Key road section road condition intelligent analysis system and method based on low-altitude remote sensing

Abstract

The invention relates to the technical field of data analysis, and discloses a key road section road condition intelligent analysis system and method based on low-altitude remote sensing. According to the method, road image historical data of a to-be-detected area are collected, the collected road image historical data are processed, a regional traffic network is constructed based on the processed road image historical data at the same time, and after construction is completed, based on the constructed regional traffic network, the traffic information of the to-be-detected area is acquired. According to the method, the traffic condition of a to-be-detected area road is monitored in real time in a low-altitude remote sensing mode, meanwhile, analysis is carried out in a data analysis mode based on traffic condition data obtained through real-time monitoring, and finally a road loss prediction model is constructed based on the obtained analyzed traffic condition data and processed road image historical data. And the collected data is intelligently analyzed based on the road loss prediction model, so that the intelligence of road condition analysis is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a system and method for intelligently analyzing road conditions of key sections based on low-altitude remote sensing. Background Art

[0002] With the rapid development of my country's highways, the vehicle traffic capacity of highways has been greatly improved, but it has caused more damage to road conditions. Excessive road wear and tear will lead to traffic congestion and increase the incidence of accidents, but at present, due to the long monitoring cycle of highway conditions, the repair of road infrastructure is slow.

[0003] The existing patent application CN110866081A, the invention analyzes the Internet map, extracts the congestion tile map, obtains the road condition distribution map; at the same time, the tile map is graded and data analysis is performed; at the same time, the coordinate system is transformed and the congestion tile map is image analyzed; at the same time, the analyzed congestion map is thinned and backboned; and then the congestion point is reversed and the congestion direction is analyzed. However, since only the congestion and location analysis of the road section is considered, the road condition is ignored, which has certain limitations. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the shortcomings of the existing technology, the present invention provides a system and method for intelligent analysis of road conditions in key sections based on low-altitude remote sensing, which has the advantages of high efficiency, accuracy, and real-time, and solves the problem that the highway road condition monitoring cycle is too long, resulting in slow repair of road infrastructure.

[0005] (II) Technical solution In order to solve the above-mentioned technical problems of poor timeliness and low efficiency of traditional video acquisition and visual interpretation, the present invention provides the following technical solutions: The present invention discloses a method for intelligently analyzing road conditions of key sections based on low-altitude remote sensing, which specifically comprises the following steps: S1. Collecting historical road image data of the area to be detected, and processing the collected historical road image data to obtain processed historical road image data; S2. Constructing a regional transportation network based on the processed road image historical data; S3. Based on the constructed regional traffic network, the traffic conditions of the roads in the detection area are monitored in real time through low-altitude remote sensing; S4. Analyze the traffic condition data obtained by real-time monitoring through a data analysis method to obtain analyzed traffic condition data; S41, processing the traffic condition data obtained by real-time monitoring to obtain processed traffic condition data; S42, calculating the average speed, traffic volume and traffic density of the road section based on the processed traffic condition data; S43, analyzing the traffic conditions based on the calculated average speed, traffic volume and traffic density of the road section, and outputting the analyzed traffic condition data; S5. Construct a road loss prediction model based on the analyzed traffic condition data and the processed road image history data, and perform intelligent analysis on the collected data based on the road loss prediction model.

[0006] The present invention collects historical road image data of the area to be detected, processes the collected historical road image data, and constructs a regional traffic network based on the processed historical road image data. After the construction is completed, based on the constructed regional traffic network, the traffic conditions of the roads in the area to be detected are monitored in real time through a low-altitude remote sensing method. At the same time, based on the traffic condition data obtained by real-time monitoring, the data is analyzed through a data analysis method. Finally, a road loss prediction model is constructed based on the analyzed traffic condition data and the processed historical road image data, and the collected data is intelligently analyzed based on the road loss prediction model, thereby improving the intelligence of the road condition analysis.

