Highway maintenance early warning and construction collaborative management method and platform

Abstract

The invention discloses a highway maintenance early warning and construction collaborative management method and platform, relates to the technical field of highway maintenance management, and realizes efficient construction collaborative management. The method comprises the following steps: firstly, determining road monitoring positions of a target area, and installing networking monitoring equipment at the positions for real-time data acquisition; thirdly, through analysis of historical vehicle flow data, dividing peak, average and low peak time periods of a road, and providing a basis for subsequent maintenance decision making; in different time periods, the system automatically adjusts the image acquisition frequency according to the traffic flow to ensure that images with higher frequency are acquired in peak periods so as to capture more road conditions; then, analyzing the collected image, automatically identifying and marking a pavement problem area, and classifying the pavement problem area into a light or serious maintenance problem area according to the severity of the problem; and finally, arranging a construction team to carry out maintenance work according to the classification results.

Description

Technical Field

[0001] The present invention relates to the technical field of highway maintenance management, and in particular to a method and platform for highway maintenance early warning and construction collaborative management. Background Art

[0002] With the continuous development of society and the increasing demand for transportation, the scale of highway construction is expanding. Smart highways are becoming an important direction for improving road management efficiency and ensuring driving safety. Smart highway construction requires not only advanced communication, monitoring, and control technologies, but also the ability to use intelligent means for real-time monitoring and timely maintenance to ensure the service life and safety of the highway.

[0003] Highway maintenance refers to the daily management, upkeep, and repair of existing highways to ensure they maintain good functionality, safety, and durability during use. This process typically includes regular inspections, repairs, restoration, and reinforcement of roads. However, in the actual construction and maintenance of highways, many maintenance issues are not addressed promptly due to insufficient technical means, limited maintenance funds, and limited personnel. This is particularly true at certain locations on highways, such as curves, bends, and uphill and downhill sections, where cracks, wear, and potholes often occur. These unrepaired maintenance issues can easily lead to accidents during driving, posing a traffic safety hazard.

[0004] In traditional highway maintenance management models, road maintenance relies heavily on periodic inspections. This can cause potential maintenance issues to remain undetected for extended periods. As traffic increases, these issues escalate, creating even greater driving hazards. Furthermore, the complex and ever-changing nature of highways creates uncertainty about the occurrence of road surface problems as vehicles pass through them, making timely detection and repair of road surface issues even more challenging. Summary of the Invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and platform for highway maintenance early warning and construction collaborative management, which solves the problems in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for highway maintenance early warning and construction collaborative management, comprising:

[0007] Step 1: Obtain the highway monitoring locations in the target area and install monitoring network equipment at the highway monitoring locations;

[0008] Step 2: Determine the benchmark passing vehicle value by dividing the time period and the historical passing vehicles, and determine whether the target time period is a peak vehicle period, a non-peak vehicle period, or a low vehicle period based on the benchmark passing vehicle value;

[0009] Step 3: At the highway monitoring location, images are collected using different image collection frequencies using monitoring network equipment during peak hours, off-peak hours, or low-peak hours.

[0010] Step 4: Acquire and analyze images collected at the highway monitoring location to identify problem areas in the images of the highway monitoring location. Once the problem areas are identified, further classify the problem areas into severe maintenance problem areas or mild maintenance problem areas.

[0011] Step 5: Based on the identified areas with severe maintenance problems and areas with mild maintenance problems, arrange for construction teams to immediately carry out maintenance at the monitoring locations of the highways in the areas with severe maintenance problems, and then arrange for construction teams to carry out maintenance at the monitoring locations of the highways in the areas with mild maintenance problems within a set time.

[0012] As a further solution of the present invention: in step 1, the highway monitoring positions include highway curve positions and highway uphill and downhill positions.

