Asphalt pavement aggregate peeling evaluation and loosening trend measurement method

Abstract

The invention provides an asphalt pavement aggregate peeling evaluation and loose trend measurement method, which comprises the following steps of: 1) acquiring pavement three-dimensional data, and preprocessing to obtain pavement three-dimensional precise data; 2) identifying and segmenting an aggregate area and a non-aggregate area in the pavement three-dimensional precision data, and judging an aggregate peeling condition; 3) according to the aggregate stripping condition, establishing an asphalt pavement looseness degree evaluation index, and evaluating the asphalt pavement aggregate looseness condition; 4) based on the pavement three-dimensional precision data, establishing an asphalt pavement three-dimensional analysis model, simulating the aggregate loosening and peeling condition of the asphalt pavement three-dimensional analysis model under the action of vehicle load, and estimating the degradation development trend of the asphalt pavement; by means of the method, the accuracy of spalling evaluation can be effectively improved, the development trend of pavement looseness can be effectively measured, the method is suitable for current asphalt pavement aggregate spalling evaluation and looseness trend measurement application, objectivity and accuracy are higher, and therefore the asphalt pavement performance condition can be more accurately obtained.

Description

Technical Field

[0001] The invention relates to the technical field of preventive maintenance of asphalt pavements, and in particular to an asphalt pavement aggregate spalling evaluation and loosening tendency measurement method. Background Art

[0002] Aggregate spalling and loosening of asphalt pavement are one of the early damages of asphalt pavement and are the focus of preventive maintenance. Pavement loosening usually manifests as the spalling of aggregates in asphalt mixture, followed by the formation of roughness, exposed bones, peeling, surface aggregate loss and even small pits on the surface of asphalt pavement. It often occurs in drainage pavements using porous asphalt mixtures, and there are mainly two forms: adhesion failure caused by the decrease in the force between the asphalt and aggregate interface, and cohesion failure caused by the aging or reduced fatigue durability of the asphalt mortar that bonds the aggregates.

[0003] However, in the existing research on the looseness evaluation of pavement, the area of ​​the loose area is mainly used as the looseness measurement index. In the degree division, the size of the missing aggregate is used as the standard. The missing of fine aggregate on the surface is considered as mild looseness, and the missing of coarse aggregate on the surface is considered as severe looseness. This type of evaluation method brings certain difficulties to the evaluation of actual loose damage due to its ambiguity and subjectivity. On the one hand, in the early damage of the pavement, the looseness is only manifested as the peeling and missing of individual aggregates, and there is no obvious aggregation behavior between the missing aggregates. This damage belongs to the local behavior phenomenon of the asphalt pavement surface, and it is difficult to determine its clear boundary, which brings certain difficulties to the measurement of the loose area. On the other hand, there may be a situation where coarse and fine aggregates are missing at the same time in the same loose area, and it is difficult to distinguish the aggregate size state before the missing area. This brings certain difficulties to the evaluation of the looseness of the pavement. In addition, no research has shown that there are obvious differences in the mechanism and difficulty of the peeling behavior of coarse and fine aggregates.

[0004] Therefore, the traditional research on the evaluation of aggregate spalling and pavement loosening of asphalt pavements is highly subjective, the meaning of indicators is vague, and there is a lack of sufficient understanding of pavement loosening and aggregate spalling, which increases the difficulty of evaluating pavement aggregate spalling and measuring loosening trends. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a method for evaluating the spalling of asphalt pavement aggregate and measuring the loosening tendency with higher objectivity and accuracy.

