Training method of road disease prediction model, road disease prediction method and product

Abstract

The invention discloses a training method of a road disease prediction model, a road disease prediction method and a product, and relates to the technical field of road disease prediction. Determining the number of heavy vehicles passing through the training road and passing time data; training a machine learning algorithm by using the number of the heavy vehicles passing through the training road, the passing time data and the historical inspection data of the training road to obtain a road disease prediction model; the output of the road disease prediction model is the probability of the road disease of the road for training. Road diseases are predicted based on the heavy traffic flow, and investment of complex hardware equipment is avoided mainly based on historical data of the diseases and the traffic flow.

Description

Technical Field

[0001] The present application relates to the technical field of road disease prediction, and particularly to a method for training a road disease prediction model, a road disease prediction method and a product. Background Art

[0002] Road disease prediction can significantly improve the inspection efficiency and accuracy, reduce the missed report rate and improve the road maintenance efficiency. The existing road disease prediction technologies mainly use digital technologies for inspection, such as using drones to collect road images, and then using image recognition technologies to identify road diseases. There are also lidar and other devices that can provide high-precision pavement detection. The void height can be estimated through the reflection characteristics of radar waves, so as to detect and predict diseases more accurately. However, these technologies have high requirements for hardware devices. Summary of the Invention

[0003] The purpose of the present application is to provide a method for training a road disease prediction model, a road disease prediction method and a product, which can predict road diseases based on the heavy vehicle flow, mainly based on the historical data of diseases and vehicle flow, and do not require complex hardware device investment.

[0004] To achieve the above purpose, the present application provides the following solutions:

[0005] In the first aspect, the present application provides a method for training a road disease prediction model. The method for training the road disease prediction model includes:

[0006] Obtain the historical and positioning data of a preset area; the historical and positioning data includes: the positioning data of historical heavy vehicles, the positioning data of the training road, and the historical inspection data of the training road; the historical inspection data includes: road disease data and corresponding inspection time data;

[0007] According to the positioning data of the historical heavy vehicles and the positioning data of the training road, determine the number of heavy vehicles passing through the training road and the passing time data;

[0008] Use the number of heavy vehicles passing through the training road, the passing time data, and the historical inspection data of the training road to train a machine learning algorithm to obtain a road disease prediction model; the output of the road disease prediction model is the probability of road diseases occurring on the training road.

[0009] Optionally, according to the positioning data of the historical heavy vehicles and the positioning data of the training road, determining the number of heavy vehicles passing through the training road and the passing time data specifically includes:

[0010] Segment the training road by intersections to obtain a number of road sub-segments;

[0011] Divide several of the road sub - segments into upward and downward directions along the road center line, and number each of the road sub - segments after being divided into upward and downward directions to obtain several numbered sub - segments;

[0012] Use an electronic fence to enclose each of the numbered sub - segments according to the positioning data of the training road to obtain several fenced sub - segments;

[0013] Judge the time when each heavy vehicle passes through each of the fenced sub - segments according to the positioning data of the historical heavy vehicles, and obtain the number of heavy vehicles passing through the training road and the passing time data.

[0014] Optionally, use the number of heavy vehicles passing through the training road, the passing time data, and the historical inspection data of the training road to train a machine learning algorithm to obtain a road disease prediction model, specifically including:

[0015] Obtain the historical weather data of a preset area;

[0016] Organize the number of heavy vehicles passing through the training road, the passing time data, the historical inspection data of the training road, and the historical weather data into a table form corresponding to time to obtain a historical data table;

[0017] Use the historical data table to train a machine learning algorithm to obtain a road disease prediction model;

[0018] The historical data table includes: date, training road name, heavy vehicle traffic volume, road disease data, and weather data.

[0019] Optionally, the road disease data includes: road crack, pothole, and subsidence data.

[0020] Optionally, the positioning data of the historical heavy vehicles is the GPS data of the historical heavy vehicles.

[0021] In a second aspect, the present application provides a road disease prediction method, and the road disease prediction method includes:

[0022] Obtain the number of heavy vehicles passing through the road to be predicted and the passing time;

[0023] Input the number of heavy vehicles passing through the road to be predicted and the passing time into the road disease prediction model to obtain the probability of road diseases occurring on the road to be predicted; the road disease prediction model is trained by the training method of the road disease prediction model described in any one of the above.

[0024] Optionally, after inputting the number of heavy vehicles passing through the road to be predicted and the passing time into the road disease prediction model to obtain the probability of road diseases occurring on the road to be predicted, it further includes:

[0025] Sort the probabilities of road diseases occurring on the to-be-predicted roads to obtain a sorting result;

[0026] Mark the to-be-predicted roads corresponding to the top n probabilities in the sorting result as key inspection roads.