[0013] As a further solution of the present invention, the specific contents of determining the target time period as a peak time period, a non-peak time period or a low-peak time period by dividing the time period and determining the benchmark passing vehicle value based on the historical passing vehicle value are as follows:

[0014] AS1: Divide the current day into x periods, and use the first period as the target period for the following analysis steps; x is a preset value;

[0015] AS2: Obtain the number of vehicles that passed through the target period every day in the previous n days and mark it as Ci, where 1≤i≤n, and the previous n days means starting from the current day and moving forward n days; n is a preset value;

[0016] AS3: Calculate the average value of the vehicle values Ci during the target period in the previous n days and record it as Cp;

[0017] AS4: Determine the deviation value Z using the standard deviation formula based on the average value Cp and the number of vehicles passing through each day Ci;

[0018] AS5: Get the deviation value Z and compare it with the preset value G1:

[0019] If Z≤G1, it means that the deviation value Z is within a reasonable range, and Cp is used as the benchmark passing vehicle value during the target period;

[0020] If Z>G1, it means the deviation value Z is too large;

[0021] AS6: When the deviation value Z is too large, the data of the vehicle passing value Ci for each of the n days is deleted. The specific method of data deletion is:

[0022] Sort the daily passing vehicle values \(C_i\) for \(n\) days in descending order according to \(|C_i - C_p|\). Each time, obtain the \(C_i\) with the first rank in the sorting and delete it. After deletion, recalculate the deviation value \(Z\) of the remaining \(C_i\) until \(Z\leq G_1\).

[0023] Then, intercept the number of deleted \(C_i\) values, mark it as \(g_1\), calculate the ratio of \(g_1\) to \(n\), mark it as \(gb_1\), and compare \(gb_1\) with the preset value \(G_2\):

[0024] If \(gb_1\leq G_2\), calculate the average value of the remaining \(C_i\) as the benchmark passing vehicle value for the target time period.

[0025] If \(gb_1 > G_2\), obtain the maximum value of \(C_i\) in the previous \(m\) days as the benchmark passing vehicle value for the target time period, where \(m < n\), and the previous \(m\) days refer to \(m\) days before the current day starting from the current day.

[0026] AS7: Determine whether the target time period on the current day is a vehicle peak time period, a vehicle flat peak time period, or a vehicle low peak time period according to the benchmark passing vehicle value for the target time period;

[0027] AS8: Take the remaining time period on the current day as the target time period, and repeat steps AS2 - AS7 to determine whether the target time period on the current day is a vehicle peak time period, a vehicle flat peak time period, or a vehicle low peak time period.

[0028] As a further solution of the present invention: In the step AS7, the specific method for determining whether the target time period is a vehicle peak time period, a vehicle flat peak time period, or a vehicle low peak time period according to the benchmark passing vehicle value is as follows:

[0029] If the benchmark passing vehicle value exceeds the preset value \(G_4\), mark the target time period as a vehicle peak time period;

[0030] If the benchmark passing vehicle value exceeds \(G_3\) but does not exceed the preset value \(G_4\), mark the target time period as a vehicle flat peak time period;

[0031] If the benchmark passing vehicle value does not exceed the preset value \(G_3\), mark the target time period as a vehicle low peak time period;

[0032] Among them, \(G_4 > G_3\).

[0033] As a further solution of the present invention: In the third step, the specific method for collecting images of the highway monitoring location at different image acquisition frequencies by the monitoring networking device during the peak time period, the flat peak time period, or the low peak time period is as follows:

[0034] When the highway monitoring location is in the peak time period: Conduct a pavement image acquisition every interval time \(T_1\);

[0035] When the highway monitoring location is in peak hours: road surface image acquisition is performed every time interval T2;

[0036] When the highway monitoring location is in the off-peak period: road surface image acquisition is performed every time interval T3;

[0037] Among them, T1, T2 and T3 are all preset values, and T3>T2>T1.