[0006] To achieve the above object, the present invention proposes a method for evaluating the spalling of asphalt pavement aggregate and measuring its loosening tendency, comprising the following steps:

[0007] S1: Acquire three-dimensional road surface data and perform preprocessing to obtain precise three-dimensional road surface data;

[0008] S2: Identify and segment the aggregate area and non-aggregate area in the 3D precise data of the pavement, and determine the aggregate spalling condition;

[0009] S3: According to the aggregate spalling condition, establish the looseness evaluation index of asphalt pavement and evaluate the looseness condition of asphalt pavement aggregate;

[0010] S4: Based on the three-dimensional precise data of the pavement, a three-dimensional analysis model of the asphalt pavement is established, and the loosening and peeling condition of aggregates of the three-dimensional analysis model of the asphalt pavement under the action of vehicle load is simulated to estimate the deterioration trend of the asphalt pavement.

[0011] Furthermore, in step S1, three-dimensional road surface data is acquired by scanning, and the three-dimensional road surface data includes an x-coordinate, a y-coordinate, a z-coordinate and a reflection intensity of a scanned point on the road surface.

[0012] Furthermore, in step S1, the preprocessing method is: based on the 3σ rule of statistical distribution, for the z coordinate and reflection intensity, the data exceeding 3σ are regarded as abnormal points and removed, and linear interpolation is used to replace them to obtain three-dimensional precise road surface data.

[0013] Furthermore, in step S2, a point cloud segmentation algorithm is used to identify and segment aggregate areas and non-aggregate areas in the three-dimensional precise data of the road surface. The point cloud segmentation algorithm includes a watershed algorithm, an edge segmentation algorithm, a region growing algorithm, a model fitting algorithm, a point cloud clustering algorithm or a deep learning method.

[0014] Furthermore, in step S2, the non-aggregate area is divided into: an aggregate spalling area, an aggregate missing area, and an algorithm missed identification area.

[0015] The aggregate spalling area refers to the non-aggregate area formed by the asphalt being peeled off from the aggregate surface under the action of rain and load, causing the aggregate to become loose or even carried away by the rolling wheels; the aggregate missing area refers to the non-aggregate area where the pavement aggregate itself does not exist due to the uneven construction; the algorithm missed identification area refers to the area where aggregate exists but is identified as non-aggregate by the model due to the limited accuracy of the algorithm.

[0016] Furthermore, in step S2, the method for determining the aggregate spalling condition is to identify the aggregate spalling area from the non-aggregate area, and the identification method is to use SVM support vector machine for identification, and the input is the z coordinate and reflection intensity of the three-dimensional precise data of the pavement, and the output is 0 or 1 or 2; wherein 0 indicates that the non-aggregate area is an aggregate spalling area; 1 indicates that the non-aggregate area is an aggregate missing area; and 2 indicates that the non-aggregate area is an area that is missed by the algorithm.

[0017] Furthermore, the z coordinate and reflection intensity of the three-dimensional precise data of the road surface are uniformly input into the SVM support vector machine after being encoded, and the encoding method is encoded by frequency distribution. For example, the z coordinate of the three-dimensional precise data of the non-aggregate area is obtained, and the frequency distribution of the z coordinate is statistically calculated. The number of distribution groups should be no less than 10 groups and no more than 100 groups. The frequency of the z coordinate of each group is input into the SVM support vector machine as the code of the z coordinate.

[0018] Furthermore, the z coordinate and reflection intensity index are used for splicing because the z coordinate distribution and reflection intensity index distribution in the aggregate spalling area are significantly different from those in other non-aggregate areas, which are specifically reflected in the following aspects: the average value of the z coordinate in the aggregate spalling area is low, and the bottom surface of the local missing area formed by aggregate spalling is relatively flat, and the distribution peak of its z coordinate is not prominent, presenting a flat peak shape; the reflection intensity index distribution in the aggregate spalling area presents obvious polarization characteristics. In the aggregate spalling area where cohesive failure occurs, the surface of the spalling area is covered with an asphalt film with low reflection intensity. In the aggregate spalling area where adhesion failure occurs, the surface of the spalling area exposes the aggregate surface, and the reflection intensity is high at this time, which is close to the reflection intensity of the rock itself.

