A method, system and equipment for real-time evaluation of the surface performance of airport pavement concrete

Abstract

This application discloses a method, system, and device for real-time evaluation of the surface performance of airport pavement concrete. The method calculates the normal vector and curvature of each 3D point cloud data set to be predicted; clusters multiple 3D point cloud data sets and calculates the area of ​​the connected regions corresponding to each cluster; based on the normal vectors, curvatures, and connected regions corresponding to each cluster, a first target prediction model is used to predict the first probability value of defects existing on the airport pavement concrete surface; multiple images to be predicted are input into a second target prediction model for defect prediction, obtaining a second probability value and center location of the predicted defects on the airport pavement concrete surface; based on the center location, the first probability value, and the second probability value, the final defect is determined; and based on the final defect, the surface performance of the airport pavement concrete is analyzed. This application can improve the accuracy of predicting defects on the airport pavement concrete surface.

Description

Technical Field

[0001] This application relates to the field of concrete performance evaluation technology, and in particular to a method, system and equipment for real-time evaluation of the surface performance of airport pavement concrete. Background Technology

[0002] Over long-term use, airport pavement concrete structures are affected by environmental factors, loads, and material aging, leading to a decline in surface performance, such as cracks, spalling, and peeling. These problems not only affect the service life of the pavement but may also jeopardize flight safety. Therefore, developing an intelligent curing and real-time surface performance evaluation method for airport pavement concrete is crucial for its maintenance and management.

[0003] Traditional surface performance testing of airport pavement concrete relies on manual inspections. This method depends on the experience of workers to identify defects such as cracks, spalling, and peeling. It is subjective, inefficient, and prone to missing defects. To address these issues, researchers have developed intelligent surface performance testing methods for airport pavement concrete, such as those using large amounts of image data to train detection models. However, these methods suffer from poor stability and accuracy in complex environments (such as uneven lighting, water accumulation, and low nighttime illumination), making it difficult to accurately classify and identify various defects (such as cracks, spalling, and peeling). This leads to frequent misjudgments and missed detections, resulting in untimely and inaccurate discovery and treatment of airport pavement defects, increasing safety hazards. Summary of the Invention

[0004] This application aims to propose a method, system, and equipment for real-time evaluation of the surface performance of airport pavement concrete, which can improve the accuracy of predicting surface defects of airport pavement concrete, thereby improving the accuracy of evaluating the surface performance of airport pavement concrete and enhancing the safety of airport pavement.

[0005] In a first aspect, embodiments of this application provide a method for real-time evaluation of the surface performance of airport pavement concrete, the method comprising:

[0006] Acquire multiple 3D point cloud data and multiple images of the concrete surface of the airport pavement to be predicted;

[0007] Calculate the normal vector and curvature of each of the three-dimensional point cloud data to be predicted;

[0008] Cluster the multiple 3D point cloud data to be predicted to obtain multiple cluster categories, and calculate the area of ​​the connected region corresponding to each cluster category;

[0009] Based on the normal vectors, curvatures, and connected region areas corresponding to each cluster category of the multiple three-dimensional point cloud data to be predicted, the first probability value of defects existing on the concrete surface of the airport pavement is predicted using the first target prediction model.

[0010] The multiple images to be predicted are input into the second target prediction model to predict defects, thereby obtaining the second probability value and center location of the defects existing on the concrete surface of the airport pavement.

[0011] The final defect is determined based on the center location of the defect, the first probability value, and the second probability value;

[0012] Based on the aforementioned final defects, the surface properties of airport pavement concrete were analyzed.

[0013] Compared with the prior art, the first aspect of this application has the following beneficial effects:

[0014] This method acquires multiple 3D point cloud data and multiple images of the airport pavement concrete surface to be predicted; calculates the normal vector and curvature of each 3D point cloud data; clusters the multiple 3D point cloud data to be predicted to obtain multiple cluster categories, and calculates the area of ​​the connected region corresponding to each cluster category; based on the normal vector, curvature, and area of ​​the connected region corresponding to each cluster category, a first target prediction model is used to predict the first probability value of defects existing on the airport pavement concrete surface; multiple images are input into a second target prediction model for defect prediction, obtaining the second probability value and center position of the predicted defects on the airport pavement concrete surface; based on the center position of the defect, the first probability value, and the second probability value, the final defect is determined; and based on the final defect, the surface performance of the airport pavement concrete is analyzed. Therefore, by combining the normal vector, curvature, and area of ​​the connected region to predict defects on the airport pavement concrete surface, the accuracy of defect prediction can be improved. Then, by comprehensively considering the prediction results obtained from 3D point cloud data and image-based prediction results, the final defect is determined, further improving the accuracy of defect prediction. This, in turn, enhances the accuracy of airport pavement concrete surface performance evaluation and improves airport pavement safety. Furthermore, both 3D point cloud data and image data can be acquired in real time, enabling real-time evaluation of airport pavement concrete surface performance.

