A road risk prediction method, device, computer equipment and storage medium

Abstract

The present invention relates to the technical field of traffic safety, and discloses a road risk prediction method, device, computer equipment and storage medium, the method comprising: obtaining comprehensive vehicle trajectory data of a section to be tested, then dividing the section to be tested into a plurality of interval sections, and correspondingly obtaining vehicle trajectory data corresponding to each interval section; calculating the risk evaluation index value of each risk evaluation index in the section to be tested and each interval section according to the comprehensive vehicle trajectory data of the section to be tested and the vehicle trajectory data of each interval section; calculating the weight of each risk evaluation index according to the risk evaluation index value of each risk evaluation index in different interval sections; calculating the risk evaluation results of the section to be tested and each interval section according to the weight of each risk evaluation index and the risk evaluation index value of each risk evaluation index in the section to be tested and different interval sections. The present invention realizes the risk evaluation of traffic management in advance and improves the accuracy of road traffic safety analysis.

Description

Technical Field

[0001] The present invention relates to the field of traffic safety technology, and in particular to a road risk prediction method, device, computer equipment and storage medium. Background Art

[0002] At present, road traffic safety risk assessment mainly uses historical violations and accident data for evaluation, and the corresponding traffic hazard inspection work is mainly based on experience and manual analysis.

[0003] However, traffic violation and accident data are lagging, highly random and low-probability, and it often takes a long period to accumulate accident data that meets the requirements of statistical analysis. In addition, different scenarios have different traffic operation characteristics, thus having different accident causes and road risk hazards. Using the same evaluation indicators and weights cannot accurately reflect road traffic safety risks. Summary of the invention

[0004] In view of this, the present invention provides a road risk prediction method to solve the problem that the road risk evaluation indicators and weights currently existing in the related art are the same and cannot accurately reflect the road traffic safety risks in a short time.

[0005] In a first aspect, the present invention provides a road risk prediction method, which includes: obtaining comprehensive vehicle trajectory data of a road section to be tested; dividing the road section to be tested into multiple intervals to obtain multiple interval sections; dividing the comprehensive vehicle trajectory data into vehicle trajectory data corresponding to each interval section according to the interval sections; calculating the risk assessment index value of each risk assessment indicator in each interval section according to the vehicle trajectory data of each interval section; calculating the weight of each risk assessment indicator according to the risk assessment index value of each risk assessment indicator in different interval sections; calculating the risk assessment result of each interval section according to the weight of each risk assessment indicator and the risk assessment index value of each risk assessment indicator in different interval sections.

[0006] The road risk prediction method provided by the present invention divides the road section to be tested into multiple interval sections, and calculates each risk assessment index value based on the vehicle trajectory data in different interval sections. After calculating the risk assessment index value in each interval section, the weight of each risk assessment index is determined according to the risk assessment index value of each risk assessment index in each interval section. That is, in the embodiment of the present invention, the weight used to calculate the risk assessment result is determined based on the comprehensive vehicle trajectory data of the road section to be tested. Therefore, the risk assessment result calculated by the embodiment of the present invention can more truly reflect the actual situation of the road section to be tested, and the risk assessment result is more accurate. In addition, the accurate risk assessment results of different interval sections in the road section to be tested are calculated in the embodiment of the present invention, which helps the road monitoring party to accurately locate the high-risk points of the road section.

[0007] In an optional implementation, the comprehensive risk assessment index values ​​corresponding to the risk assessment indicators are calculated according to the comprehensive vehicle trajectory data; and the comprehensive risk assessment result of the road section to be tested is calculated according to the comprehensive risk assessment index values ​​and the weights of the risk assessment indicators.

[0008] The road risk prediction method provided by the present invention, in addition to calculating the risk assessment results of each interval section, also calculates the comprehensive risk assessment index value of the section to be tested, and then calculates the comprehensive risk assessment result of the section to be tested based on the weight of each risk assessment index and the comprehensive risk assessment index value to realize the monitoring of the overall risk of the section to be tested.