[0019] Furthermore, in step S3, the calculation formula of the asphalt pavement looseness evaluation index (RD, Rave ling Degree) is:

[0020]

[0021] Among them, S l It refers to the asphalt pavement spalling area caused by aggregate loss and spalling, which can be calculated by the number of pixels in the aggregate spalling area in the aggregate spalling identification result; S o Refers to the overall area of ​​the road surface, which can be calculated by counting the number of pixels in the area.

[0022] Furthermore, in step S4, the simulation includes an asphalt pavement simulation model, a vehicle load simulation model, and an interaction simulation. The asphalt pavement simulation model is a pavement simulation model with physical properties established based on precise three-dimensional pavement data; the vehicle load simulation model is a vehicle standard axle load set according to standard specifications, including static axle load and dynamic load; the interaction simulation is a force simulation of the asphalt pavement simulation model and the vehicle load simulation model in series.

[0023] Furthermore, in step S4, the specific method of estimating the deterioration trend of the asphalt pavement is: using simulation to obtain the change of the looseness RD of the asphalt pavement under the vehicle load over time.

[0024] Compared with the prior art, the advantages of the present invention are:

[0025] The present invention uses high-precision laser scanning equipment to obtain precise three-dimensional data of the road surface, and combines the point cloud segmentation algorithm to automatically identify the spalling condition of the road surface aggregate, so as to achieve the purpose of evaluating the spalling of the asphalt pavement aggregate; on this basis, an evaluation index for the looseness of the asphalt pavement is established to evaluate the looseness and spalling condition of the aggregate, and then the simulation software is used to simulate the decline in the service level of the asphalt pavement, effectively measuring the development trend of the looseness of the road surface, which is suitable for the current asphalt pavement aggregate spalling evaluation and looseness trend measurement application, and has higher objectivity and accuracy, can effectively improve the accuracy of the spalling evaluation, and more accurately obtain the performance status of the asphalt pavement. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic flow chart of the method for evaluating the spalling of asphalt pavement aggregate and measuring the loosening tendency of the present invention;

[0027] Figure 2 This is a graph showing the aggregate identification and segmentation results in an embodiment of the present invention;

[0028] Figure 3 This is a schematic diagram of the non-aggregate area in an embodiment of the present invention;

[0029] Figure 4 Schematic diagram of aggregate missing distribution of loose damaged pavement in an embodiment of the present invention;

[0030] Figure 5 This is a diagram of an aggregate spalling region identification algorithm based on a support vector machine (SVM) in an embodiment of the present invention;

[0031] Figure 6 This is a diagram showing the result of identifying the aggregate spalling area in an embodiment of the present invention;

[0032] Figure 7 A discrete element simulation model diagram of an asphalt pavement in an embodiment of the present invention;

[0033] Figure 8 A diagram of a vehicle load simulation model in an embodiment of the present invention;

[0034] Fig. 9 This is a simulation result diagram of the loose damage and deterioration trend of the asphalt pavement in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be further described below.

[0036] like Figure 1As shown, the present invention proposes a method for evaluating the spalling of aggregates and measuring the loose trend of asphalt pavement. The method is mainly divided into data collection and preprocessing, point cloud segmentation and aggregate identification, spalling judgment and loose evaluation, and simulation loose trend measurement. The specific steps are as follows:

[0037] Step 1: Data collection and preprocessing

[0038] 1.1: First, select landmark asphalt pavement points for data collection (it is advisable to select loose areas of the pavement where aggregate peeling is obvious and visible to the naked eye); use high-precision 3D scanning equipment to scan the pavement. The vertical resolution of the scanning equipment should be no less than 0.01mm, the horizontal and vertical collection intervals should be no higher than 0.05mm, and the area scanned at one time should be no less than 10cm×10cm. The data obtained by the scan should at least include the x-coordinate, y-coordinate, z-coordinate of the pavement point and the reflection intensity of the scan.