[0007] Preferably, the collecting of the road image historical data of the area to be detected and processing the collected road image historical data to obtain the processed road image historical data comprises the following steps: The road image historical data includes: road images, road geographical locations and road vehicle traffic volume; Sort the collected road image history data by time; Set monitoring cycles and record the number of vehicles passing through the road during each monitoring cycle; The historical data of the road image in the first cycle is set as the benchmark, and the road damage index of each cycle is calculated; The calculation formula of the pavement damage index is as follows: ; ; in, Indicates Damaged area of ​​damaged road surface; is the product of the actual road length and the effective road width. Indicates The weight of the pavement damage class, Indicates existence Item damage type, Represents the road damage index. Indicates the damage rate of the road surface of the current section; Calculate the pavement damage index caused by each vehicle based on the changes in the pavement damage index in each period and the number of vehicles passing through the road; The processed road image historical data is set to include: road damage condition index, road image, road geographical location and road vehicle traffic volume.

[0008] The present invention collects historical road image data and calculates a road damage condition index, and preliminarily judges the road condition based on the calculated road damage condition index, thereby improving the rationality of road condition analysis.

[0009] Preferably, the construction of a regional transportation network based on the processed road image historical data comprises the following steps: Based on the geographical location of the roads in the processed road image historical data, the road segment data set in the set area is: ; in, Represents a collection of road segment data within the area. Represents the mth road section data in the area; Road segment data By a triple Indicates that Indicates the starting point of the road segment data. Indicates the end point of the road segment data. Indicates the distance of the road segment; The road segment data set within the region is aggregated to construct a regional transportation network.

[0010] The present invention sets and records the road segment data in the area by processing the road geographical location in the historical data of the road image, and at the same time summarizes the road segment data in the area to construct a regional traffic network, thereby improving the efficiency of road condition analysis.

[0011] Preferably, the real-time monitoring of the traffic conditions of the roads in the detection area by low-altitude remote sensing based on the constructed regional traffic network includes the following steps: S31, setting a reference point in the area to be detected; Based on the constructed regional transportation network, the intersection of the main road and the non-main road is set as the reference point; S32, based on all the reference points in the area to be detected, number each reference point by setting a UID and save it; S33, establishing a three-dimensional coordinate system based on the first reference point, recording and saving the coordinate positions of the reference points corresponding to the numbers; Through low-altitude remote sensing, multiple groups of drones are arranged for real-time monitoring based on the recorded coordinate positions of corresponding reference points.

[0012] The present invention sets reference points in the area to be detected, and based on all the reference points in the area to be detected, each reference point is numbered by setting a UID, and at the same time a three-dimensional coordinate system is constructed and multiple groups of drones are arranged for real-time monitoring based on the coordinate positions of the corresponding reference points through low-altitude remote sensing, thereby improving the reliability of road condition monitoring data.

[0013] Preferably, the processing of the traffic condition data obtained by real-time monitoring to obtain the processed traffic condition data comprises the following steps: Set monitoring intervals and record vehicle data on the corresponding road section based on real-time monitoring results; Remove the vehicle data with a speed of 0 in the corresponding road section; The removed vehicle data is set as processed traffic condition data.

[0014] Preferably, the calculating the average speed, traffic volume and traffic density of a road section based on the processed traffic condition data comprises the following steps: The formula for calculating the average speed of a road section is as follows: ; in, represents the average speed of the road section, It indicates the number of vehicles passing through the road section per unit time. Indicates The instantaneous speed of the vehicle; The traffic volume calculation formula is as follows: ; in, It represents the traffic volume of the road section per unit time. Indicates time; The traffic density calculation formula is as follows: ; in, Represents the traffic density in the road segment.

[0015] Preferably, the evaluating the traffic condition based on the calculated average speed, traffic volume and traffic density of the road section and setting the evaluation result as the analyzed traffic condition data comprises the following steps: Since the faster the speed, the better the traffic condition of the road section, the average speed of the road section is set to be proportional to the traffic condition data, the larger the traffic volume, the slower the speed, and the traffic volume is set to be inversely proportional to the traffic condition data, and the larger the traffic density, the slower the speed, and the traffic density is set to be inversely proportional to the traffic condition data; Traffic Condition Data = ×Average speed of the section+ × Traffic volume+ × Traffic density; in, , , Represents the weight coefficient.

[0016] The present invention processes the traffic condition data obtained through real-time monitoring, calculates the average speed, traffic volume and traffic density of the road section, and evaluates the traffic condition based on the calculation results, thereby improving the accuracy of road condition analysis.