[0009] In an optional implementation, the road section to be tested is divided into a plurality of intervals, and a method of obtaining the plurality of interval sections includes: adopting an equal-length division method to divide the road section to be tested into a plurality of interval sections of equal length.

[0010] The road risk prediction method provided by the present invention adopts an equal-length division method which is suitable for a road section to be tested with a single functional unity. The division method is simple to calculate and the size of the divided section can be determined according to the required detection accuracy.

[0011] In an optional implementation, the road section to be tested is divided into multiple intervals, and the method of obtaining multiple interval sections includes: dividing the road section to be tested into multiple interval sections according to the functional attributes of different intervals in the road section to be tested.

[0012] The road risk prediction method provided by the present invention divides the interval sections according to the functional attributes of the sections in the section to be tested. The division method takes into account the differences in vehicle trajectory data of different sections in the section to be tested. When finally calculating the risk assessment results of the interval sections, the risk assessment results matching each interval section will be calculated based on the values ​​of each risk evaluation index in different interval sections, which helps to find high-risk locations in the section to be tested and prevent the occurrence of road risks in a targeted manner.

[0013] In an optional implementation, a method for calculating the weight of each risk assessment indicator based on the risk assessment index value of each risk assessment indicator in different interval sections includes: using the entropy weight method to calculate the risk assessment index value of each risk assessment indicator in different interval sections to obtain the weight of each risk assessment indicator; if the number of risk assessment indicators is greater than the number of interval sections, arranging the risk assessment indicators from large to small according to the weight, and determining the first n risk assessment indicators as the evaluation indicators used to calculate the risk assessment results, n is the number of interval sections; using the entropy weight method to calculate the risk assessment index values ​​of the evaluation indicators used to calculate the risk assessment results in different interval sections to obtain the weight of each risk assessment indicator.

[0014] The road risk prediction method provided by the present invention controls the number of risk assessment indicators, and compensates for the defect that the entropy weight method cannot reduce the dimension by the method of indicator dimensionality reduction. The indicators with high entropy values, that is, small numerical changes, low weights, and low contribution to risk assessment are excluded. Under the premise of retaining high-weight indicators and achieving evaluation accuracy, the calculation cost is reduced and the evaluation efficiency is improved.

[0015] In an optional implementation, the entropy weight method is used to calculate the risk assessment index value of each risk assessment index in different interval sections to obtain the weight of each risk assessment index, including: establishing an original matrix according to the risk assessment index value of each risk assessment index in each interval section; standardizing the original matrix to obtain the index value of each interval section; each interval section index value corresponds to a risk assessment indicator in an interval section; calculating the corresponding weight according to the index value of each interval section, and calculating each information entropy according to the weight; each information entropy corresponds to a risk assessment indicator; and obtaining the risk assessment index weight according to the information entropy calculation.

[0016] The road risk prediction method provided by the present invention calculates the risk assessment index weights corresponding to each risk assessment index through the entropy weight method. This method can be used when there is correlation and interaction between different indicators, objectively reflects the differences and importance between different indicators, and the calculation results are relatively accurate, objective and reliable.

[0017] In an optional implementation, obtaining comprehensive vehicle trajectory data of the road section to be tested includes: obtaining a video image of the road section to be tested; and extracting the comprehensive vehicle driving trajectory according to the video image.

[0018] The road risk prediction method provided by the present invention extracts the vehicle driving trajectory of the video image of the road section to be tested through multiple algorithms, ensuring that detailed and accurate data is extracted from the video image of the road section to be tested, and providing reliable vehicle trajectory data for subsequent calculation of the risk prediction result of the road section to be tested.

[0019] In a second aspect, the present invention provides a road risk prediction device, which includes: a trajectory data acquisition module, used to acquire comprehensive vehicle trajectory data of a road section to be tested; a road section division module, used to divide the road section to be tested into multiple sections to obtain multiple section sections; a trajectory data division module, used to divide the comprehensive vehicle trajectory data into vehicle trajectory data corresponding to each section section according to the section sections; an index value calculation module, used to calculate the risk assessment index value of each risk assessment index in each section section according to the vehicle trajectory data of each section section; a weight calculation module, used to calculate the weight of each risk assessment index according to the risk assessment index value of each risk assessment index in different section sections; a risk assessment module, used to calculate the risk assessment result of each section section according to the weight of each risk assessment index and the risk assessment index value of each risk assessment index in different section sections.