[0038] As a further solution of the present invention: in step 4, the image of the highway monitoring location is acquired and analyzed to determine the problem area in the image of the highway monitoring location. When the problem area is determined, the problem area is further divided into a serious maintenance problem area or a mild maintenance problem area. The specific content is:

[0039] BS1: Acquire images of highway monitoring locations and use object detection technology to detect and annotate vehicle locations and areas in the images;

[0040] BS2: For the marked vehicle area, the vehicle area is separated from the road surface area through image segmentation technology, the vehicle information in the image is eliminated, and only the road surface part of the highway monitoring position is retained;

[0041] BS3: Process images from highway monitoring locations for crack detection, wear detection, and pothole detection;

[0042] BS4: When cracks, wear, or potholes are detected in the processed images of highway monitoring locations, the areas with the problematic images are marked for maintenance.

[0043] BS5: If maintenance marking is performed on the same area of the processed image at the highway monitoring location three times in a row, the area is determined to be a problem area.

[0044] As a further solution of the present invention: also include:

[0045] BS6: Obtain the area of the problem area, recorded as MG, and determine whether the problem area is a serious maintenance problem area based on the area size:

[0046] If MG>Q1, it means that there is a driving hazard when the vehicle passes through the problem area, and the problem area is determined to be a serious maintenance problem area;

[0047] If MG≤Q1, it means that it is unlikely for a vehicle to pose a driving hazard when passing through the problem area, and the problem area is determined to be a light maintenance problem area;

[0048] Among them, Q1 is the preset value.

[0049] A highway maintenance early warning and construction collaborative management platform, including:

[0050] Monitoring location acquisition module: used to obtain the highway monitoring location of the target area and install monitoring network equipment at the highway monitoring location;

[0051] Vehicle time division module: determines the benchmark passing vehicle value through time division and historical passing vehicles, and determines the target time period as a vehicle peak period, a vehicle off-peak period or a vehicle low-peak period according to the benchmark passing vehicle value;

[0052] Image frequency setting module: used for collecting images of highway monitoring locations at different image collection frequencies by monitoring network devices during peak hours, off-peak hours or off-peak hours;

[0053] Problem area identification module: used to acquire and analyze images collected at highway monitoring locations to identify problem areas in the images. Once the problem areas are identified, they are further divided into areas with severe maintenance problems or areas with mild maintenance problems.

[0054] Construction team scheduling module: Based on the identified areas with severe maintenance problems and areas with mild maintenance problems, construction teams are scheduled to carry out immediate maintenance at the monitoring locations of the highways in the areas with severe maintenance problems, and then construction teams are scheduled to carry out maintenance at the monitoring locations of the highways in the areas with mild maintenance problems within a set time.

[0055] The present invention provides a method and platform for highway maintenance early warning and construction collaborative management. Compared with existing technologies, it has the following advantages:

[0056] This invention utilizes modern monitoring technology, data analysis, and intelligent scheduling to significantly improve the efficiency and safety of highway maintenance. By installing efficient monitoring equipment at key locations on the highway and acquiring real-time data on vehicle flow and road conditions, maintenance issues such as cracks, wear, and potholes on the highway can be promptly discovered and accurately assessed. This intelligent monitoring method effectively avoids the lag and errors of traditional manual inspections, ensuring highway safety and the rapid resolution of road surface problems.

[0057] The present invention optimizes the adjustment of monitoring frequency by accurately dividing different time periods and combining them with traffic flow data. It can achieve high-frequency image acquisition during peak hours to comprehensively capture road surface problems, while conducting appropriate monitoring during off-peak hours, thereby reducing redundant data while ensuring that road surface problems can be identified in a timely and effective manner. In addition, through intelligent image analysis, the system can automatically identify problem areas and divide them into mild or severe maintenance problem areas according to their severity, ensuring the priority of maintenance work and the rational allocation of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The present invention will be further described below with reference to the accompanying drawings.