[0015] In some implementations, calculating the normal vector and curvature of each of the three-dimensional point cloud data to be predicted includes:

[0016] Calculate the normal vector for each of the three-dimensional point cloud data to be predicted:

[0017]

[0018] Calculate the curvature of each of the three-dimensional point cloud data to be predicted:

[0019]

[0020] Among them, D i Represents each 3D point cloud data d i Let P be the set of neighborhood points d. j The covariance matrix, where T denotes the transpose. Represents the neighborhood point d j The centroid of the point cloud is given by γ1, γ2 and γ3, where γ1 ≤ γ2 ≤ γ3 and F represents the curvature of the three-dimensional point cloud data.

[0021] In some implementations, calculating the area of ​​the connected regions corresponding to each cluster category includes:

[0022] Convert the 3D point cloud data in each cluster category into a mesh and construct a mesh model;

[0023] Calculate the sum of the areas of all grids in the grid model to obtain the area of ​​the connected region corresponding to each cluster category.

[0024] In some implementations, the step of predicting a first probability value of defects on the concrete surface of the airport pavement using a first target prediction model based on the normal vectors, curvatures, and connected region areas corresponding to the plurality of three-dimensional point cloud data to be predicted, and each cluster category, includes:

[0025] Compare the size of the connected region area corresponding to each cluster category, remove the 3D point cloud data of the cluster category with the largest connected region area, and obtain the 3D point cloud data of the remaining cluster categories.

[0026] The normal vector, curvature, and area of ​​the connected region corresponding to the three-dimensional point cloud data of each of the remaining cluster categories are input into the first target prediction model to predict the category and first probability value of each of the multiple defects corresponding to each of the remaining cluster categories. The defects are defects existing on the concrete surface of the airport pavement.

[0027] In some implementations, determining the final defect based on the center location of the defect, the first probability value, and the second probability value includes:

[0028] Calculate the center point cloud coordinates of the 3D point cloud data for each of the remaining cluster categories;

[0029] Associate the center point cloud coordinates with the respective categories of the multiple defects;

[0030] Determine the matching degree between the center point cloud coordinates of the three-dimensional point cloud data of each of the remaining cluster categories and the center position of the defect, and determine the final defect based on the matching degree, the first probability value and the second probability value.

[0031] In some implementations, determining the matching degree between the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories and the center position of the defect, and determining the final defect based on the matching degree, the first probability value, and the second probability value, includes:

[0032] If the center point cloud coordinates of the three-dimensional point cloud data of each of the remaining cluster categories match the center position of the defect, then the first probability value of each defect corresponding to the center point cloud coordinates and the second probability value of the defect corresponding to the center position are added together to obtain the total probability value of each defect, wherein the defect corresponding to the first probability value and the defect corresponding to the second probability value are the same.

[0033] The defect with the highest total probability value is taken as the final defect.

[0034] In some implementations, determining the matching degree between the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories and the center position of the defect, and determining the final defect based on the matching degree, the first probability value, and the second probability value, includes:

[0035] If the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories do not match the center position of the defect, then the defect with the highest probability corresponding to the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories is taken as the final defect, or the defect corresponding to the center position is taken as the final defect.

[0036] Secondly, embodiments of this application also provide a real-time evaluation system for the surface performance of airport pavement concrete, the system comprising:

[0037] The data acquisition unit is used to acquire multiple three-dimensional point cloud data and multiple images of the concrete surface of the airport pavement to be predicted.

[0038] The first computing unit is used to calculate the normal vector and curvature of each of the three-dimensional point cloud data to be predicted;

[0039] The second calculation unit is used to cluster the multiple 3D point cloud data to be predicted, obtain multiple cluster categories, and calculate the area of ​​the connected region corresponding to each cluster category.

[0040] The first prediction unit is used to predict the first probability value of defects existing on the concrete surface of the airport pavement based on the normal vector, curvature and the area of ​​the connected region corresponding to each cluster category of the multiple three-dimensional point cloud data to be predicted, using the first target prediction model.

[0041] The second prediction unit is used to input the multiple images to be predicted into the second target prediction model to predict defects, and obtain the second probability value and center position of the defects existing on the concrete surface of the airport pavement.

[0042] A defect determination unit is used to determine the final defect based on the center position of the defect, the first probability value, and the second probability value;

[0043] The performance analysis unit is used to analyze the surface properties of airport pavement concrete based on the final defects.

[0044] Thirdly, embodiments of this application also provide an electronic device, including at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which, when executed by the at least one control processor, enable the at least one control processor to perform a real-time evaluation method for the surface performance of airport pavement concrete as described above.

[0045] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer-executable instructions for causing a computer to execute a real-time evaluation method for the surface performance of airport pavement concrete as described above.

[0046] It is understood that the beneficial effects of the second to fourth aspects compared with the related technologies are the same as the beneficial effects of the first aspect compared with the related technologies. Please refer to the relevant description in the first aspect above, which will not be repeated here. Attached Figure Description

[0047] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0048] Figure 1 This is a flowchart illustrating an embodiment of the real-time evaluation method for the surface performance of airport pavement concrete provided in this application.