[0039] 1.2: Preprocess the 3D road surface data obtained by scanning to remove abnormal data. Specifically, based on the 3σ rule of statistical distribution, the data exceeding 3σ are regarded as abnormal points for the z coordinate and reflection intensity respectively, and linear interpolation is used to obtain precise 3D road surface data.

[0040] Step 2: Point cloud segmentation and aggregate identification

[0041] According to the obtained three-dimensional precise data of the road surface, the point cloud data belonging to the aggregate area and the point cloud data belonging to the non-aggregate area are identified; the watershed algorithm is used to segment the three-dimensional precise data of the road surface (the watershed algorithm is to obtain the contours of different objects by imitating the terrain such as mountains, gullies, basins, etc. in the geographical structure. It is a segmentation algorithm based on geographical morphology. It compares the input three-dimensional data to geographical space, and regards the local minimum and its affected area as a watershed. The boundaries of different watersheds, that is, the contours of the target, constitute the watershed). In the segmentation process, the lowest point of the algorithm is the starting point of the threshold, and the watershed area is gradually covered. As the threshold gradually increases, the two adjacent but isolated watersheds tend to be connected. At this time, the pixels of the confluence point are connected to form a watershed. After the three-dimensional precise data of the road surface is deeply inverted, the aggregate area is turned into a valley, and the aggregate boundary area is turned into a peak, thereby completing the watershed segmentation. After the segmentation is completed, the aggregate area is identified, such as Figure 2 As shown in Figure 1, (a) is the aggregate area identification mask map, and (b) is the aggregate area identification overlay map. On this basis, the remaining part of the aggregate area is taken as the non-aggregate area, as shown in Figure 1. Figure 3 As shown, (a) is the non-aggregate area identification map, and (b) is the non-aggregate area identification map after superimposed display.

[0042] Step 3: Delamination and looseness assessment

[0043] Schematic diagram of loose road damage Figure 4 As shown in the figure, it can be seen that the looseness of the road surface is mainly manifested in the spalling and missing of aggregates. Therefore, the identification of aggregate spalling is the basis for judging the looseness of the road surface and evaluating its degree. Therefore, the problem can be described as identifying the aggregate spalling area from the non-aggregate area. However, from the results of aggregate segmentation and identification, there are mainly three situations in the non-aggregate area: 1) Aggregate spalling area: Aggregate spalling occurs, resulting in the missing of aggregates at local sites and unidentified; 2) Aggregate missing area: During the road construction process, the uneven distribution of aggregates leads to the fact that there is no aggregate at this site and it is not identified; 3) Algorithm missed identification area: The performance of the segmentation model is limited, and some aggregates are missed.

[0044] For the above three types of situations, by analyzing the causes and combining on-site observations, we summarized their differentiated characteristics in the laser point cloud data. The results showed that the z coordinate distribution and reflection intensity intensity distribution of the aggregate spalling area were significantly different from those of other non-aggregate areas. Specifically, the average value of the z coordinate of the aggregate spalling area was low, and the bottom surface of the local missing area formed by aggregate spalling was relatively flat, and the distribution peak of its z coordinate was not prominent, showing a flat peak shape. In addition, the reflection intensity intensity distribution of the aggregate spalling area showed obvious polarization characteristics compared with other non-aggregate areas. In the aggregate spalling area where cohesive failure occurred, the surface of the spalling area was covered with an asphalt film with low reflection intensity. In the aggregate spalling area where adhesion failure occurred, the surface of the spalling area exposed the aggregate surface, and the reflection intensity was high at this time, close to the reflection intensity of the rock itself.