[0059] Figure 1 This is a flowchart of the steps of a method for highway maintenance early warning and construction collaborative management according to the present invention;

[0060] Figure 2 This is a structural framework diagram of a highway maintenance early warning and construction collaborative management platform according to the present invention. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0062] Example 1

[0063] See also Figure 1 ,The present invention provides a method for highway maintenance early warning and construction collaborative management, comprising;

[0064] Step 1: Obtain the highway monitoring locations in the target area and install monitoring network equipment at the highway monitoring locations;

[0065] The highway monitoring locations include highway bend locations, and highway uphill and downhill locations;

[0066] Specifically, monitoring network equipment will be installed at the bends and uphill and downhill locations of the highway in the target area to obtain information about passing vehicles. At the same time, the road surface conditions at the monitoring locations will be obtained and analyzed to promptly identify driving safety issues caused by unmaintained roads.

[0067] By installing radar vehicle detectors at appropriate locations to obtain information about passing vehicles, and by installing high-definition industrial cameras at appropriate locations to obtain road surface conditions at highway monitoring locations;

[0068] Step 2: Determine the benchmark passing vehicle value by dividing the time period and the historical passing vehicles, and determine whether the target time period is a peak vehicle period, a non-peak vehicle period, or a low vehicle period based on the benchmark passing vehicle value;

[0069] The specific method of determining the benchmark passing vehicle value by dividing the time period and the historical passing vehicles and determining whether the target time period is a peak time period, a non-peak time period or a low-peak time period according to the benchmark passing vehicle value is as follows:

[0070] AS1: Divide the current day evenly into x time periods, and take the first time period as the target time period for the following step analysis; where x is a preset value, specifically determined by professional staff. In this embodiment, x = 24;

[0071] AS2: Obtain the vehicle passing values for each day in the target time period in the previous n days, and mark them as Ci, where 1 ≤ i ≤ n. The previous n days mean n days before the current day; n is a preset value, specifically determined by professional staff. In this embodiment, n = 45;

[0072] AS3: Calculate the average value of the vehicle passing values Ci in the target time period in the previous n days, and denote it as Cp;

[0073] AS4: According to the average value Cp and the vehicle passing value Ci for each day, determine the deviation value Z of the corresponding n daily vehicle passing values Ci through the following formula:

[0074]

[0075] AS5: Obtain the deviation value Z and compare it with the preset value G1:

[0076] If Z ≤ G1, it means that the deviation value Z is within a reasonable range, and Cp is used as the reference vehicle passing value for the target time period;

[0077] If Z > G1, it means that the deviation value Z is too large, where the preset value G1 is set by professional staff;

[0078] AS6: When it is indicated that the deviation value Z is too large, perform data deletion on the vehicle passing values Ci for each day in the n days. The specific method of data deletion is:

[0079] Sort the vehicle passing values Ci for each day in the n days in descending order according to |Ci - Cp|, and each time obtain the Ci with the first rank in the sorting for deletion. After deletion, recalculate the deviation value Z of the remaining Ci until Z ≤ G1;

[0080] Then intercept the number of the deleted Ci values, and mark it as g1. Calculate the ratio of g1 to n, and mark it as gb1. Compare gb1 with the preset value G2:

[0081] If gb1 ≤ G2, calculate the average value of the remaining Ci as the reference vehicle passing value for the target time period;

[0082] If gb1 > G2, obtain the maximum value of Ci in the previous m days as the reference vehicle passing value for the target time period, where m < n. The previous m days mean m days before the current day; the value of m is determined by professional staff;

[0083] AS7: Determine whether the current target period is a peak period, a flat period, or a low period based on the benchmark passing vehicle value of the target period;

[0084] If the benchmark passing vehicle value exceeds the preset value G4, the target period is marked as the vehicle peak period;

[0085] If the benchmark passing vehicle value exceeds G3 but does not exceed the preset value G4, the target period is marked as a vehicle off-peak period;

[0086] If the benchmark vehicle passing value does not exceed the preset value G3, the target period is marked as a vehicle off-peak period;

[0087] The preset value G3 and the preset value G4 are both determined by professional staff, and G4>G3;

[0088] AS8: Taking the remaining time period of the current day as the target time period, repeating steps AS2-AS7 to determine whether the target time period of the current day is a peak time period, a non-peak time period, or a low-peak time period;

[0089] By dividing a day into multiple time periods and calculating the baseline number of vehicles passing through based on historical vehicle data, the traffic flow conditions for each time period can be scientifically determined. This data-driven approach can accurately determine whether a particular time period is peak, off-peak, or low-peak.