[0017] Preferably, the step of constructing a road loss prediction model based on the obtained analyzed traffic condition data and the processed road image history data, and performing intelligent analysis on the collected data based on the road loss prediction model comprises the following steps: The structure of BP neural network includes: input layer, hidden layer and output layer; Set the input layer to ,in Represents the input road image history data of the first group of road segments, represents the input road image history data of the mth group of road sections, and the output of the output layer is ,in represents the output road loss prediction data for the first group of road sections, It represents the output road loss prediction data of the sth group of road sections, and the hidden layer contains q neurons; Construct a road loss prediction model based on the forward propagation process of BP neural network signals; The forward propagation of the signal is that the data is input from the input layer, passes through the hidden layer, and finally reaches the output layer; The calculation formula for the forward propagation of the signal is: The calculation formula for transferring parameter data from the input layer to the hidden layer is: ; in, represents the input of the hth hidden layer neuron, represents the weight of the i-th input of the input layer to the h-th hidden layer neuron, and M represents the number of input data of the input layer; Represents the i-th input of the input layer; Represents the bias value of the hth hidden layer neuron; The calculation formula for transferring parameter data from the hidden layer to the output layer is: ; in, represents the input of the jth output layer neuron, represents the weight from the hth hidden layer neuron to the jth output layer neuron, and q represents the number of hidden layer neurons; represents the output of the hth hidden layer neuron; Represents the bias value of the jth output layer neuron; After receiving the parameter data transmitted by the hidden layer, the output layer directly outputs the parameter data received from the hidden layer; Right now , represents the output of the jth output layer neuron; Calculate the error between the output layer and the expected value, set the error threshold between the output layer and the expected value, and when the error between the output layer and the expected value is greater than or equal to the set error threshold, adjust the weight in a step-by-step manner, and use the adjusted weight as the new weight in the forward propagation calculation of the signal for iterative calculation until the data converges to obtain a road loss prediction model; S52. Perform intelligent analysis on the collected data based on the road loss prediction model.

[0018] Preferably, the intelligent analysis of the collected data based on the road loss prediction model comprises the following steps: The road image data of the area to be detected collected in the previous cycle and the current traffic condition data are input into a road loss prediction model to predict the road loss of the area to be detected in the next week; Set a road loss threshold. When the road loss in the detection area expected in the next week exceeds the set threshold, report it and inspect the road.

[0019] The present invention constructs a road loss prediction model through the obtained analyzed traffic condition data and processed road image history data through the BP neural network continuous iteration method, and at the same time, intelligently analyzes the collected data based on the road loss prediction model to improve the efficiency of road condition analysis.

[0020] The present invention also discloses a key road section intelligent road condition analysis system based on low-altitude remote sensing, which is used to implement a key road section intelligent road condition analysis method based on low-altitude remote sensing. The system includes: a road data acquisition module, a traffic planning module, a low-altitude remote sensing monitoring module, a traffic condition module and a road loss prediction module; The road data acquisition module is used to collect road image data of the area to be detected; The traffic planning module is used to construct a traffic network based on the collected road image data; The low-altitude remote sensing monitoring module is used to arrange drones to monitor the road in real time; The traffic condition module is used to analyze the traffic condition based on the real-time monitoring data of the low-altitude remote sensing monitoring module; The road loss prediction module is used to predict road loss based on road image data and traffic condition data.

[0021] (III) Beneficial effects Compared with the prior art, the present invention provides a system and method for intelligent analysis of key road conditions based on low-altitude remote sensing, which has the following beneficial effects: 1. The invention collects historical road image data of the area to be detected, processes the collected historical road image data, and constructs a regional traffic network based on the processed historical road image data. After the construction is completed, based on the constructed regional traffic network, the traffic conditions of the roads in the detection area are monitored in real time by low-altitude remote sensing. At the same time, the traffic condition data obtained by real-time monitoring is analyzed by data analysis. Finally, a road loss prediction model is constructed based on the analyzed traffic condition data and the processed historical road image data, and the collected data is intelligently analyzed based on the road loss prediction model, thereby improving the intelligence of road condition analysis.

