surface area cumulative density curve of OGFC asphalt mixture. DETAILED DESCRIPTION

[0045] Specific embodiment 1: The specific process of the defect definition method of asphalt mixture in this embodiment is as follows:

[0046] Step 1: obtaining a tomographic image of an asphalt mixture specimen, and then segmenting and extracting asphalt mixture voids in the tomographic image of the asphalt mixture specimen;

[0047] Step 11: Perform X-Ray scanning on the asphalt mixture specimen based on industrial CT, adjust relevant CT parameters, and obtain a tomographic image of the asphalt mixture specimen;

[0048] The asphalt mixture specimens include: Marshall specimens, rotary compaction specimens, wheel rolling specimens, road coring specimens, etc.; industrial CT parameters include scanning voltage, scanning current, scanning time, etc.; and a suitable slice spacing is selected to output the scanned image.

[0049] Acquire high-precision scanned images of asphalt mixtures, laying the foundation for extracting and analyzing voids in the mixture. Perform imaging accuracy analysis on asphalt mixture specimens and adjust industrial CT parameters such as scanning voltage, scanning current, and scanning time to ensure reliable scanning accuracy. Determine the appropriate tomographic slice spacing based on the asphalt mixture specimen size and scanning accuracy requirements to output tomographic images of the asphalt mixture.

[0050] Step 12: Import the tomographic scan image obtained in step 11 into the digital image processing software Avizo, perform batch cutting on the scanned image, and cut the scanned image into the size of the asphalt mixture cross section to reduce the image storage size and improve the calculation efficiency;

[0051] Step 13: Preprocess the cut image:

[0052] First, the slice image is preprocessed through non-local mean filtering, grayscale equalization, background detection correction and other algorithms to improve the image quality; then the "inner island points" generated by scanning noise are removed through island moving algorithm, hole filling algorithm and other means to reduce experimental errors.

[0053] Step 14: Use threshold segmentation methods, such as "watershed segmentation", to accurately identify and extract the asphalt mixture voids in the pre-processed image. The final recognition result is as follows: Figure 1 As shown in the figure, from left to right are the scanned original image, the gap extraction result, and the gap marking result.

[0054] Step 2: Perform three-dimensional reconstruction on the asphalt mixture voids obtained in step 1, divide the asphalt mixture voids into multiple voids, and obtain the surface area of ​​each void:

[0055] like Figure 2 As shown in the figure (AC-13, SMA-13, OGFC-13 from left to right), the connected large voids in the open-graded asphalt mixture are segmented. The existing cutting algorithm is used to segment the connected void throats into several independent voids, which is more in line with the actual mechanical bearing. Label analysis is performed on the voids to mark and distinguish them. Area3d is selected to read the void surface area. Figure 3 As shown, from left to right are: connected gaps before segmentation and connected gaps after segmentation.

[0056] Step 3: Obtain a defect threshold value based on the void surface area, compare each void surface area with the defect threshold value, and if there is a void surface area greater than or equal to the defect threshold value, then the current asphalt mixture specimen has a defect;

[0057] The defect definition threshold is obtained by:

[0058] S1. Accumulate the gap surface areas in ascending order, and label the gap surface areas obtained by each accumulation in sequence;

[0059] S2. Obtain the ratio A of the void surface area obtained in each accumulation to the total surface area of ​​all voids, and draw a cumulative curve based on the relationship between A and the label obtained in S1. The void surface area corresponding to the label of the first turning point of the cumulative curve is the defect definition threshold; for the cumulative curve with unclear turning points, select tangents on both sides of the bending line, draw an angle bisector after the tangents intersect, and use the intersection of the angle bisector and the cumulative curve as the turning point of the cumulative curve.

[0060] Specific embodiment 2: A defect definition system for asphalt mixture, comprising: an asphalt mixture void extraction module, a void surface area acquisition module, and an asphalt mixture defect determination module;

[0061] The asphalt mixture extraction module is configured to obtain a tomographic image of an asphalt mixture specimen and then extract asphalt mixture voids from the tomographic image of the asphalt mixture specimen;

[0062] The asphalt mixture specimens include: Marshall specimens, rotary compaction specimens, wheel rolling specimens, and road coring specimens;

[0063] The extraction of asphalt mixture voids from the tomographic image of the asphalt mixture specimen is specifically as follows:

[0064] First, the tomographic image of the asphalt mixture specimen is cut into the size of the asphalt mixture specimen cross section;

[0065] Then, the tomographic image of the cut asphalt mixture specimen is preprocessed to obtain the preprocessed image:

[0066] The tomographic scanning images of the cut asphalt mixture specimens were subjected to non-local mean filtering, grayscale equalization, background detection correction, island moving processing, and hole filling processing in sequence.