[0027] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the training method or the road disease prediction method of any one of the above road disease prediction models.

[0028] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the training method or the road disease prediction method of any one of the above road disease prediction models.

[0029] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the training method or the road disease prediction method of any one of the above road disease prediction models.

[0030] According to the specific embodiments provided by the present application, the following technical effects are disclosed:

[0031] The present application provides a training method for a road disease prediction model, a road disease prediction method, and a product. The training method for the road disease prediction model includes: obtaining historical and positioning data of a preset area; the historical and positioning data includes: positioning data of historical heavy vehicles, positioning data of training roads, and historical inspection data of training roads; the historical inspection data includes: road disease data and corresponding inspection time data; according to the positioning data of historical heavy vehicles and the positioning data of training roads, determining the number of heavy vehicles passing through the training roads and the passing time data; using the number of heavy vehicles passing through the training roads and the passing time data and the historical inspection data of the training roads to train a machine learning algorithm to obtain a road disease prediction model; the output of the road disease prediction model is the probability of road diseases occurring on the training roads. The present application predicts road diseases based on heavy vehicle traffic, mainly based on historical data of diseases and traffic flow, avoiding the investment in complex hardware devices. Description of the Drawings

[0032] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0033] Figure 1 It is an application environment diagram of a training method for a road disease prediction model and a road disease prediction method provided by an embodiment of the present application;

[0034] Figure 2 It is a schematic flowchart of a training method for a road disease prediction model provided by Embodiment 1 of the present application;

[0035] Figure 3 It is a schematic diagram of a road disease prediction method provided by Embodiment 3 of the present application;

[0036] Figure 4 It is a schematic diagram of a training data table provided by Embodiment 3 of the present application;

[0037] Figure 5 It is a schematic diagram of the structure of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0039] The present application predicts road diseases based on the heavy vehicle flow, mainly based on the historical data of diseases and vehicle flow, avoiding the investment in hardware equipment, enabling targeted road inspections, making the inspection work more efficient and accurate, promptly discovering and handling road diseases, and ensuring the safety and smoothness of road traffic.

[0040] To make the above objects, features, and advantages of the present application more obvious and understandable, the following will further describe the present application in detail in conjunction with the accompanying drawings and specific implementation manners.

[0041] The training method for the road disease prediction model or the road disease prediction method provided by the embodiments of the present application can be applied to, for example Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set separately, integrated on the server 104, placed on the cloud or other servers. Among them, the terminal 102 can be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers, and can also be a cloud server.

[0042] Embodiment 1:

[0043] In an exemplary embodiment, as Figure 2 shown, a method for training a road disease prediction model is provided. This method is executed by a computer device, and can be specifically executed by a computer device such as a terminal or a server alone, or jointly executed by a terminal and a server. In the embodiments of the present application, taking this method applied to Figure 1 the server 104 in it as an example for illustration, it includes the following steps S1 to S3. Among them:

[0044] S1. Obtain the historical and positioning data of a preset area; the historical and positioning data includes: the positioning data of historical heavy vehicles, the positioning data of the training road, and the historical inspection data of the training road; the historical inspection data includes: road disease data and corresponding inspection time data. The road disease data includes: road crack, pothole, and subsidence data. The positioning data of historical heavy vehicles is the GPS data of historical heavy vehicles. Among them, the positioning data of historical heavy vehicles is the positioning data of multiple heavy vehicles in this preset area in history. The preset area is an area containing the training road. For example, when predicting the disease situation of a certain road in Shanghai, the preset area can be set as Shanghai.

[0045] S2. Determine the number of heavy vehicles passing through the training road and the passing time data according to the positioning data of the historical heavy vehicles and the positioning data of the training road.

[0046] The specific steps include:

[0047] Segment the training road by intersections to obtain several road sub-segments;

[0048] Divide several of the road sub-segments into upstream and downstream according to the road center line, and number each of the road sub-segments after being divided into upstream and downstream to obtain several numbered sub-segments;

[0049] Using an electronic fence, enclose each of the numbered sub - segments according to the positioning data of the training road to obtain a number of fence sub - segments;

[0050] Judge the time when each heavy vehicle passes through each of the fence sub - segments according to the positioning data of the historical heavy vehicles, and obtain the number of heavy vehicles passing through the training road and the passing time data.

[0051] S3. Use the number of heavy vehicles passing through the training road, the passing time data, and the historical inspection data of the training road to train a machine learning algorithm to obtain a road disease prediction model; the output of the road disease prediction model is the probability of road diseases occurring on the training road. The machine learning algorithm is a random forest algorithm.