[0045] On this basis, a classification and recognition model for non-aggregate areas is established to identify aggregate spalling areas in non-aggregate areas. The algorithm model is as follows: Figure 5 As shown in the figure: 1) First, the z coordinate and reflection intensity of the non-aggregate area in all the three-dimensional precise data of the road surface are extracted; 2) The distribution of the z coordinate and reflection intensity are counted respectively. For the z coordinate, it is divided into 50 groups from 0.25 to 0.5, and the frequency of each group is counted respectively as the z coordinate code of the non-aggregate area; for the reflection intensity intensity, it is divided into 50 groups from 0 to 3000, and the frequency of each group is counted as the reflection intensity intensity code of the non-aggregate area; 3) After splicing the two groups of codes, a 1×100 vector is formed, which is input into the SVM support vector machine, and the support vector machine SVM outputs a 1×3 vector, which indicates that the non-aggregate area belongs to 0—aggregate peeling area, 1—aggregate missing area, and 2—algorithm missed recognition area. The recognition results are shown in Figure 6As shown, (a) is a schematic diagram of the three-dimensional accuracy data of the pavement, (b) is the identified non-aggregate area, and (c) is the identified aggregate spalling area.

[0046] Finally, the area S of the aggregate spalling area is calculated separately. l and the area of ​​the scanned area S o The ratio of the two is used as the evaluation index RD of the looseness of asphalt pavement. The expression formula is:

[0047]

[0048] In the formula, S l It refers to the asphalt pavement spalling area caused by aggregate loss and spalling, which can be calculated by the number of pixels in the aggregate spalling area in the aggregate spalling identification result; S o Refers to the total area of ​​the road surface, which can be calculated by the total number of pixels in the scanned area.

[0049] Step 4: Simulate the loose trend measure

[0050] After identifying the aggregate spalling area and calculating the evaluation index of the looseness of the pavement, the looseness of the current asphalt pavement can be understood. By further simulating the aging process of the asphalt pavement, the deterioration trend of the loose damage of the asphalt pavement can be obtained, providing a theoretical basis for road maintenance and management. The details are as follows:

[0051] 4.1: First, based on the three-dimensional precise data of the pavement and the identified aggregate area, a discrete element simulation model of the asphalt pavement is established, such as Figure 7 As shown in the figure, at the same time, the vehicle load simulation model is established as Figure 8 As shown, in addition, when establishing the interaction model, the main considerations are 1) dynamically attenuating the bonding strength of the asphalt elements while applying the moving load to simulate the impact of the dynamic water pressure on the asphalt film and the aging effect of the asphalt; 2) simplifying the definition conditions for the spalling of the surface aggregate elements, and simplifying all the contact fractures of the original elements and the substantial displacement from the road surface to only judging whether the contact relationship is broken. Because under the action of a simple load, the asphalt and aggregate elements are less subject to upward forces, even if the elements have broken in contact, they may not necessarily have substantial displacement behavior from the road surface. Therefore, this simplification of the definition can more realistically measure the proportion of aggregate spalling. On this basis, the discrete element software was used for simulation and the degree of looseness RD of the asphalt pavement was evaluated, and the results were obtained as shown below Fig. 9 shown.

[0052] According to the loose rate change diagram, it can be seen that the early loose damage of asphalt pavement mainly goes through three stages: early loose formation, middle loose aggravation, and late loose stability. In the early loose formation, the looseness of the pavement is low and its deterioration trend is also slow. At this time, the overall level of aggregate peeling rate is low, but the fluctuation is large; when the aggregate peeling ratio in the asphalt mixture reaches a certain level, the pavement enters the middle loose aggravation period. At this stage, the overall contact force network of the surface elements undergoes obvious structural changes and loses its original stable state. At this time, the deterioration trend of the loose pavement and the peeling rate of the aggregate accelerate, and the pavement undergoes rapid loose aggravation; when the looseness of the pavement reaches a higher level, the pavement enters the late loose stability period. At this time, the local contact force network structure is destroyed, and only local areas exist to resist traffic loads and dynamic water pressure alone. However, the remaining unpeeled areas are usually areas with strong adhesion and strong anti-peeling ability. At this time, the aggregate peeling rate decreases significantly, and enters the loose stability period.