[0067] Finally, the threshold segmentation algorithm is used to identify and extract the asphalt mixture voids in the preprocessed image;

[0068] The void surface area acquisition module is configured to divide the asphalt mixture void into a plurality of voids and obtain the surface area of ​​each void;

[0069] The asphalt mixture void is divided into a plurality of voids, specifically: a cutting algorithm is used to divide the asphalt mixture void at the throat to obtain a plurality of independent voids;

[0070] The asphalt mixture defect determination module obtains a defect threshold value based on the void surface area, compares each void surface area with the defect threshold value, and if there is a void surface area greater than or equal to the defect threshold value, the current asphalt mixture specimen has a defect;

[0071] The defect definition threshold is obtained by:

[0072] S1. Accumulate the gap surface areas in ascending order, and label the gap surface areas obtained by each accumulation in sequence;

[0073] S2. Obtain the ratio A of the void surface area obtained in each accumulation to the total surface area of ​​all voids, and draw a cumulative curve based on the relationship between A and the labels obtained in S1. The void area represented by the label corresponding to the turning point of the cumulative curve is the defect definition threshold;

[0074] Among them, if the cumulative curve has no turning point, then draw tangents on both sides of the cumulative curve, and the tangents intersect to form an angle bisector of the angle. The intersection of the angle bisector and the cumulative curve is taken as the turning point of the cumulative curve.

[0075] Example: A method for defining defects in an asphalt mixture comprises the following steps:

[0076] Marshall specimens of asphalt mixtures of various profiles, including AC-10, AC-13, AC-16, SMA-10, SMA-13, SMA-16, OGFC-10, OGFC-13, and OGFC-16, were prepared and X-rayed using industrial CT. Based on image accuracy, a scanning voltage of 190 kV, a scanning current of 110 μA, and a scanning duration of 50 minutes were determined. Based on the Marshall specimen size, a scanning interval of 0.1 mm was selected to balance image accuracy and computational efficiency, and tomographic images of the asphalt mixture specimens were generated.

[0077] The tomographic scan images were imported into the digital image processing software Avizo, and the scanned images were batch cut into the size of asphalt mixture cross sections to reduce the image storage size and improve the computing efficiency. The sliced ​​images were preprocessed using algorithms such as non-local mean filtering, grayscale equalization, and background detection correction to improve the image quality. The "inner island points" generated by scanning noise were removed using methods such as island moving algorithm and hole filling algorithm to reduce experimental errors. The "watershed segmentation" algorithm was used to accurately identify and extract asphalt mixture voids, and finally a tomographic image of asphalt mixture voids was obtained.

[0078] Based on the void tomographic image, three-dimensional reconstruction of voids is performed, and the connected large voids of open-graded asphalt mixtures (OGFC-10, OGFC-13, and OGFC-16) are segmented. A cutting algorithm is used to segment the connected void throats into several independent voids, which is more in line with the actual mechanical bearing capacity. Label analysis is performed on the voids to mark and distinguish them, and the void surface area is read.

[0079] The void surface area is used as the only indicator to define the defects of this type of asphalt mixture. A surface area cumulative proportion method is proposed. The void surface areas are sorted and numbered from small to large. The proportion of the void surface area to the total void surface area is accumulated according to the sequence number. The turning point of the void surface area cumulative proportion is determined as the defect definition condition. The void number corresponding to the turning point is searched for the void. For the cumulative curve with unclear turning points, tangents are selected on both sides of the curved line. After the tangents intersect, an angle bisector is drawn. The intersection with the cumulative curve is the turning point. The surface area of ​​the void corresponding to the turning point is the definition surface area of ​​the defect. Void surface areas greater than or equal to this value are considered as defects of this type of asphalt mixture. Specifically, Figure 4 、 Figure 5 、 Figure 6 shown.

[0080] Based on the surface area cumulative density curve, the defective surface area of ​​each type of asphalt mixture is determined as shown in Table 1.

[0081] Table 1 Definition of asphalt mixture defects

[0082]

Claims

1. A method for defining defects in asphalt mixture, characterized in that The specific process of the method is: Step 1: obtaining a tomographic image of an asphalt mixture specimen, and then extracting asphalt mixture voids from the tomographic image of the asphalt mixture specimen; Step 2: dividing the asphalt mixture voids obtained in step 1 into a plurality of voids, and obtaining the surface area of ​​each void; Step 3: Obtain a defect threshold value based on the void surface area, compare each void surface area with the defect threshold value, and if there is a void surface area greater than or equal to the defect threshold value, then the current asphalt mixture specimen has a defect; The defect definition threshold is obtained by: S1. Accumulate the gap surface areas in ascending order, and label the gap surface areas obtained by each accumulation in sequence; S2. Obtain the ratio A of the void surface area obtained in each accumulation to the total surface area of ​​all voids, and draw a cumulative curve based on the relationship between A and the labels obtained in S1. The void area represented by the label corresponding to the turning point of the cumulative curve is the defect definition threshold; Among them, if the cumulative curve has no turning point, then draw tangents on both sides of the cumulative curve, and the tangents intersect to form an angle bisector of the angle. The intersection of the angle bisector and the cumulative curve is taken as the turning point of the cumulative curve.