[0052] Among them, as an optional implementation manner, weather data can also be obtained in this embodiment and added during training to improve the prediction accuracy.

[0053] When considering weather data, use the number of heavy vehicles passing through the training road, the passing time data, and the historical inspection data of the training road to train a machine learning algorithm to obtain a road disease prediction model, which specifically includes:

[0054] Obtain the historical weather data of a preset area.

[0055] Organize the number of heavy vehicles passing through the training road, the passing time data, the historical inspection data of the training road, and the historical weather data into a table form corresponding to time to obtain a historical data table.

[0056] Use the historical data table to train a machine learning algorithm to obtain a road disease prediction model.

[0057] The historical data table includes: date, training road name, heavy vehicle traffic volume, road disease data, and weather data.

[0058] In this embodiment, by collecting and analyzing the GPS data of heavy vehicles on the road, combining the historical road disease data and weather data, and performing modeling calculations, a prediction of the road disease incidence rate is formed, providing data support for road inspection and road maintenance, and effectively improving the maintenance efficiency.

[0059] Embodiment Two:

[0060] In an exemplary embodiment, a road disease prediction method is provided, including:

[0061] A1. Obtain the number of heavy vehicles passing through the road to be predicted and the passing time.

[0062] A2. Input the number of heavy vehicles passing through the road to be predicted and the passing time into the road disease prediction model to obtain the probability of road diseases occurring on the road to be predicted; the road disease prediction model is trained by the training method of the road disease prediction model described in Embodiment 1.

[0063] A3. Sort the probabilities of road diseases occurring on the road to be predicted to obtain a sorting result.

[0064] A4. Mark the roads to be predicted corresponding to the top n probabilities in the sorting result as key inspection roads.

[0065] Among them, when considering weather data, it is also necessary to obtain the weather data of the area corresponding to the road to be predicted.

[0066] Organize the number of heavy vehicles passing through the road to be predicted, the passing time, and the weather data of the corresponding area into a table form corresponding to time, and input it into the road disease prediction model to obtain the probability of road diseases occurring on the road to be predicted.

[0067] Embodiment 3:

[0068] In an exemplary embodiment, as Figure 3 shown, a road disease prediction method is provided, including:

[0069] S0. Segment the roads within the analysis scope according to the road basic information, and the segmentation steps are as follows:

[0070] S0-1. Break a road at intersections to obtain road sub-segments.

[0071] S0-2. Divide the sub-segments into upstream and downstream according to the road center line.

[0072] S0-3. Number each segment.

[0073] S1. Draw an electronic fence for the segments generated in S0 to enclose each segment.

[0074] S2. Match the heavy vehicle GPS data with the electronic fences of the road segments to obtain a matching table of license plates, hours, and road segments, which can reflect which heavy vehicles pass through a specific segment within a specific hour.

[0075] S3. According to the matching table of license plates, dates, and road segments obtained in S2, the number of heavy vehicles passing through each segment every day can be counted.

[0076] S4. According to the result of S3, the heavy vehicle data of the segments to which the road belongs can be accumulated to count the daily heavy vehicle traffic volume of each road.

[0077] S5. Obtain the number of road diseases detected in daily inspections from the road inspection department. Road diseases include road cracks, potholes, subsidence, etc.

[0078] S6. Generate a historical data table of date, road name, heavy vehicle traffic volume, number of road diseases, and weather.

[0079] S7. Use the random forest algorithm to predict the number of road diseases on the target date. The specific prediction steps are as follows:

[0080] S7-1. Preparation of random forest model training data:

[0081] According to the table generated in S6, prepare historical data for a period of time to generate random forest model training data. Dd represents the date feature of the d-th day, Jd represents the holiday feature of the d-th day, Vd represents the heavy vehicle traffic volume value of the d-th day, Wd represents the weather condition of the d-th day, Md represents the historical number of diseases on the d-th day, and Pd represents the predicted number of diseases, that is, the target value expected to be obtained through these features, as Figure 4 shown.

[0082] S7-2. Construct a random forest model: Use the training set data to construct a random forest model. A random forest consists of multiple decision trees, and each decision tree is constructed based on randomly selected samples and features. When constructing each decision tree, the Bootstrap sampling method is used to randomly draw multiple subsets from the original training set with replacement, and each subset is used to train a decision tree.

[0083] S7-3. Train the decision tree: For each decision tree, use randomly selected samples and features for training. The training process of the decision tree usually uses a recursive splitting algorithm to split according to the impurity of the features.