[0022] 2. The invention collects historical road image data and calculates a road damage condition index, and makes a preliminary judgment on the road condition based on the calculated road damage condition index, thereby improving the rationality of road condition analysis.

[0023] 3. The invention processes the geographical location of roads in the historical data of road images, sets and records the road section data in the area, and at the same time aggregates the road section data in the area to build a regional traffic network, thereby improving the efficiency of road condition analysis.

[0024] 4. The invention sets reference points in the area to be detected, and based on all the reference points in the area to be detected, each reference point is numbered by setting UID, and at the same time, a three-dimensional coordinate system is constructed and multiple groups of drones are arranged for real-time monitoring based on the coordinate positions of the corresponding reference points through low-altitude remote sensing, thereby improving the reliability of road condition monitoring data.

[0025] 5. The invention processes the traffic condition data obtained from real-time monitoring, calculates the average speed, traffic volume and traffic density of the road section, and evaluates the traffic condition based on the calculation results, thereby improving the accuracy of road condition analysis.

[0026] 6. The invention constructs a road loss prediction model by continuously iterating the BP neural network through the analyzed traffic condition data and the processed road image history data. At the same time, the collected data is intelligently analyzed based on the road loss prediction model to improve the efficiency of road condition analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the structure of the intelligent analysis process of key road sections of the present invention; DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0029] Example 1 See also Figure 1 This embodiment discloses a method for intelligent analysis of key road conditions based on low-altitude remote sensing, which specifically includes the following steps: S1. Collecting historical road image data of the area to be detected, and processing the collected historical road image data to obtain processed historical road image data; S2. Constructing a regional transportation network based on the processed road image historical data; S3. Based on the constructed regional traffic network, the traffic conditions of the roads in the detection area are monitored in real time through low-altitude remote sensing; S4. Analyze the traffic condition data obtained by real-time monitoring through a data analysis method to obtain analyzed traffic condition data; S41, processing the traffic condition data obtained by real-time monitoring to obtain processed traffic condition data; S42, calculating the average speed, traffic volume and traffic density of the road section based on the processed traffic condition data; S43, analyzing the traffic conditions based on the calculated average speed, traffic volume and traffic density of the road section, and outputting the analyzed traffic condition data; S5. constructing a road loss prediction model based on the analyzed traffic condition data and the processed road image history data, and performing intelligent analysis on the collected data based on the road loss prediction model; For further information, see Figure 1 , collecting the road image history data of the area to be detected, and processing the collected road image history data, and obtaining the processed road image history data includes the following steps: The road image historical data includes: road images, road geographical locations and road vehicle traffic volume; Sort the collected road image history data by time; Furthermore, a monitoring period is set, and the number of vehicles passing through the road in each monitoring period is recorded; The historical data of the road image in the first cycle is set as the benchmark, and the road damage index of each cycle is calculated; The calculation formula of the pavement damage index is as follows: ; ; in, Indicates The damaged area of ​​the damaged road surface, is the product of the actual road length and the effective road width. Indicates The weight of the pavement damage class, Indicates existence Item damage type, Represents the road damage index. Indicates the damage rate of the road surface of the current section; Calculate the pavement damage index caused by each vehicle based on the changes in the pavement damage index in each period and the number of vehicles passing through the road; Furthermore, the processed road image historical data is set to include: road damage condition index, road image, road geographical location and road vehicle traffic volume; For further information, see Figure 1 Based on the processed road image historical data, building a regional transportation network includes the following steps: Based on the geographical location of the roads in the processed road image historical data, the road segment data set in the set area is: ; in, Represents a collection of road segment data within the area. Represents the mth road section data in the area; Road segment data By a triple Indicates that Indicates the starting point of the road segment data. Indicates the end point of the road segment data. Indicates the distance of the road segment; Furthermore, the road segment data set in the region is aggregated to construct a regional transportation network; For further information, see Figure 1 Based on the constructed regional traffic network, real-time monitoring of the traffic conditions of the roads in the detection area through low-altitude remote sensing includes the following steps: S31, setting a reference point in the area to be detected; Based on the constructed regional transportation network, the intersection of the main road and the non-main road is set as the reference point; S32, based on all the reference points in the area to be detected, number each reference point by setting a UID and save it; S33, establishing a three-dimensional coordinate system based on the first reference point, recording and saving the coordinate positions of the reference points corresponding to the numbers; Arrange multiple groups of drones for real-time monitoring based on the coordinate positions of