[0049] Figure 2 This is a schematic diagram of an embodiment of the real-time evaluation system for the surface performance of airport pavement concrete provided in this application;

[0050] Figure 3 This is a schematic diagram of the structure of an embodiment of the electronic device provided in this application. Detailed Implementation

[0051] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0052] In the description of this application, the use of terms such as "first," "second," etc., is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0053] In the description of this application, it should be understood that the orientation descriptions, such as up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0054] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0055] Existing intelligent surface performance testing methods for airport pavement concrete, such as those using large amounts of image data to train detection models, suffer from poor stability and accuracy under complex environments (e.g., uneven lighting, water accumulation, and low nighttime illumination). These methods struggle to accurately classify and identify various defects, such as cracks, spalling, and peeling on the concrete surface. This leads to frequent misjudgments and missed detections, resulting in the inability to promptly and accurately identify and address airport pavement defects, thus increasing safety hazards.

[0056] To address the problem that existing technologies struggle to accurately classify and identify various defects, leading to frequent misjudgments and omissions that increase safety risks, this application proposes a method, system, and equipment for real-time evaluation of the surface performance of airport pavement concrete.

[0057] Reference Figure 1 This application provides a schematic flowchart of a method for real-time evaluation of the surface performance of airport pavement concrete. This method is applied to an electronic device, which may be a server or a mobile terminal, etc. Figure 1As shown, the real-time evaluation method for the surface performance of airport pavement concrete may include the following steps:

[0058] Step S100: Obtain multiple three-dimensional point cloud data and multiple images of the concrete surface of the airport pavement to be predicted.

[0059] Step S200: Calculate the normal vector and curvature of each 3D point cloud data to be predicted;

[0060] Step S300: Cluster multiple 3D point cloud data to be predicted to obtain multiple cluster categories, and calculate the area of ​​the connected region corresponding to each cluster category;

[0061] Step S400: Based on the normal vectors, curvatures, and connected region areas corresponding to each cluster category of multiple three-dimensional point cloud data to be predicted, the first probability value of defects existing on the concrete surface of the airport pavement is predicted using the first target prediction model.

[0062] Step S500: Input multiple images to be predicted into the second target prediction model to predict defects, and obtain the second probability value and center position of the defects existing on the concrete surface of the airport pavement.

[0063] Step S600: Based on the center location of the defect, the first probability value, and the second probability value, determine the final defect;

[0064] Step S700: Analyze the surface properties of airport pavement concrete based on the final defects.

[0065] In this embodiment, multiple three-dimensional point cloud data and multiple images of the airport pavement concrete surface to be predicted are acquired; the normal vector and curvature of each three-dimensional point cloud data to be predicted are calculated; the multiple three-dimensional point cloud data to be predicted are clustered to obtain multiple cluster categories, and the area of ​​the connected region corresponding to each cluster category is calculated; based on the normal vector, curvature and the area of ​​the connected region corresponding to each cluster category, a first target prediction model is used to predict the first probability value of defects existing on the airport pavement concrete surface; the multiple images to be predicted are input into a second target prediction model to predict defects, and a second probability value and center position of the predicted defects existing on the airport pavement concrete surface are obtained; based on the center position of the defect, the first probability value and the second probability value, the final defect is determined; and the surface performance of the airport pavement concrete is analyzed based on the final defect. Therefore, by combining the normal vector, curvature, and area of ​​the connected region to predict defects on the airport pavement concrete surface, the accuracy of defect prediction can be improved. Then, by comprehensively considering the prediction results obtained from 3D point cloud data and image-based prediction results, the final defect is determined, further improving the accuracy of defect prediction. This, in turn, enhances the accuracy of airport pavement concrete surface performance evaluation and improves airport pavement safety. Furthermore, both 3D point cloud data and image data can be acquired in real time, enabling real-time evaluation of airport pavement concrete surface performance.

[0066] In some implementations, calculating the normal vector and curvature of each 3D point cloud data to be predicted includes:

[0067] Calculate the normal vector for each 3D point cloud data to be predicted:

[0068]

[0069] Calculate the curvature of each 3D point cloud data to be predicted:

[0070]

[0071] Among them, D i Represents each 3D point cloud data d i Let P be the set of neighborhood points d. j The covariance matrix, where T denotes the transpose. Represents the neighborhood point d j The centroid of the point cloud is given by γ1, γ2 and γ3, where γ1 ≤ γ2 ≤ γ3 and F represents the curvature of the three-dimensional point cloud data.

[0072] In this embodiment, by accurately calculating the normal vector and curvature of each 3D point cloud data to be predicted, the accuracy of subsequent defect prediction can be improved.

[0073] In some implementations, calculating the area of ​​the connected regions corresponding to each cluster category includes:

[0074] Convert the 3D point cloud data in each cluster category into a mesh and construct a mesh model;

[0075] Calculate the sum of the areas of all grids in the grid model to obtain the area of ​​the connected region corresponding to each cluster category.