2. The method for defining defects in asphalt mixture according to claim 1, wherein: The asphalt mixture specimens in step 1 include: Marshall specimens, rotary compaction specimens, wheel rolling specimens, and road coring specimens.

3. The defect definition method for asphalt mixture according to claim 1, characterized in that: The extraction of asphalt mixture voids from the tomographic image of the asphalt mixture specimen is specifically as follows: First, the tomographic image of the asphalt mixture specimen is cut into the size of the asphalt mixture specimen cross section; Then, the tomographic image of the cut asphalt mixture specimen is preprocessed to obtain a preprocessed image; Finally, the threshold segmentation algorithm is used to identify and extract the asphalt mixture voids in the preprocessed image.

4. The method for defining defects in asphalt mixture according to claim 3, wherein: The preprocessing of the tomographic scanning image of the cut asphalt mixture specimen is specifically as follows: non-local mean filtering, grayscale equalization, background detection correction, island moving processing, and hole filling processing are sequentially performed on the tomographic scanning image of the cut asphalt mixture specimen.

5. The method for defining defects in asphalt mixture according to claim 4, characterized in that: The threshold segmentation algorithm adopts the watershed segmentation algorithm.

6. The method for defining defects in asphalt mixture according to claim 5, characterized in that: The asphalt mixture void obtained in step 1 is divided into a plurality of voids, specifically by using a cutting algorithm to divide the asphalt mixture void at the throat to obtain a plurality of independent voids.

7. The method for defining defects in asphalt mixture according to any one of claims 1 to 6, characterized in that: The surface area of ​​each gap is obtained using Area3d.

8. A defect definition system for asphalt mixture, characterized by: The system includes: an asphalt mixture void extraction module, a void surface area acquisition module, and an asphalt mixture defect determination module; The asphalt mixture extraction module is configured to obtain a tomographic image of an asphalt mixture specimen and then extract asphalt mixture voids from the tomographic image of the asphalt mixture specimen; The void surface area acquisition module is configured to divide the asphalt mixture void into a plurality of voids and obtain the surface area of ​​each void; The asphalt mixture void is divided into a plurality of voids, specifically: a cutting algorithm is used to divide the asphalt mixture void at the throat to obtain a plurality of independent voids; The asphalt mixture defect determination module obtains a defect threshold value based on the void surface area, compares each void surface area with the defect threshold value, and if there is a void surface area greater than or equal to the defect threshold value, the current asphalt mixture specimen has a defect; The defect definition threshold is obtained by: S1. Accumulate the gap surface areas in ascending order, and label the gap surface areas obtained by each accumulation in sequence; S2. Obtain the ratio A of the void surface area obtained in each accumulation to the total surface area of ​​all voids, and draw a cumulative curve based on the relationship between A and the labels obtained in S1. The void area represented by the label corresponding to the turning point of the cumulative curve is the defect definition threshold; Among them, if the cumulative curve has no turning point, then draw tangents on both sides of the cumulative curve, and the tangents intersect to form an angle bisector of the angle. The intersection of the angle bisector and the cumulative curve is taken as the turning point of the cumulative curve.

9. The defect definition system for asphalt mixture according to claim 8, characterized in that: The asphalt mixture specimens include: Marshall specimens, rotary compaction specimens, wheel rolling specimens, and road coring specimens.

10. The defect definition system for asphalt mixture according to claim 9, characterized in that: The extraction of asphalt mixture voids from the tomographic image of the asphalt mixture specimen is specifically as follows: First, the tomographic image of the asphalt mixture specimen is cut into the size of the asphalt mixture specimen cross section; Then, the tomographic image of the cut asphalt mixture specimen is preprocessed to obtain a preprocessed image; The preprocessing of the tomographic scan image of the cut asphalt mixture specimen is specifically as follows: sequentially performing non-local mean filtering, grayscale equalization, background detection correction, island moving processing, and hole filling processing on the tomographic scan image of the cut asphalt mixture specimen; Finally, the threshold segmentation algorithm is used to identify and extract the asphalt mixture voids in the preprocessed image.

Images