[0020] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the road risk prediction method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0021] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the road risk prediction method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a schematic flow chart of a road risk prediction method according to an embodiment of the present invention;

[0024] Figure 2 is a flow chart of another road risk prediction method according to an embodiment of the present invention;

[0025] Figure 3 is a flow chart of another road risk prediction method according to an embodiment of the present invention;

[0026] Figure 4 is a schematic structural diagram of a road risk prediction device according to an embodiment of the present invention;

[0027] Figure 5 Schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0029] Since the data of traffic violations and accidents are lagging, low-probability and random, it takes a long period to analyze road risks by relying on real-time data; and when choosing to make predictions based on existing traffic conditions, since the probability of risk and the factors causing risk are usually different in different traffic scenes, if the same road risk prediction index and risk prediction weight are used in different traffic scenes, the actual traffic safety risk cannot be accurately reflected. Therefore, the present invention provides a road risk prediction method, which divides the road section to be tested into multiple interval sections, calculates the weight of each risk evaluation index according to the evaluation index value of each evaluation index in different interval sections, and then calculates the risk assessment result of each interval section according to the weight of each risk assessment index and the risk assessment index value of each risk assessment index in each interval section, so that the risk assessment result conforms to the scene of each interval section.

[0030] According to an embodiment of the present invention, an embodiment of a road risk prediction method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0031] In this embodiment, a road risk prediction method is provided. Figure 1 is a flow chart of a road risk prediction method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0032] Step S101, obtaining comprehensive vehicle trajectory data of the road section to be tested.

[0033] In an optional embodiment, to evaluate the risk assessment result of the road section to be tested, it is necessary to analyze the vehicle trajectory data of the road section to be tested. Therefore, the vehicle driving trajectory of the road section to be tested is first obtained, and the comprehensive vehicle trajectory data in the vehicle driving trajectory is extracted.

[0034] In an optional embodiment, the vehicle trajectory data refers to the data information obtained by sampling the movement process of the car in the road section to be tested, including the sampling point location, sampling time, vehicle speed, etc. These sampling point data information constitute the trajectory data according to the sampling order.

[0035] Step S102, dividing the road section to be tested into a plurality of intervals to obtain a plurality of interval sections.

[0036] In an optional embodiment, the road section to be tested may include different road conditions such as straight roads, intersections, diversion areas, and confluence areas, and the risk conditions of different sections are different. Therefore, the entire road section to be tested is divided into multiple interval sections, and the risk assessment results of each interval section are subsequently calculated.

[0037] Step S103, dividing the comprehensive vehicle trajectory data into vehicle trajectory data corresponding to each section according to the section.

[0038] In an optional embodiment, after the entire road section to be tested is divided into a plurality of interval sections, it is necessary to obtain vehicle trajectory data corresponding to each interval section to calculate the risk assessment result of each interval section.

[0039] Step S104, calculating the risk assessment index value of each risk assessment index in each section according to the vehicle trajectory data of each section.

[0040] In an optional embodiment, before calculating the risk assessment results of each section, the risk assessment index content to be used is determined, and the risk assessment index value corresponding to each risk assessment index in each section is calculated based on the vehicle trajectory data of each section.

[0041] In an optional embodiment, the risk assessment index refers to an index that can be used to evaluate the risk situation of a road section. Any influencing factor that may bring risks to cars traveling on the road can be used as a risk assessment index. For example, traffic flow characteristic indicators: traffic flow, average speed, vehicle composition, number of lane changes; traffic safety alternative evaluation indicators: distance collision time, headway, etc. The risk assessment index value corresponding to the risk assessment index in each section is calculated through vehicle trajectory data. Take traffic flow as an example: traffic flow is the number of vehicles passing through the section per unit time.

[0042] Step S105, calculating the weight of each risk assessment indicator according to the risk assessment indicator value of each risk assessment indicator in different interval sections.