[0076] In this embodiment, a grid model is constructed by converting the 3D point cloud data in each cluster category into a mesh; the area of ​​all grids in the grid model is calculated to obtain the area of ​​the connected regions corresponding to each cluster category. This allows for the accurate determination of the connected region area for each cluster category, laying a solid data foundation for accurate defect prediction in the later stages.

[0077] In some implementations, based on the normal vectors, curvatures, and connected region areas corresponding to multiple 3D point cloud data to be predicted, a first target prediction model is used to predict the first probability value of defects existing on the concrete surface of the airport pavement, including:

[0078] Compare the size of the connected region area corresponding to each cluster category, remove the 3D point cloud data of the cluster category with the largest connected region area, and obtain the 3D point cloud data of the remaining cluster categories.

[0079] The normal vector, curvature, and area of ​​the connected region corresponding to the 3D point cloud data of each of the remaining cluster categories are input into the first target prediction model to predict the category and first probability value of each of the multiple defects corresponding to each of the remaining cluster categories. The defects are defects existing on the concrete surface of the airport pavement.

[0080] In this embodiment, by combining the normal vector, curvature, and area of ​​the connected region to predict defects on the concrete surface of the airport pavement, the accuracy of defect prediction can be improved.

[0081] In some implementations, the final defect is determined based on the center location of the defect, a first probability value, and a second probability value, including:

[0082] Calculate the center point cloud coordinates of the 3D point cloud data for each of the remaining cluster categories;

[0083] Associate the center point cloud coordinates with the respective categories of multiple defects;

[0084] Determine the matching degree between the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories and the center location of the defect, and determine the final defect based on the matching degree, the first probability value and the second probability value.

[0085] In this embodiment, the prediction results (i.e., the first probability value) obtained based on three-dimensional point cloud data and the prediction results (i.e., the second probability value) obtained based on the image are comprehensively considered to determine the final defect, which further improves the accuracy of defect prediction, thereby improving the accuracy of airport pavement concrete surface performance evaluation and improving airport pavement safety.

[0086] In some implementations, determining the matching degree between the center point cloud coordinates of the 3D point cloud data of each cluster in the remaining cluster categories and the center location of the defect, and determining the final defect based on the matching degree, a first probability value, and a second probability value, includes:

[0087] If the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories match the center position of the defect, then the first probability value of each defect corresponding to the center point cloud coordinates and the second probability value of the defect corresponding to the center position are added together to obtain the total probability value of each defect. Among them, the defect corresponding to the first probability value is the same as the defect corresponding to the second probability value.

[0088] The defect with the highest total probability value is taken as the final defect.

[0089] In this embodiment, by using the defect with the highest total probability value as the final defect, the accuracy of the predicted defect can be further ensured, preventing prediction errors when using only 3D point cloud data or image data for prediction. Combining 3D point cloud data and image data can further improve the accuracy of the prediction results.

[0090] In some implementations, determining the matching degree between the center point cloud coordinates of the 3D point cloud data of each cluster in the remaining cluster categories and the center location of the defect, and determining the final defect based on the matching degree, a first probability value, and a second probability value, includes:

[0091] If the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories do not match the center location of the defect, then the defect with the highest probability corresponding to the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories will be taken as the final defect, or the defect corresponding to the center location will be taken as the final defect.

[0092] In this embodiment, when the center point cloud coordinates of the three-dimensional point cloud data of each cluster category in the remaining cluster categories do not match the center position of the defect, by retaining the defect with the highest probability corresponding to the center point cloud coordinates as the final defect, or by retaining the defect corresponding to the center position as the final defect, it is possible to prevent some defects from being missed, reduce the probability of defect omission, and improve the safety of the airport pavement.

[0093] To facilitate understanding by those skilled in the art, a set of preferred embodiments is provided below:

[0094] Over long-term use, airport pavement concrete structures are affected by environmental factors, loads, and material aging, leading to a decline in surface performance, such as cracks, spalling, and peeling. These problems not only affect the service life of the pavement but may also jeopardize flight safety. Therefore, developing an intelligent curing and real-time surface performance evaluation method for airport pavement concrete is crucial for its maintenance and management.

[0095] Traditional surface performance testing of airport pavement concrete relies on manual inspections. This method depends on the experience of workers to identify defects such as cracks, spalling, and peeling. It is subjective, inefficient, and prone to missing defects. To address these issues, researchers have developed intelligent surface performance testing methods for airport pavement concrete, such as those using large amounts of image data to train detection models. However, these methods suffer from poor stability and accuracy in complex environments (such as uneven lighting, water accumulation, and low nighttime illumination), making it difficult to accurately classify and identify various defects (such as cracks, spalling, and peeling). This leads to frequent misjudgments and missed detections, resulting in untimely and inaccurate discovery and treatment of airport pavement defects, increasing safety hazards.