[0090] Step 3: At the highway monitoring location, images are collected using different image collection frequencies using monitoring network equipment during peak hours, off-peak hours, or low-peak hours.

[0091] The specific method of collecting images of the highway monitoring location at peak hours, off-peak hours, or low-peak hours by using the monitoring network device with different image collection frequencies is as follows:

[0092] When the highway monitoring location is in peak hours: road surface image acquisition is performed every time interval T1;

[0093] When the highway monitoring location is in peak hours: road surface image acquisition is performed every time interval T2;

[0094] When the highway monitoring location is in the off-peak period: road surface image acquisition is performed every time interval T3;

[0095] Among them, T1, T2 and T3 are all preset values, and T3>T2>T1, which are specifically determined by professional staff. In this embodiment, they are T3=10min, T2=5min, and T1=1min respectively;

[0096] It should be noted that during peak hours when traffic is heavy, high-frequency image acquisition is required to capture the road surface conditions at the highway monitoring location. Subsequent analysis of the high-frequency acquired images can provide a more comprehensive understanding of the road surface conditions at the highway monitoring location to determine whether there are any maintenance issues that affect vehicle traffic during peak hours.

[0097] According to the changes in traffic flow at different times, different image acquisition frequencies are used to more accurately monitor road conditions. During peak traffic hours, due to the large volume of traffic, the damage caused by vehicles to the road surface increases, and areas with mild maintenance problems may quickly develop into areas with serious maintenance problems. At the same time, road surface problems will also affect the traffic efficiency of vehicles. Therefore, in order to improve the efficiency of maintenance warnings, the image acquisition frequency needs to be increased to capture more road surface details and identify potential maintenance problems as early as possible. During off-peak hours, due to the low volume of traffic, the frequency of image acquisition can be reduced, thereby reducing data redundancy and improving processing efficiency. This strategy helps to improve the adaptability of the monitoring system under different traffic conditions and ensure that road surface monitoring work can be carried out efficiently and accurately at any time.

[0098] Step 4: Acquire and analyze images collected at the highway monitoring location to identify problem areas in the images of the highway monitoring location. Once the problem areas are identified, further classify the problem areas into severe maintenance problem areas or mild maintenance problem areas.

[0099] The specific method of acquiring and analyzing images of the highway monitoring location to determine problem areas in the images of the highway monitoring location and further classifying the problem areas into serious maintenance problem areas or minor maintenance problem areas is as follows:

[0100] BS1: Obtain images of highway monitoring locations and use object detection technology (such as YOLO, Faster R-CNN, and other deep learning algorithms) to detect and annotate vehicle locations and areas in the images.

[0101] BS2: For the marked vehicle area, the vehicle area is separated from the road surface area through image segmentation technology, the vehicle information in the image is eliminated, and only the road surface part of the highway monitoring position is retained;

[0102] BS3: Process images from highway monitoring locations for crack detection, wear detection, and pothole detection;

[0103] Specifically, crack detection uses a convolutional neural network (CNN) model to analyze images processed at highway monitoring locations to detect pavement cracks. The model can identify different types of cracks (such as horizontal cracks, longitudinal cracks, and network cracks) by learning from historical data.

[0104] Wear detection: Image processing techniques (such as edge detection and texture analysis) are used to analyze road wear in images processed at highway monitoring locations. Wear areas are typically characterized by decreased surface flatness, color changes, and fluctuations in road surface roughness. The model can identify these features and automatically mark them as wear areas.

[0105] Pothole detection: Deep learning models (such as Mask R-CNN) are used to detect potholes and uneven road surfaces in images processed from highway monitoring locations. Potholes typically have varying depths and sizes, and image analysis can identify them based on surface texture and height differences.