[0053] Implementation Cases

[0054] This embodiment proposes a method for evaluating the spalling of asphalt pavement aggregate and measuring the loosening tendency, which is as follows:

[0055] (1) Scan the road surface to obtain three-dimensional road surface information and preprocess it

[0056] In this embodiment, a road surface texture laser scanner is used to scan the road surface. The vertical resolution of the scanning instrument is 0.01 mm, the lateral collection interval is 0.0415 mm, the longitudinal collection interval is 0.0496 mm, and the area scanned at one time is 10 cm × 10 cm. The scanning obtains three-dimensional information, including the x-coordinate, y-coordinate, and z-coordinate of the road surface point and the reflection intensity information of the scan.

[0057] The preprocessing adopts the 3σ rule based on statistical distribution. The data exceeding 3σ in the z-coordinate and reflection intensity information are regarded as abnormal points and removed, and linear interpolation is used to replace them to obtain the three-dimensional precise data of the road surface.

[0058] (2) Aggregate identification and spalling judgment:

[0059] Based on the pre-processed 3D precise data of the pavement, the watershed algorithm is used to identify the aggregate area, and the remaining part of the aggregate area is taken as the non-aggregate area. The z-coordinate distribution and reflection intensity index distribution of each aggregate area are statistically analyzed, and the support vector machine (SVM) is used to identify the part of the non-aggregate area that belongs to the aggregate spalling area.

[0060] In this example, a total of 29 10cm×10cm asphalt pavement points were collected, and 369 non-aggregate areas were obtained. Manual annotation was used to confirm that 196 of the non-aggregate areas were aggregate spalling areas, 160 were aggregate missing areas, and 40 were algorithm missed areas. The results of using SVM to identify aggregate spalling areas in non-aggregate areas are as follows:

[0061] Table 1

[0062] Number of samples The true value is 0 The true value is 1 The true value is 2 The predicted value is 0 183 1 0 184 The predicted value is 1 5 157 2 164 The predicted value is 2 8 2 38 48 196 160 40 396

[0063] Table 1 above shows the result of the support vector machine (SVM) identifying the non-aggregate area. It can be seen from Table 1 that the recognition accuracy of the support vector machine (SVM) model is 95.45%.

[0064] (3) Looseness assessment

[0065] For each scanned 10cm×10cm area, the size of the aggregate spalling area is counted and the looseness degree RD is calculated. The calculation formula is:

[0066]

[0067] From the above calculation results, it can be seen that the looseness RD of the scanned road surface area is 9.75%.

[0068] (4) Simulation of loosening trend measurement: Using discrete element simulation software, a three-dimensional analysis model of asphalt pavement is established, and the loosening and peeling of aggregates in the three-dimensional analysis model of asphalt pavement under vehicle load is simulated to estimate the deterioration trend of asphalt pavement.

[0069] The three-dimensional analysis model of asphalt pavement established by discrete element simulation software is as follows: Figure 7 As shown in Figure 2, the established vehicle load simulation model is as follows: Figure 8 shown.

[0070] Among them, the corresponding material parameters of the established pavement model are given to the model according to the measured data, see Table 2 below:

[0071] Table 2

[0072] Structural layer Macro modulus (MPa) Contact stiffness (N / m) Asphalt stabilized macadam 1200 <![CDATA[2.4×10 11 ]]> Graded crushed stone 400 <![CDATA[8×10 6 ]]> Lime soil 400 <![CDATA[8×10 6 ]]> Soil foundation 50 <![CDATA[2×10 6 ]]>

[0073] In addition, in Figure 8 In the simulation model of the vehicle load shown, the design axle load and wheel parameters are shown in Tables 3 and 4 below:

[0074] Table 3

[0075]

[0076] Table 4

[0077] Tire weight (kg) Tread width(m) Tire diameter(m) Double wheel center distance (m) 50 0.28 1.085 0.3195

[0078] In this embodiment, the loose damage deterioration trend of the asphalt pavement obtained by simulation is as follows: Fig. 9 shown.