[0096] To address the problem that existing technologies struggle to accurately classify and identify various defects, leading to frequent misjudgments and omissions that increase safety hazards, this application proposes a real-time evaluation method for the surface performance of airport pavement concrete. This embodiment of the method includes the following steps:

[0097] Step S1: Real-time 3D point cloud data acquisition of the airport pavement concrete is performed using LiDAR to obtain multiple 3D point cloud data points of the airport pavement to be predicted. Real-time image acquisition of the airport pavement concrete is performed using a telephoto camera to obtain multiple images of different areas of the airport pavement to be predicted. Buried sensors are installed in the airport pavement concrete; these buried sensors may include temperature sensors and humidity sensors, etc.

[0098] Step S2: Calculate the normal vector and curvature of each 3D point cloud data in the multiple 3D point cloud datasets to be predicted. A clustering method is used to cluster the multiple 3D point cloud datasets to be predicted, making the 3D point cloud data in each cluster relatively compact, i.e., the spatial distance between the 3D point cloud data in each cluster is small, resulting in a point cloud clustering result containing K clusters. The area of ​​the connected regions is calculated based on the 3D point cloud data in each cluster, obtaining the area of ​​the connected regions for each cluster. It should be noted that the clustering method can be the DBSCAN clustering method, which is well-known to those skilled in the art; this embodiment will not describe it specifically.

[0099] (1) Calculate the normal vector for each 3D point cloud data in the following manner:

[0100] For each 3D point cloud data d i The neighborhood radius method is used to select each 3D point cloud data d. i The neighborhood point set D i Calculate these neighborhood points d j The covariance matrix is ​​calculated using the following formula:

[0101]

[0102] Where P represents the neighborhood point d j The covariance matrix, where T denotes the transpose. Represents the neighborhood point d j The center of mass.

[0103] Eigenvalues ​​are obtained by performing eigenvalue decomposition on the covariance matrix:

[0104] P·v k =γ k ·v k

[0105] Where, γ k Let v represent the k-th eigenvalue. k This represents the eigenvector corresponding to the k-th eigenvalue.

[0106] The eigenvector corresponding to the smallest eigenvalue among multiple eigenvalues ​​is used as the 3D point cloud data x. i The normal vector, for example, if multiple eigenvalues ​​are obtained as γ1, γ2, and γ3, and γ1≤γ2≤γ3, then the eigenvector corresponding to γ1 is taken as the 3D point cloud data x. i The normal vector.

[0107] (2) Based on the obtained multiple feature values, calculate the curvature of each 3D point cloud data. The specific calculation formula is as follows:

[0108]

[0109] Where F represents the curvature of the 3D point cloud data.

[0110] (3) The area of ​​the connected region of each cluster category is calculated by performing the connected region calculation based on the three-dimensional point cloud data of each cluster category in the following manner.

[0111] Specifically, for each cluster category's 3D point cloud data, a surface reconstruction algorithm (such as the Poisson surface reconstruction method and the Delaunay triangulation method) is used to convert all 3D point cloud data of each cluster category into a triangular mesh (or a quadrilateral mesh, which is not specifically limited in this embodiment), constructing a mesh model. Therefore, this mesh model contains multiple triangles, and one mesh model corresponds to one connected region. That is, one mesh model contains all 3D point cloud data of one cluster category, only the 3D point cloud data has been converted into a mesh. Based on the constructed mesh model, the area of ​​the connected region is calculated as follows:

[0112]

[0113] Where S represents the area of ​​the connected region. and Let represent the two side vectors of a triangle, ‖·‖ represent the magnitude of the vectors, and N represent the number of triangles.

[0114] Step S3: Based on multiple 3D point cloud data to be predicted and the corresponding normal vectors, curvatures, and connected region areas of the multiple 3D point cloud data to be predicted, a support vector machine is used to predict the probability values ​​of defects such as cracks, spalling, and peeling on the surface of airport pavement concrete.

[0115] Specifically, since the defect-free airport pavement corresponds to the largest connected region area, while the others are basically defective connected regions (the connected regions in other cases (such as grass, branches, or leaves) can be ignored and not removed), the 3D point cloud data of the cluster category corresponding to the largest connected region area is removed to obtain the 3D point cloud data of the remaining cluster categories.

[0116] The support vector machine (SVM) is trained using the first training dataset to obtain the trained SVM (i.e., the first target prediction model). The first training dataset includes 3D point cloud data, the normal vectors, curvatures, and areas of connected regions corresponding to the 3D point cloud data, and includes defect category labels set for the 3D point cloud data.

[0117] The normal vector, curvature, and area of ​​the connected region corresponding to the 3D point cloud data of each of the remaining cluster categories are input into the support vector machine to obtain the category and probability value (i.e., the first probability value) of multiple defects corresponding to each of the remaining cluster categories. That is, the probability value that a cluster category may simultaneously correspond to defects such as cracks, peeling, and flaking.

[0118] Step S4: Perform image enhancement on each of the multiple images to be predicted to obtain multiple enhanced images to be predicted. Input the enhanced images to be predicted into the image defect prediction model to predict defects such as cracks, spalling, and peeling on the concrete surface of the airport pavement.