[0106] The above crack detection, wear detection and pothole detection are achieved through existing image analysis technology, which will not be described in detail here;

[0107] BS4: When cracks, wear, or potholes are detected in the processed images of highway monitoring locations, the areas with the problematic images are marked for maintenance.

[0108] BS5: If maintenance marking is performed on the same area of the processed image at the highway monitoring location three times in a row, the area is determined to be a problem area;

[0109] Specifically, the processed image of the highway monitoring location is divided into preset grids. If a grid is marked for maintenance three times in a row, the grid is determined to be a problem area.

[0110] It should be noted that during peak vehicle traffic hours, each time an image of the highway monitoring location is acquired, a large area of the image may be occupied by vehicles. Therefore, the occupied areas cannot be used for analysis, and only the unoccupied areas are subject to maintenance inspection. After these areas have been inspected, maintenance-marked areas can be determined. When images of the highway monitoring location are subsequently acquired, the maintenance-marked areas may be occupied by vehicles and cannot be analyzed again. Step BS5 mentions that if the same area of the highway monitoring location image is subsequently marked for maintenance three times in a row, then the area is determined to be a problem area. This is the maintenance-marked area determined by performing maintenance inspections on the same area three times in a row when images of the highway monitoring location are acquired multiple times. The maintenance-marked area is determined to be a problem area.

[0111] BS6: Obtain the area of the problem area, recorded as MG, and determine whether the problem area is a serious maintenance problem area based on the area size:

[0112] If MG>Q1, it means that there is a driving hazard when the vehicle passes through the problem area, and the problem area is determined to be a serious maintenance problem area;

[0113] If MG≤Q1, it means that it is unlikely for a vehicle to pose a driving hazard when passing through the problem area, and the problem area is determined to be a light maintenance problem area;

[0114] Q1 is the preset value, which will be determined by professional staff;

[0115] Step 5: Based on the identified areas with severe maintenance problems and areas with mild maintenance problems, arrange for construction teams to immediately carry out maintenance at the monitoring locations of the highways in the areas with severe maintenance problems, and then arrange for construction teams to carry out maintenance at the monitoring locations of the highways in the areas with mild maintenance problems within a set time.

[0116] Through detection and analysis, it is possible to clearly identify areas on the highway with serious maintenance problems and areas with minor maintenance problems. For these problem areas, the system can prioritize the nearest construction team for maintenance. Areas with serious problems will be given priority to ensure that areas that may pose a threat to vehicle safety are repaired as soon as possible. For areas with minor problems, corresponding maintenance work will be arranged according to traffic flow and road conditions. Through this optimized construction team scheduling mechanism, the response speed and efficiency of maintenance work can be improved, resource waste can be avoided, and the long-term safety and traffic capacity of the highway can be ensured.

[0117] Example 2

[0118] In the specific implementation process, this embodiment is based on the embodiment 1, and the difference from the embodiment 1 is that, refer to Figure 2 This embodiment also includes a highway maintenance early warning and construction collaborative management platform, which is composed of the following modules, including:

[0119] Monitoring location acquisition module: used to obtain the highway monitoring location of the target area and install monitoring network equipment at the highway monitoring location;

[0120] Vehicle time division module: determines the benchmark passing vehicle value through time division and historical passing vehicles, and determines the target time period as a vehicle peak period, a vehicle off-peak period or a vehicle low-peak period according to the benchmark passing vehicle value;

[0121] Image frequency setting module: used for collecting images of highway monitoring locations at different image collection frequencies by monitoring network devices during peak hours, off-peak hours or off-peak hours;

[0122] Problem area identification module: used to acquire and analyze images collected at highway monitoring locations to identify problem areas in the images. Once the problem areas are identified, they are further divided into areas with severe maintenance problems or areas with mild maintenance problems.