corresponding reference points through low-altitude remote sensing; For further information, see Figure 1 , processing the traffic condition data obtained by real-time monitoring, and obtaining the processed traffic condition data includes the following steps: Set monitoring intervals and record vehicle data on the corresponding road section based on real-time monitoring results; Remove the vehicle data with a speed of 0 in the corresponding road section; The removed vehicle data is set as processed traffic condition data; For further information, see Figure 1 Calculating the average speed, traffic volume and traffic density of a road section based on the processed traffic condition data includes the following steps: The formula for calculating the average speed of a road section is as follows: ; in, represents the average speed of the road section, It indicates the number of vehicles passing through the road section per unit time. Indicates The instantaneous speed of the vehicle; The traffic volume calculation formula is as follows: ; in, It represents the traffic volume of the road section per unit time. Indicates time; The traffic density calculation formula is as follows: ; in, Indicates the traffic density in the road segment; For further information, see Figure 1 , based on the calculated average speed, traffic volume and traffic density of the road section, the traffic condition is evaluated, and setting the evaluation result as the analyzed traffic condition data includes the following steps: Since the faster the speed, the better the traffic condition of the road section, the average speed of the road section is set to be proportional to the traffic condition data, the larger the traffic volume, the slower the speed, and the traffic volume is set to be inversely proportional to the traffic condition data, and the larger the traffic density, the slower the speed, and the traffic density is set to be inversely proportional to the traffic condition data; Traffic Condition Data = ×Average speed of the section+ × Traffic volume+ × Traffic density; in, , , represents the weight coefficient; For further information, see Figure 1 , building a road loss prediction model based on the analyzed traffic condition data and the processed road image history data, and intelligently analyzing the collected data based on the road loss prediction model includes the following steps: S51, constructing a road loss prediction model through BP neural network; The structure of BP neural network includes: input layer, hidden layer and output layer; Set the input layer to ,in Represents the input road image history data of the first group of road segments, represents the input road image history data of the mth group of road sections, and the output of the output layer is ,in represents the output road loss prediction data for the first group of road sections, It represents the output road loss prediction data of the sth group of road sections, and the hidden layer contains q neurons; Construct a road loss prediction model based on the forward propagation process of BP neural network signals; The forward propagation of the signal is that the data is input from the input layer, passes through the hidden layer, and finally reaches the output layer; The calculation formula for the forward propagation of the signal is: The calculation formula for transferring parameter data from the input layer to the hidden layer is: ; in, represents the input of the hth hidden layer neuron, represents the weight of the i-th input of the input layer to the h-th hidden layer neuron, and M represents the number of input data of the input layer; Represents the i-th input of the input layer; Represents the bias value of the hth hidden layer neuron; The calculation formula for transferring parameter data from the hidden layer to the output layer is: ; in, represents the input of the jth output layer neuron, represents the weight from the hth hidden layer neuron to the jth output layer neuron, and q represents the number of hidden layer neurons; represents the output of the hth hidden layer neuron; Represents the bias value of the jth output layer neuron; After receiving the parameter data transmitted by the hidden layer, the output layer directly outputs the parameter data received from the hidden layer; Right now , represents the output of the jth output layer neuron; Calculate the error between the output layer and the expected value, set the error threshold between the output layer and the expected value, and when the error between the output layer and the expected value is greater than or equal to the set error threshold, adjust the weight in a step-by-step manner, and use the adjusted weight as the new weight in the forward propagation calculation of the signal for iterative calculation until the data converges to obtain a road loss prediction model; S52, performing intelligent analysis on the collected data based on the road loss prediction model; The road image data of the area to be detected collected in the previous cycle and the current traffic condition data are input into a road loss prediction model to predict the road loss of the area to be detected in the next week; Set a road loss threshold. When the road loss in the detection area expected to be detected in the next week exceeds the set threshold, report it and conduct road maintenance. Example 2 See also Figure 1 ,This embodiment also discloses a key road section intelligent road condition analysis system based on low-altitude remote sensing, which is used to implement a key road section intelligent road condition analysis method based on low-altitude remote sensing. The system includes: a road data acquisition module, a traffic planning module, a low-altitude remote sensing monitoring module, a traffic condition module, and a road loss prediction module; The road data acquisition module is used to collect road image data of the area to be detected; The traffic planning module is used to construct a traffic network based on the collected road image data; The low-altitude remote sensing monitoring module is used to arrange drones to monitor the road in real time; The traffic condition module is used to analyze the traffic condition based on the real-time monitoring data of the low-altitude remote sensing monitoring module; The road loss prediction module is used to predict road loss based on road image data and traffic condition data.