[0119] Specifically, image enhancement can be performed on each image using methods such as grayscale transformation, histogram equalization, noise reduction, and sharpening to obtain the enhanced image.

[0120] Based on the YOLO series architecture, an image defect prediction model can be constructed, which can adopt YOLOv3, YOLOv5, and YOLOv8 structures, etc. This embodiment does not provide specific descriptions or limitations.

[0121] Obtain a training dataset containing multiple enhanced images and defect category labels. This training dataset can be a manually constructed dataset.

[0122] The image defect prediction model is trained using a training dataset containing multiple enhanced images and defect category labels to obtain a trained image defect prediction model (i.e., the second target prediction model).

[0123] The trained image defect prediction model is used to predict defects in the enhanced image to be predicted, and the prediction results of defects such as cracks, peeling and flaking in the image to be predicted are output. That is, the probability value (i.e. the second probability value) of defects such as cracks, peeling and flaking and the center position of each defect are predicted (i.e. the center point coordinates of each defect in the bounding box coordinates of each defect).

[0124] Step S5: Analyze the surface performance of airport pavement concrete based on the categories and probability values ​​of multiple defects corresponding to each cluster category obtained in step S3 (i.e., the first probability value) and the probability values ​​of defects such as cracks, spalling and peeling obtained in step S4 (i.e., the second probability value).

[0125] Specifically, the center point cloud coordinates of the 3D point cloud data for each of the remaining cluster categories are calculated as follows:

[0126]

[0127] Where C represents the center point cloud coordinates, n represents the number of 3D point cloud data points for each cluster category, and x i Represents the x-coordinate and y-coordinate of the i-th 3D point cloud data in each cluster category. i The z-coordinate represents the y-coordinate of the i-th 3D point cloud data in each cluster category. i This represents the z-coordinate of the i-th 3D point cloud data in each cluster category.

[0128] Associate the categories of multiple defects corresponding to each cluster category in the remaining cluster categories with the center point cloud coordinates of the 3D point cloud data of each cluster category.

[0129] The center point cloud coordinates of the 3D point cloud data for each cluster category are matched with the center position of each defect in step S4. If the center point cloud coordinates of a cluster category match the center position of a defect, the value of the defect corresponding to the center point cloud coordinates (i.e., the first probability value) and the probability value of the defect corresponding to the center position (i.e., the second probability value) are added together to obtain the total probability value of each defect. The defect with the highest total probability value is taken as the final defect at the current center position or center point cloud coordinates. This further ensures the accuracy of defect prediction and prevents prediction errors when using only 3D point cloud data or image data. Combining 3D point cloud data and image data can further improve the accuracy of prediction results.

[0130] If the center point cloud coordinates of the 3D point cloud data of a single cluster do not match the center positions of all defects in step S4, then the defect with the highest probability corresponding to the center point cloud coordinates of the 3D point cloud data of each cluster is taken as the final defect, and the center point cloud coordinates of the final defect are retained. If the center position of a single defect in step S4 does not match the center point cloud coordinates of the 3D point cloud data of all clusters, then the single defect and its center position in step S4 are retained. Retaining these defects and their positions can prevent some defects from being missed. Even if the prediction error is not a defect, but some grass, branches, or leaves, they can be directly checked by observing the image, reducing the probability of missing defects. Grass, branches, or leaves also need to be removed because they pose a safety hazard, so they can be considered as a special type of defect.

[0131] The center point cloud coordinates need to be transformed by a coordinate system transformation, projecting the three-dimensional center point cloud coordinates onto the image plane to obtain two-dimensional coordinates, and then matching them with the center positions of all defects in step S4. It should be noted that the transformation of three-dimensional point cloud coordinates into two-dimensional coordinates can be done using existing techniques known to those skilled in the art, and will not be specifically described in this embodiment.

[0132] The surface performance of airport pavement concrete is analyzed based on the type of defect (cracks, spalling, and peeling, etc.), the location of the defect, the corresponding information of the defect (e.g., the area of ​​the connected region), and the total number of each defect. For example, if the number of cracks is less than a preset value or the area of ​​the connected region of spalling is less than a preset value, it indicates that the surface performance of the airport pavement concrete is good. This can be set according to actual conditions, and this embodiment does not impose specific limitations. If the surface performance of the airport pavement concrete is poor and falls below a certain value, an early warning will be issued, reminding staff to pay attention to the defects and locate and repair the corresponding defects as needed. Furthermore, based on the type of defect (cracks, spalling, and peeling, etc.) and the location of the defect, it can help inspection staff speed up the inspection process, improve the inspection effect, reduce the probability of missed defects, and improve the safety of the airport pavement. Since three-dimensional point cloud data and image data can be acquired in real time, this embodiment also realizes real-time evaluation of the surface performance of airport pavement concrete.

[0133] Step S6: Based on the data obtained from the buried sensors (i.e., temperature sensors and humidity sensors, etc.) and the surface properties of the airport pavement concrete obtained in Step 5, the airport pavement concrete is cured.