[0123] Construction team scheduling module: Based on the identified areas with severe maintenance problems and areas with mild maintenance problems, construction teams are scheduled to carry out immediate maintenance at the monitoring locations of the highways in the areas with severe maintenance problems, and then construction teams are scheduled to carry out maintenance at the monitoring locations of the highways in the areas with mild maintenance problems within a set time.

[0124] Example 3

[0125] The specific implementation process of this embodiment includes the entire implementation process of the above two groups of embodiments.

[0126] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0127] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for highway maintenance early warning and construction collaborative management, characterized in that: Including: Step 1: Obtain the highway monitoring locations in the target area and install monitoring networking devices at the highway monitoring locations; Step 2: Determine the reference vehicle passing value through time period division and historical passing vehicles, and determine whether the target time period is a vehicle peak time period, a vehicle flat peak time period, or a vehicle low peak time period according to the reference vehicle passing value; Step 3: At the highway monitoring locations, in the peak time period, flat peak time period, or low peak time period respectively, collect images of the highway monitoring locations through the monitoring networking devices at different image acquisition frequencies; Step 4: Obtain the images collected at the highway monitoring locations and analyze them to determine the problem areas in the images of the highway monitoring locations. When the problem areas are determined, further divide the problem areas into serious maintenance problem areas or minor maintenance problem areas; Step 5: Based on the determined serious maintenance problem areas and minor maintenance problem areas, arrange a construction team to immediately carry out construction maintenance at the highway monitoring locations where the serious maintenance problem areas are located, and then arrange a construction team to carry out construction maintenance at the highway monitoring locations where the minor maintenance problem areas are located within a set time.

2. A method for highway maintenance early warning and construction collaborative management according to claim 1, characterized in that: In the above Step 1, the highway monitoring locations include highway bend locations and highway uphill and downhill locations.

3. A method for highway maintenance early warning and construction collaborative management according to claim 2, characterized in that: The specific content of determining the reference vehicle passing value through time period division and historical passing vehicles, and determining whether the target time period is a vehicle peak time period, a vehicle flat peak time period, or a vehicle low peak time period according to the reference vehicle passing value is as follows: AS1: Divide the current day evenly into x time periods, and take the first time period as the target time period for the following step analysis; where x is a preset value; AS2: Obtain the vehicle passing values during the target time period for each of the previous n days, and mark them as Ci, where 1 ≤ i ≤ n, and the previous n days means n days before the current day; n is a preset value; AS3: Calculate the average value of the vehicle passing values Ci during the target time period for the previous n days, and denote it as Cp; AS4: Determine the deviation value Z according to the average value Cp and the daily vehicle passing value Ci through the standard deviation formula; AS5: Obtain the deviation value Z and compare it with the preset value G1: If Z ≤ G1, it means that the deviation value Z is within a reasonable range, and take Cp as the reference vehicle passing value for the target time period; If Z > G1, it means that the deviation value Z is偏大; AS6: When it is indicated that the deviation value Z is偏大, perform data deletion on the daily vehicle passing values Ci for n days. The specific method of data deletion is as follows: Sort the daily vehicle passing values Ci for n days in descending order according to |Ci - Cp|, each time obtain the Ci with the first ranking for deletion, and recalculate the deviation value Z of the remaining Ci until Z ≤ G1; Then intercept the number of deleted Ci values, and mark it as g1, calculate the ratio of g1 to n, and mark it as gb1, and compare gb1 with the preset value G2: If gb1 ≤ G2, calculate the average value of the remaining Ci as the reference vehicle passing value for the target time period; If gb1 > G2, obtain the maximum value of Ci for the previous m days as the reference vehicle passing value for the target time period, where m < n, and the previous m days means m days before the current day; AS7: Determine whether the current target period is a peak period, a flat period, or a low period based on the benchmark passing vehicle value of the target period; AS8: Taking the remaining time period of the current day as the target time period, repeat steps AS2-AS7 to determine whether the target time period of the current day is a peak time period, a non-peak time period, or a low-peak time period.