[0043] In an optional embodiment, after obtaining the risk assessment index value of a section, it is further necessary to determine the weight coefficient corresponding to each risk assessment index value. The risk assessment index with a larger weight has a more obvious impact on the risk assessment result.

[0044] Step S106, calculating the risk assessment result of each section according to the weight of each risk assessment indicator and the risk assessment indicator value of each risk assessment indicator in different sections.

[0045] In an optional embodiment, after obtaining the weight of each risk assessment index, a weighted method is used to calculate the final risk assessment result. For a section, the weight of each risk assessment index and the value of each risk assessment index are weighted and calculated to obtain the risk assessment result of the section.

[0046] The road risk prediction method provided in this embodiment divides the road section to be tested into multiple interval sections, and calculates each risk assessment index value based on the vehicle trajectory data in different interval sections. After calculating the risk assessment index value in each interval section, the weight of each risk assessment index is determined according to the risk assessment index value of each risk assessment index in each interval section. That is, in the embodiment of the present invention, the weight used to calculate the risk assessment result is determined based on the comprehensive vehicle trajectory data of the road section to be tested. Therefore, the risk assessment result calculated by the embodiment of the present invention can more truly reflect the actual situation of the road section to be tested, and the risk assessment result is more accurate. In addition, the accurate risk assessment results of different interval sections in the road section to be tested are calculated in the embodiment of the present invention, which helps the road monitoring party to accurately locate the high-risk points of the road section.

[0047] In some optional implementations, comprehensive risk assessment index values ​​corresponding to respective risk assessment indicators are calculated based on comprehensive vehicle trajectory data; and comprehensive risk assessment results of the road section to be tested are calculated based on the comprehensive risk assessment index values ​​and weights of the risk assessment indicators.

[0048] The road risk prediction method provided in this embodiment, in addition to calculating the risk assessment results of each interval section, also calculates the comprehensive risk assessment index value of the section to be tested, and then calculates the comprehensive risk assessment result of the section to be tested based on the weight of each risk assessment index and the comprehensive risk assessment index value to monitor the overall risk of the section to be tested.

[0049] In some optional implementations, the road section to be tested is divided into a plurality of intervals to obtain a plurality of interval sections, including: using an equal-length division method to divide the road section to be tested into a plurality of interval sections of equal length.

[0050] Specifically, the road section to be tested is divided into multiple interval sections, and the method includes equal length division, that is, the road section to be tested is divided into n equal length interval sections, and the length of each interval section is the same. This allocation method is used for straight roads, curves with the same curvature, steep slopes with the same longitudinal gradient, and other road sections with relatively uniform road lines or functions. The specific length of each interval section is determined according to the road section to be tested and the detection accuracy.

[0051] The road risk prediction method provided in this embodiment adopts an equal-length division method that is suitable for a road section to be tested with a single functional integration. This division method is simple to calculate, and the size of the divided section can also be determined according to the required detection accuracy.

[0052] In some optional implementations, the road section to be tested is divided into multiple intervals to obtain multiple interval sections, including: dividing the road section to be tested into multiple interval sections according to the functional attributes of different intervals in the road section to be tested.

[0053] Specifically, the road section to be tested is divided into multiple interval sections, and the key node division method is used, that is, according to the functional attributes of the road section in the road section to be tested, the road section to be tested is divided into n interval sections with different functions. This allocation method is used in intersections, merging areas, diverging areas and other sections with complex linear structures and clear functional segments. The functional attributes of each interval section are determined by the highway itself.

[0054] The road risk prediction method provided in this embodiment divides the interval sections according to the functional attributes of the sections in the section to be tested. The division method takes into account the differences in vehicle trajectory data in different sections in the section to be tested. When finally calculating the risk assessment results of the interval sections, the risk assessment results that match each interval section will be calculated based on the values ​​of each risk evaluation indicator in different interval sections. This helps to find high-risk locations in the section to be tested and prevent road risks in a targeted manner.