[0134] Specifically, when the surface performance of the airport pavement concrete is good, the temperature and humidity of the airport pavement concrete can be set to a certain value; when the surface performance of the airport pavement concrete is poor, the temperature and humidity of the airport pavement concrete can be set to another value. Then, based on the temperature obtained from the temperature sensor and the humidity obtained from the humidity sensor, the temperature and humidity of the airport pavement concrete are adjusted to reach the set values, so as to realize intelligent curing of the airport pavement concrete.

[0135] Reference Figure 2 This application also provides a real-time evaluation system for the surface performance of airport pavement concrete. The system includes a data acquisition unit 100, a first calculation unit 200, a second calculation unit 300, a first prediction unit 400, a second prediction unit 500, a defect determination unit 600, and a performance analysis unit 700, wherein:

[0136] The data acquisition unit 100 is used to acquire multiple three-dimensional point cloud data and multiple images to be predicted from the concrete surface of the airport pavement.

[0137] The first computing unit 200 is used to calculate the normal vector and curvature of each 3D point cloud data to be predicted;

[0138] The second computing unit 300 is used to cluster multiple 3D point cloud data to be predicted, obtain multiple cluster categories, and calculate the area of ​​the connected region corresponding to each cluster category.

[0139] The first prediction unit 400 is used to predict the first probability value of defects existing on the concrete surface of the airport pavement based on the normal vectors, curvatures and the area of ​​the connected regions corresponding to each cluster category of multiple three-dimensional point cloud data to be predicted, using the first target prediction model.

[0140] The second prediction unit 500 is used to input multiple images to be predicted into the second target prediction model to predict defects and obtain the second probability value and center position of the defects existing on the concrete surface of the airport pavement.

[0141] The defect determination unit 600 is used to determine the final defect based on the center location of the defect, a first probability value, and a second probability value.

[0142] The performance analysis unit 700 is used to analyze the surface properties of airport pavement concrete based on the final defects.

[0143] It should be noted that since the real-time evaluation system for the surface performance of airport pavement concrete in this embodiment is based on the same inventive concept as the aforementioned real-time evaluation method for the surface performance of airport pavement concrete, the corresponding content in the method embodiment is also applicable to this system embodiment, and will not be described in detail here.

[0144] Reference Figure 3 This application also provides an electronic device, which includes:

[0145] At least one memory;

[0146] At least one processor;

[0147] At least one program;

[0148] The program is stored in memory, and the processor executes at least one program to implement the above-described method for real-time evaluation of the surface performance of airport pavement concrete.

[0149] This electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), and in-vehicle computers.

[0150] The electronic devices according to embodiments of this application will now be described in detail.

[0151] The processor 1600 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this disclosure.

[0152] The memory 1700 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1700 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1700 and is called and executed by the processor 1600 to execute the real-time evaluation method for the surface performance of airport pavement concrete according to the embodiments of this disclosure.

[0153] The input / output interface 1800 is used to implement information input and output.

[0154] The communication interface 1900 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0155] Bus 2000 transmits information between various components of the device (e.g., processor 1600, memory 1700, input / output interface 1800, and communication interface 1900);

[0156] The processor 1600, memory 1700, input / output interface 1800 and communication interface 1900 are connected to each other within the device via bus 2000.

[0157] This disclosure also provides a storage medium, which is a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-described method for real-time evaluation of the surface performance of airport pavement concrete.

[0158] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0159] The embodiments described in this disclosure are for the purpose of more clearly illustrating the technical solutions of this disclosure and do not constitute a limitation on the technical solutions provided by this disclosure. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by this disclosure are also applicable to similar technical problems.

[0160] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this disclosure, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0161] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0162] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0163] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0164] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0165] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0166] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0167] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0168] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks. The embodiments of this application have been described in detail above with reference to the accompanying drawings, but this application is not limited to the above embodiments. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of this application.

[0169] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application.

Claims

1. A method for real-time evaluation of the surface performance of an airport pavement concrete, characterized in that, The method includes: Acquire multiple 3D point cloud data and multiple images of the concrete surface of the airport pavement to be predicted; Calculate the normal vector and curvature of each of the three-dimensional point cloud data to be predicted; Cluster the multiple 3D point cloud data to be predicted to obtain multiple cluster categories, and calculate the area of ​​the connected region corresponding to each cluster category; Based on the normal vectors, curvatures, and connected region areas corresponding to each cluster category of the multiple 3D point cloud data to be predicted, a first probability value for defects existing on the concrete surface of the airport pavement is predicted using a first target prediction model, including: Compare the size of the connected region area corresponding to each cluster category, remove the 3D point cloud data of the cluster category with the largest connected region area, and obtain the 3D point cloud data of the remaining cluster categories. The normal vector, curvature, and area of ​​the connected region corresponding to the three-dimensional point cloud data of each of the remaining cluster categories are input into the first target prediction model to predict the category and first probability value of each of the multiple defects corresponding to each of the remaining cluster categories. The defects are defects existing on the concrete surface of the airport pavement. The multiple images to be predicted are input into the second target prediction model to predict defects, thereby obtaining the second probability value and center location of the defects existing on the concrete surface of the airport pavement. Based on the center location of the defect, the first probability value, and the second probability value, the final defect is determined, including: Calculate the center point cloud coordinates of the 3D point cloud data for each of the remaining cluster categories; Associate the center point cloud coordinates with the respective categories of the multiple defects; Determine the matching degree between the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories and the center position of the defect, and determine the final defect based on the matching degree, the first probability value, and the second probability value, including: If the center point cloud coordinates of the three-dimensional point cloud data of each of the remaining cluster categories match the center position of the defect, then the first probability value of each defect corresponding to the center point cloud coordinates and the second probability value of the defect corresponding to the center position are added together to obtain the total probability value of each defect, wherein the defect corresponding to the first probability value and the defect corresponding to the second probability value are the same. The defect with the highest total probability value is taken as the final defect; Based on the aforementioned final defects, the surface properties of airport pavement concrete were analyzed.