4. A method for highway maintenance early warning and construction collaborative management according to claim 3, characterized in that: In step AS7, the specific method of determining the target time period as a peak time period, a non-peak time period or a low-peak time period according to the reference number of vehicles passing through is as follows: If the benchmark passing vehicle value exceeds the preset value G4, the target period is marked as the vehicle peak period; If the benchmark passing vehicle value exceeds G3 but does not exceed the preset value G4, the target period is marked as a vehicle off-peak period; If the benchmark vehicle passing value does not exceed the preset value G3, the target period is marked as a vehicle off-peak period; Among them, G4>G3.

5. The method for highway maintenance early warning and construction collaborative management according to claim 1, characterized in that: In step 3, the specific method of collecting images of the highway monitoring location at peak hours, off-peak hours, or low-peak hours by using the monitoring network device at different image collection frequencies is as follows: When the highway monitoring location is in peak hours: road surface image acquisition is performed every time interval T1; When the highway monitoring location is in peak hours: road surface image acquisition is performed every time interval T2; When the highway monitoring location is in the off-peak period: road surface image acquisition is performed every time interval T3; Among them, T1, T2 and T3 are all preset values, and T3>T2>T1.

6. A method for highway maintenance early warning and construction collaborative management according to claim 1, characterized in that: In step 4, the image of the highway monitoring location is acquired and analyzed to determine the problem area in the image of the highway monitoring location. When the problem area is determined, the problem area is further divided into a serious maintenance problem area or a minor maintenance problem area. The specific content is: BS1: Acquire images of highway monitoring locations and use object detection technology to detect and annotate vehicle locations and areas in the images; BS2: For the marked vehicle area, the vehicle area is separated from the road surface area through image segmentation technology, the vehicle information in the image is eliminated, and only the road surface part of the highway monitoring position is retained; BS3: Process images from highway monitoring locations for crack detection, wear detection, and pothole detection; BS4: When cracks, wear, or potholes are detected in the processed images of highway monitoring locations, the areas with the problematic images are marked for maintenance. BS5: If maintenance marking is performed on the same area of the processed image at the highway monitoring location three times in a row, the area is determined to be a problem area.

7. A method for highway maintenance early warning and construction collaborative management according to claim 6, characterized in that: Also includes: BS6: Obtain the area of the problem area, recorded as MG, and determine whether the problem area is a serious maintenance problem area based on the area size: If MG>Q1, it means that there is a driving hazard when the vehicle passes through the problem area, and the problem area is determined to be a serious maintenance problem area; If MG≤Q1, it means that it is unlikely for a vehicle to pose a driving hazard when passing through the problem area, and the problem area is determined to be a light maintenance problem area; Where Q1 is the preset value.

8. A highway maintenance early warning and construction collaborative management platform, implemented in a highway maintenance early warning and construction collaborative management method according to any one of claims 1 to 7, characterized in that: include: Monitoring location acquisition module: used to obtain the highway monitoring location of the target area and install monitoring network equipment at the highway monitoring location; Vehicle time division module: determines the benchmark passing vehicle value through time division and historical passing vehicles, and determines the target time period as a vehicle peak period, a vehicle off-peak period or a vehicle low-peak period according to the benchmark passing vehicle value; Image frequency setting module: used for collecting images of highway monitoring locations at different image collection frequencies by monitoring network devices during peak hours, off-peak hours or off-peak hours; Problem area identification module: used to acquire and analyze images collected at highway monitoring locations to identify problem areas in the images. Once the problem areas are identified, they are further divided into areas with severe maintenance problems or areas with mild maintenance problems. Construction team scheduling module: Based on the identified areas with severe maintenance problems and areas with mild maintenance problems, construction teams are scheduled to carry out immediate maintenance at the monitoring locations of the highways in the areas with severe maintenance problems, and then construction teams are scheduled to carry out maintenance at the monitoring locations of the highways in the areas with mild maintenance problems within a set time.

Images