[0055] In this embodiment, a road risk prediction method is provided. Figure 2 is a flow chart of a road risk prediction method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0056] Step S201, obtain comprehensive vehicle trajectory data of the road section to be tested. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0057] Step S202: Divide the road section to be tested into multiple sections to obtain multiple sections. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0058] Step S203: divide the comprehensive vehicle trajectory data into vehicle trajectory data corresponding to each section according to the section. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0059] Step S204: Calculate the risk assessment index value of each risk assessment index in each section according to the vehicle trajectory data of each section. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0060] Step S205, calculating the weight of each risk assessment indicator according to the risk assessment indicator value of each risk assessment indicator in different interval sections.

[0061] Specifically, the above step S205 includes:

[0062] Step S2051, using the entropy weight method to calculate the risk assessment index value of each risk assessment index in different interval sections to obtain the weight of each risk assessment index.

[0063] In an optional embodiment, after obtaining the risk assessment index values ​​in different intervals of road sections, considering that different risk assessment indicators have different influences on road risk results, it is necessary to recalculate the risk assessment index weights corresponding to different risk assessment indicators. Here, the entropy weight method is adopted to substitute the risk assessment index values ​​to obtain the weights of each risk assessment indicator.

[0064] Step S2052, determining whether the number of risk assessment indicators is greater than the number of interval sections.

[0065] In an optional embodiment, since the risk assessment indicators are manually selected, the number of risk assessment indicators may be greater than the number of interval sections, or may be less than or equal to the number of interval sections; and the number of risk assessment indicators does not need to be too many, especially the risk assessment indicators with small weights will basically not affect the final road risk assessment results. Therefore, the number of interval sections is chosen as a measure to determine whether the number of risk assessment indicators is greater than the number of interval sections.

[0066] Step S2052, arrange the risk assessment indicators from large to small according to the weight, determine the first n risk assessment indicators as the assessment indicators used to calculate the risk assessment result, n is the number of interval sections; return to execute step S2051.

[0067] In an optional embodiment, when the number of risk assessment indicators is greater than the number of interval road sections, the evaluation indicator weights are sorted from large to small, and the evaluation indicators whose ranking is greater than the number of interval road sections are discarded. Only the first n risk assessment indicators are retained as the evaluation indicators ultimately used to calculate the risk assessment results. After removing some risk assessment indicators, the remaining risk assessment indicator weights need to be recalculated.

[0068] Step S206, according to the weight of each risk assessment indicator and the risk assessment indicator value of each risk assessment indicator in different sections, the risk assessment result of each section is calculated. Figure 1 Step S106 of the illustrated embodiment will not be described in detail here.

[0069] This embodiment proposes an improved entropy weight method based on discarding high entropy values. The information entropy of indicators with large weights is small, which means that the discrete degree of the indicator between the interval sections is large and the state is relatively unstable. Therefore, it has a greater impact on the safety of the road, which conforms to objective laws. Abandoning evaluation indicators with small weights can reduce the calculation cost and improve the evaluation efficiency while retaining high-weight indicators and achieving evaluation accuracy. In addition, for the retained large-weight evaluation indicators, it means that the indicator has a large contribution to the risk assessment of the scene section. The impact of the dynamic characteristics of traffic flow on the indicator can be used to infer the road risk type and the cause of the accident. For example, if the vehicle composition indicator has a large weight, it means that the larger the proportion of large vehicles, the less safe it is. The road section has a road risk with a high proportion of trucks, and targeted truck restriction control can be carried out.

[0070] The road risk prediction method provided in this embodiment controls the number of risk assessment indicators. The indicator dimensionality reduction method in this embodiment makes up for the defect that the entropy weight method cannot reduce the dimensionality. The indicators with high entropy values, that is, small numerical changes, low weights, and low contribution to risk assessment are excluded. Under the premise of retaining high-weight indicators and achieving evaluation accuracy, the calculation cost is reduced and the evaluation efficiency is improved.