[0079] The above is only a preferred embodiment of the present invention and does not limit the present invention in any way. Any technician in the relevant technical field, without departing from the scope of the technical solution of the present invention, makes any form of equivalent replacement or modification to the technical solution and technical content disclosed in the present invention, which does not depart from the content of the technical solution of the present invention and still falls within the protection scope of the present invention.

Claims

1. A method for evaluating the spalling of asphalt pavement aggregate and measuring its loosening tendency, characterized in that: The steps include: S1: Acquire three-dimensional road surface data and perform preprocessing to obtain precise three-dimensional road surface data; S2: Identify and segment the aggregate area and non-aggregate area in the 3D precise data of the pavement, and determine the aggregate spalling condition; S3: According to the aggregate spalling condition, establish the looseness evaluation index of asphalt pavement and evaluate the looseness condition of asphalt pavement aggregate; S4: Based on the three-dimensional precise data of the pavement, a three-dimensional analysis model of the asphalt pavement is established, and the loosening and peeling condition of aggregates of the three-dimensional analysis model of the asphalt pavement under the action of vehicle load is simulated to estimate the deterioration trend of the asphalt pavement.

2. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 1 is characterized in that: In step S1, three-dimensional road surface data is acquired by scanning, and the three-dimensional road surface data includes an x-coordinate, a y-coordinate, a z-coordinate and a scanned reflection intensity of a road surface point.

3. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 2 is characterized in that: In step S1, the preprocessing method is: based on the 3σ rule of statistical distribution, for the z coordinate and reflection intensity, the data exceeding 3σ are regarded as abnormal points and removed, and linear interpolation is used to replace them to obtain three-dimensional precise road surface data.

4. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 1 is characterized in that: In step S2, a point cloud segmentation algorithm is used to identify and segment the aggregate area and non-aggregate area in the three-dimensional precise data of the road surface. The point cloud segmentation algorithm includes a watershed algorithm, an edge segmentation algorithm, a region growing algorithm, a model fitting algorithm, a point cloud clustering algorithm or a deep learning method.

5. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 1 is characterized in that: In step S2, the non-aggregate area is divided into: an aggregate spalling area, an aggregate missing area, and an algorithm missed identification area.

6. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 5 is characterized in that: In step S2, the method for determining the aggregate spalling condition is to identify the aggregate spalling area from the non-aggregate area. The identification method is to use SVM support vector machine for identification. The input is the z coordinate and reflection intensity intensity of the three-dimensional precise data of the pavement, and the output is 0 or 1 or 2; wherein 0 indicates that the non-aggregate area is an aggregate spalling area; 1 indicates that the non-aggregate area is an aggregate missing area; and 2 indicates that the non-aggregate area is an area that is missed by the algorithm.

7. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 6, characterized in that: The z coordinate and reflection intensity of the three-dimensional precise data of the road surface are uniformly input into the SVM support vector machine after being encoded, and the encoding method is encoded through frequency distribution.

8. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 1 is characterized in that: In step S3, the calculation formula of the asphalt pavement looseness evaluation index is: Among them, S l It refers to the asphalt pavement spalling area caused by aggregate loss and spalling, which can be calculated by the number of pixels in the aggregate spalling area in the aggregate spalling identification result; S o Refers to the overall area of ​​the road surface, which can be calculated by counting the number of pixels in the area.

9. The method for evaluating the spalling and loosening tendency of asphalt pavement aggregate according to claim 1, characterized in that: In step S4, the simulation includes an asphalt pavement simulation model, a vehicle load simulation model and an interaction simulation.

10. The method for evaluating the spalling of asphalt pavement aggregate and measuring the loosening tendency according to claim 1, characterized in that: In step S4, the specific method of estimating the deterioration trend of the asphalt pavement is: using simulation to obtain the change of the looseness RD of the asphalt pavement under the vehicle load over time.

Images