[0084] S7-4. Prediction: Use the trained random forest to predict the test set. For regression prediction, the random forest will average or weighted average the prediction results of multiple decision trees to obtain the final predicted value.

[0085] S8. Sort the number of road diseases predicted in S7-4. The top ten roads with the highest probability of disease occurrence need to be inspected with key emphasis.

[0086] In this embodiment, the prediction of road diseases based on heavy vehicle traffic volume is mainly based on the historical data of diseases and traffic volume, avoiding the investment in hardware equipment. The invention can make the road inspection targeted, and the inspection work can be more efficient and accurate, timely discover and handle road diseases, and ensure the safety and smoothness of road traffic. Among them, by using the inspection vehicle to collect road images and using AI technology to identify road images, road diseases can also be discovered. However, this solution requires a comprehensive inspection of all roads and cannot achieve strong pertinence.

[0087] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal, and its internal structure diagram may be as shown in Figure 5 . The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for training a road disease prediction model or a road disease prediction method.

[0088] Those skilled in the art can understand that Figure 5 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0089] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0090] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0091] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0092] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0093] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the various embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0094] The databases involved in the various embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the various embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0095] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0096] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for training a road damage prediction model, characterized in that: The training method of the road damage prediction model includes: Obtaining historical and positioning data of a preset area; the historical and positioning data include: historical heavy vehicle positioning data, training road positioning data, and historical inspection data of training roads; the historical inspection data include: road disease data and corresponding inspection time data; Determining the number of heavy vehicles passing through the training road and the passing time data according to the historical heavy vehicle positioning data and the positioning data of the training road; The number of heavy vehicles passing through the training road and the passing time data as well as the historical inspection data of the training road are used to train a machine learning algorithm to obtain a road damage prediction model; the output of the road damage prediction model is the probability of road damage occurring on the training road.

2. The method for training a road damage prediction model according to claim 1, characterized in that: Determining the number of heavy vehicles passing through the training road and the passing time data according to the historical heavy vehicle positioning data and the training road positioning data, specifically includes: Divide the training road into sections according to intersections to obtain several road sub-segments; Dividing the plurality of road sub-segments into an upward direction and a downward direction according to the center line of the road, and numbering each of the road sub-segments after being divided into an upward direction and a downward direction to obtain a plurality of numbered sub-segments; Using an electronic fence, according to the positioning data of the training road, each of the numbered sub-segments is enclosed to obtain a plurality of fence sub-segments; The time when each heavy-duty vehicle passes through each fence sub-segment is determined according to the historical heavy-duty vehicle positioning data, and the number of heavy-duty vehicles passing through the training road and the passing time data are obtained.

3. The method for training a road damage prediction model according to claim 1, characterized in that: Using the data on the number of heavy vehicles passing through the training road and the time of passing through, as well as the historical inspection data of the training road, a machine learning algorithm is trained to obtain a road disease prediction model, specifically including: Get historical weather data for a preset area; Arrange the number of heavy vehicles passing through the training road and the passing time data, the historical inspection data of the training road and the historical weather data into a table format corresponding to time to obtain a historical data table; Using the historical data table to train a machine learning algorithm to obtain a road disease prediction model; The historical data table includes: date, name of training road, heavy vehicle traffic, road disease data and weather data.

4. The method for training a road damage prediction model according to claim 1, characterized in that: The road damage data include: road cracks, potholes and subsidence data.

5. The method for training a road damage prediction model according to claim 1, characterized in that: The historical heavy vehicle positioning data is the GPS data of the historical heavy vehicle.

6. A road disease prediction method, characterized in that: The road damage prediction method comprises: Obtain the number and passing time of heavy vehicles passing the road to be predicted; The number of heavy vehicles passing through the road to be predicted and the passing time are input into a road damage prediction model to obtain the probability of road damage occurring on the road to be predicted; the road damage prediction model is trained by the road damage prediction model training method described in any one of claims 1-5.

7. The road damage prediction method according to claim 6, characterized in that: After the number of heavy vehicles passing the road to be predicted and the passing time are input into the road damage prediction model to obtain the probability of the road damage occurring on the road to be predicted, the method further includes: Sorting the probability of occurrence of road damage on the to-be-predicted road to obtain a sorting result; The to-be-predicted roads corresponding to the first n probabilities in the sorting results are marked as key inspection roads.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the road damage prediction model training method described in any one of claims 1 to 5 or the road damage prediction method described in any one of claims 6 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the road damage prediction model training method described in any one of claims 1-5 or the road damage prediction method described in any one of claims 6-7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the road damage prediction model training method described in any one of claims 1-5 or the road damage prediction method described in any one of claims 6-7.

Images