[0030] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent analysis of key road conditions based on low-altitude remote sensing, characterized in that: The following steps are involved: S1. Collecting historical road image data of the area to be detected, and processing the collected historical road image data to obtain processed historical road image data; S2. Constructing a regional transportation network based on the processed road image historical data; S3. Based on the constructed regional traffic network, the traffic conditions of the roads in the detection area are monitored in real time through low-altitude remote sensing; S4. Analyze the traffic condition data obtained by real-time monitoring through a data analysis method to obtain analyzed traffic condition data; S41, processing the traffic condition data obtained by real-time monitoring to obtain processed traffic condition data; S42, calculating the average speed, traffic volume and traffic density of the road section based on the processed traffic condition data; S43, analyzing the traffic conditions based on the calculated average speed, traffic volume and traffic density of the road section, and outputting the analyzed traffic condition data; S5. Construct a road loss prediction model based on the analyzed traffic condition data and the processed road image history data, and perform intelligent analysis on the collected data based on the road loss prediction model.

2. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 1 is characterized in that: The collecting of the road image history data of the area to be detected and processing the collected road image history data to obtain the processed road image history data comprises the following steps: The road image historical data includes: road images, road geographical locations and road vehicle traffic; Sort the collected road image history data by time; Set monitoring cycles and record the number of vehicles passing through the road during each monitoring cycle; The historical data of the road image in the first cycle is set as the benchmark, and the road damage index of each cycle is calculated; The calculation formula of the pavement damage index is as follows: ； ； in, Indicates Damaged area of ​​damaged road surface; is the product of the actual road length and the effective road width. Indicates The weight of the pavement damage class, Indicates existence Item damage type, Represents the road damage index. Indicates the damage rate of the road surface of the current section; Calculate the pavement damage index caused by each vehicle based on the changes in the pavement damage index in each period and the number of vehicles passing through the road; The processed road image historical data is set to include: road damage condition index, road image, road geographical location and road vehicle traffic volume.

3. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 1 is characterized in that: The construction of a regional transportation network based on the processed road image historical data comprises the following steps: Based on the geographical location of the roads in the processed road image historical data, the road segment data set in the set area is: ; in, Represents a collection of road segment data within the area. Represents the mth road section data in the area; Road segment data By a triple Indicates that Indicates the starting point of the road segment data. Indicates the end point of the road segment data. Indicates the distance of the road segment; The road segment data set within the region is aggregated to construct a regional transportation network.

4. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 1 is characterized in that: The real-time monitoring of the traffic conditions of the roads in the detection area by low-altitude remote sensing based on the constructed regional traffic network includes the following steps: S31, setting a reference point in the area to be detected; Based on the constructed regional transportation network, the intersection of the main road and the non-main road is set as the reference point; S32, based on all the reference points in the area to be detected, number each reference point by setting a UID and save it; S33, establishing a three-dimensional coordinate system based on the first reference point, recording and saving the coordinate positions of the reference points corresponding to the numbers; Through low-altitude remote sensing, multiple groups of drones are arranged for real-time monitoring based on the recorded coordinate positions of corresponding reference points.

5. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 1 is characterized in that: The processing of the traffic condition data obtained by real-time monitoring to obtain the processed traffic condition data comprises the following steps: Set monitoring intervals and record vehicle data on the corresponding road section based on real-time monitoring results; Remove the vehicle data with a speed of 0 in the corresponding road section; The removed vehicle data is set as processed traffic condition data.

6. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 1 is characterized in that: The calculation of the average speed, traffic volume and traffic density of the road section based on the processed traffic condition data comprises the following steps: The formula for calculating the average speed of a road section is as follows: ； in, represents the average speed of the road section, Indicates the number of vehicles passing through the road section per unit time. Indicates The instantaneous speed of the vehicle; The traffic volume calculation formula is as follows: ； in, It represents the traffic volume of the road section per unit time. Indicates time; The traffic density calculation formula is as follows: ； in, Represents the traffic density in the road segment.

7. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 1 is characterized in that: The step of evaluating the traffic condition based on the calculated average speed, traffic volume and traffic density of the road section and setting the evaluation result as the analyzed traffic condition data comprises the following steps: Since the faster the speed, the better the traffic condition of the road section, the average speed of the road section is set to be proportional to the traffic condition data, the larger the traffic volume, the slower the speed, and the traffic volume is set to be inversely proportional to the traffic condition data, and the larger the traffic density, the slower the speed, and the traffic density is set to be inversely proportional to the traffic condition data; Traffic Condition Data = ×Average speed of the section+ × Traffic volume+ × Traffic density; in, , , Represents the weight coefficient.

8. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 1 is characterized in that: The road loss prediction model is constructed based on the analyzed traffic condition data and the processed road image history data, and the collected data is intelligently analyzed based on the road loss prediction model, including the following steps: S51, constructing a road loss prediction model through BP neural network; The structure of BP neural network includes: input layer, hidden layer and output layer; Set the input layer to ,in Represents the input road image history data of the first group of road segments, represents the input road image history data of the mth group of road sections, and the output of the output layer is ,in represents the output road loss prediction data for the first group of road sections, It represents the output road loss prediction data of the sth group of road sections, and the hidden layer contains q neurons; Construct a road loss prediction model based on the forward propagation process of BP neural network signals; The forward propagation of the signal is that the data is input from the input layer, passes through the hidden layer, and finally reaches the output layer; The calculation formula for the forward propagation of the signal is: The calculation formula for transferring parameter data from the input layer to the hidden layer is: ； in, represents the input of the hth hidden layer neuron, represents the weight of the i-th input of the input layer to the h-th hidden layer neuron, and M represents the number of input data of the input layer; Represents the i-th input of the input layer; Represents the bias value of the hth hidden layer neuron; The calculation formula for transferring parameter data from the hidden layer to the output layer is: ； in, represents the input of the jth output layer neuron, represents the weight from the hth hidden layer neuron to the jth output layer neuron, and q represents the number of hidden layer neurons; represents the output of the hth hidden layer neuron; Represents the bias value of the jth output layer neuron; After receiving the parameter data transmitted by the hidden layer, the output layer directly outputs the parameter data received from the hidden layer; Right now , represents the output of the jth output layer neuron; Calculate the error between the output layer and the expected value, set the error threshold between the output layer and the expected value, and when the error between the output layer and the expected value is greater than or equal to the set error threshold, adjust the weight in a step-by-step manner, and use the adjusted weight as the new weight in the forward propagation calculation of the signal for iterative calculation until the data converges to obtain a road loss prediction model; S52. Perform intelligent analysis on the collected data based on the road loss prediction model.

9. The method for intelligent analysis of key road conditions based on low-altitude remote sensing according to claim 8 is characterized in that: The intelligent analysis of the collected data based on the road loss prediction model comprises the following steps: The road image data of the area to be detected collected in the previous cycle and the current traffic condition data are input into a road loss prediction model to predict the road loss of the area to be detected in the next week; Set a road loss threshold. When the road loss in the detection area expected in the next week exceeds the set threshold, report it and inspect the road.

10. A system for implementing the intelligent analysis method of key road conditions based on low-altitude remote sensing according to any one of claims 1 to 9, characterized in that: include: Road data acquisition module, traffic planning module, low-altitude remote sensing monitoring module, traffic status module and road loss prediction module; The road data acquisition module is used to collect road image data of the area to be detected; The traffic planning module is used to construct a traffic network based on the collected road image data; The low-altitude remote sensing monitoring module is used to arrange drones to monitor the road in real time; The traffic condition module is used to analyze the traffic condition based on the real-time monitoring data of the low-altitude remote sensing monitoring module; The road loss prediction module is used to predict road loss based on road image data and traffic condition data.

Images