[0071] In this embodiment, a road risk prediction method is provided. Figure 3 is a flow chart of a road risk prediction method according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0072] Step S301, obtain comprehensive vehicle trajectory data of the road section to be tested. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0073] Step S302: Divide the road section to be tested into multiple sections to obtain multiple sections. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0074] Step S303: divide the comprehensive vehicle trajectory data into vehicle trajectory data corresponding to each section according to the section. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0075] Step S304: Calculate the risk assessment index value of each risk assessment index in each section according to the vehicle trajectory data of each section. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0076] Step S305, calculating the weight of each risk assessment indicator according to the risk assessment indicator value of each risk assessment indicator in different interval sections.

[0077] Specifically, step S305 includes:

[0078] Step S3051, using the entropy weight method to calculate the risk assessment index value of each risk assessment index in different interval sections to obtain the weight of each risk assessment index.

[0079] Specifically, the above step S3051 includes:

[0080] Step S30511, establishing an original matrix according to the risk assessment index values ​​of each risk assessment index in each section.

[0081] In an optional embodiment, the road risk evaluation index value set of the interval section is used as the original matrix X ij , this original matrix looks like this:

[0082]

[0083] Among them, X represents the risk assessment index value, i represents the i-th section, and j represents the j-th risk assessment index.

[0084] Step S30512, normalize the original matrix to obtain the index value of each section.

[0085] In an optional embodiment, each interval road section index value corresponds to a risk assessment index in an interval road section.

[0086] In an optional embodiment, the original matrix is ​​normalized to obtain the interval section index value Y ij In each section, a risk assessment index value corresponds to an interval section index value. In addition, the interval section index value can only be a positive index or a negative index. The calculation method of the interval section index value is:

[0087] For positive indicators, that is, evaluation indicators with higher values ​​and higher risks, the calculation formula is:

[0088]

[0089] For negative indicators, that is, evaluation indicators with smaller values ​​and higher risks, the calculation formula is:

[0090]

[0091] Step S30513, calculating the corresponding weight according to the interval road section index value, and calculating each information entropy according to the weight.

[0092] In an optional embodiment, each information entropy corresponds to a risk assessment indicator.

[0093] In an optional embodiment, the above-mentioned interval road section index value is used to calculate its corresponding proportion, and the calculation formula is:

[0094]

[0095] P ij That is, it represents the weight corresponding to the jth risk assessment index of the ith section. Then, the weight corresponding to the same risk assessment index in all sections is used to calculate the information entropy of the risk assessment index, and the calculation formula is:

[0096]

[0097] Step S30514, obtaining the risk assessment index weight based on information entropy calculation.

[0098] In an optional embodiment, the weight W of the jth evaluation index is calculated j The formula is as follows:

[0099]

[0100] In an optional embodiment, as mentioned above, the weight W j It is not necessarily the final risk assessment index weight, and it is necessary to perform an index dimension reduction. If the number of evaluation indicators is less than or equal to the number of divided interval sections, then the current W j is the risk assessment index weight; if the number of assessment indicators is greater than the number of divided sections, the assessment index weight W j Sort from large to small, discard the evaluation index whose ranking is greater than the number of interval sections, and repeat the process from step S30511 to step S30514, recalculate the weight, and determine it as the final risk evaluation index weight W' j .

[0101] Step S3052: determine whether the number of risk assessment indicators is greater than the number of interval sections. Figure 2 Step S2052 of the illustrated embodiment will not be described in detail here.

[0102] Step S3053: Arrange the risk assessment indicators from large to small according to the weight, determine the first n risk assessment indicators as the assessment indicators for calculating the risk assessment result, n is the number of interval sections, and return to execute step S3051. For details, please refer to Figure 2 Step S2053 of the illustrated embodiment will not be described in detail here.

[0103] Step S306: Calculate the risk assessment result of each section according to the weight of each risk assessment indicator and the risk assessment indicator value of each risk assessment indicator in different sections. Figure 1 Step S106 of the illustrated embodiment will not be described in detail here.