2. The method for real-time evaluation of the surface performance of airport pavement concrete according to claim 1, characterized in that, The calculation of the normal vector and curvature of each of the three-dimensional point cloud data to be predicted includes: Calculate the normal vector for each of the three-dimensional point cloud data to be predicted: Calculate the curvature of each of the three-dimensional point cloud data to be predicted: in, Represents each 3D point cloud data The neighborhood point set, Representing neighborhood points The covariance matrix, Indicates transpose. Representing neighborhood points The center of mass, , and Represents the eigenvalue, and , Represents the curvature of 3D point cloud data.

3. The method for real-time evaluation of the surface performance of airport pavement concrete according to claim 1, characterized in that, The calculation of the area of ​​the connected region corresponding to each cluster category includes: Convert the 3D point cloud data in each cluster category into a mesh and construct a mesh model; Calculate the sum of the areas of all grids in the grid model to obtain the area of ​​the connected region corresponding to each cluster category.

4. The method for real-time evaluation of the surface performance of airport pavement concrete according to claim 1, characterized in that, The step of determining the matching degree between the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories and the center position of the defect, and determining the final defect based on the matching degree, the first probability value, and the second probability value, includes: If the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories do not match the center position of the defect, then the defect with the highest probability corresponding to the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories is taken as the final defect, or the defect corresponding to the center position is taken as the final defect.

5. A real-time evaluation system for the surface performance of airport pavement concrete, characterized in that, The system includes: The data acquisition unit is used to acquire multiple three-dimensional point cloud data and multiple images of the concrete surface of the airport pavement to be predicted. The first computing unit is used to calculate the normal vector and curvature of each of the three-dimensional point cloud data to be predicted; The second calculation unit is used to cluster the multiple 3D point cloud data to be predicted, obtain multiple cluster categories, and calculate the area of ​​the connected region corresponding to each cluster category. The first prediction unit is used to predict the first probability value of defects existing on the concrete surface of the airport pavement based on the normal vectors, curvatures, and connected region areas corresponding to each cluster category of the multiple three-dimensional point cloud data to be predicted, using a first target prediction model, including: Compare the size of the connected region area corresponding to each cluster category, remove the 3D point cloud data of the cluster category with the largest connected region area, and obtain the 3D point cloud data of the remaining cluster categories. The normal vector, curvature, and area of ​​the connected region corresponding to the three-dimensional point cloud data of each of the remaining cluster categories are input into the first target prediction model to predict the category and first probability value of each of the multiple defects corresponding to each of the remaining cluster categories. The defects are defects existing on the concrete surface of the airport pavement. The second prediction unit is used to input the multiple images to be predicted into the second target prediction model to predict defects, and obtain the second probability value and center position of the defects existing on the concrete surface of the airport pavement. The defect determination unit is configured to determine the final defect based on the center position of the defect, the first probability value, and the second probability value, including: Calculate the center point cloud coordinates of the 3D point cloud data for each of the remaining cluster categories; Associate the center point cloud coordinates with the respective categories of the multiple defects; Determine the matching degree between the center point cloud coordinates of the 3D point cloud data of each of the remaining cluster categories and the center position of the defect, and determine the final defect based on the matching degree, the first probability value, and the second probability value, including: If the center point cloud coordinates of the three-dimensional point cloud data of each of the remaining cluster categories match the center position of the defect, then the first probability value of each defect corresponding to the center point cloud coordinates and the second probability value of the defect corresponding to the center position are added together to obtain the total probability value of each defect, wherein the defect corresponding to the first probability value and the defect corresponding to the second probability value are the same. The defect with the highest total probability value is taken as the final defect; The performance analysis unit is used to analyze the surface properties of airport pavement concrete based on the final defects.

6. An electronic device, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which, when executed by the at least one control processor, enable the at least one control processor to perform the real-time evaluation method for the surface performance of airport pavement concrete as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the real-time evaluation method for the surface performance of airport pavement concrete as described in any one of claims 1 to 4.

Images