[0104] In an optional embodiment, the method of finally calculating the risk assessment results of each section and the comprehensive risk assessment results of the section to be tested is as follows:

[0105] (1) Using the normalized interval section index value Y ij Obtain the comprehensive road risk score S for each section i , that is, the risk assessment results of each section of the road to be tested are obtained, and the calculation formula is:

[0106]

[0107] Among them, S i (i∈(1,n]) is the comprehensive risk evaluation value of the i-th section, which is used to analyze the risk points and types of accidents that are prone to occur on the section. The larger the value, the higher the risk.

[0108] (2) Using the normalized comprehensive risk assessment index value to obtain the comprehensive risk assessment result of the road section to be tested, and obtain the risk value of the road section to be tested. The steps are as follows: ① Using the normalization method in step S30512 to calculate the comprehensive risk assessment index value Y of the road section to be tested allj ; ② Calculate the comprehensive risk assessment results of the road section to be tested, and the calculation is publicized as follows:

[0109] S all =w′ j Y all j (8)

[0110] Similarly, the comprehensive risk assessment result is expressed as the higher the comprehensive score, the greater the risk.

[0111] The road risk prediction method provided in this embodiment calculates the risk assessment index weights corresponding to each risk assessment index through the entropy weight method. This method can be used when there is correlation and interaction between different indicators, objectively reflects the differences and importance between different indicators, and the calculation results are relatively accurate, objective and reliable.

[0112] In some optional implementations, obtaining comprehensive vehicle trajectory data of the road section to be tested includes: obtaining a video image of the road section to be tested; and extracting the comprehensive vehicle driving trajectory according to the video image.

[0113] Specifically, in this embodiment, after obtaining the video images with vehicle driving tracks on both sides of the road, the lane line recognition algorithm based on the U-Net3+ network is selected to identify the lane lines, and then the vehicle is identified based on the Mask-RCNN vehicle target recognition algorithm. After the vehicle is identified, the vehicle tracking algorithm based on SORT is used to track the vehicle, and finally the vehicle driving track data is extracted.

[0114] In an optional embodiment, this embodiment can also extract the vehicle-mounted electronic information of the vehicle by connecting to the vehicle. The vehicle-mounted electronic information includes but is not limited to the vehicle's sensor device data or the navigation track data recorded by the vehicle navigation device, and the collision event data recorded by the vehicle event data recorder system (EDR). In addition, the navigation track data can be obtained from the relevant vehicle enterprise platform, navigation platform, etc., and the vehicle event data record is extracted from the vehicle EDR. However, after selecting to extract the vehicle-mounted electronic information of the vehicle, it is also necessary to align and integrate the data. For example, for navigation trajectory data, the first frame image recorded by the vehicle-mounted electronic data after the vehicle appears in the video is used as the reference, and the longitude and latitude coordinates of the image video and the vehicle-mounted electronic data are aligned. According to the positioning (longitude, latitude) and actual vehicle speed in the electronic vehicle information, the vehicle coordinates (x, y), vehicle longitudinal speed, and vehicle type in the driving trajectory data are corrected. The interpolation method can be used to fill in the values ​​at the time when there is no record; for collision event data extracted from EDR, the starting point of the collision event (i.e., time zero) is used as the reference, based on the characteristics of the accident process in the video image, time, and the speed change trend of the two, the video image and the time zero recorded by the EDR are aligned, and the vehicle longitudinal speed, longitudinal acceleration, lateral acceleration, yaw angular velocity, and vehicle type in the driving trajectory data are corrected. The interpolation method can be used to fill in the values ​​at the time when there is no record. Since the process of extracting vehicle trajectory data through the vehicle-mounted electronic information of the car involves more algorithms, this embodiment does not use this method as a necessary option for extracting vehicle trajectory data.

[0115] The road risk prediction method provided in this embodiment uses multiple algorithms to extract vehicle driving trajectories in the video image of the road section to be tested, ensuring that detailed and accurate data is extracted from the video image of the road section to be tested, and providing reliable vehicle trajectory data for subsequent calculation of risk prediction results for the road section to be tested.

[0116] In this embodiment, a road risk prediction device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0117] This embodiment provides a road risk prediction device, such as Figure 4 As shown, including:

[0118] The trajectory data acquisition module 1 is used to acquire the comprehensive vehicle trajectory data of the road section to be tested.

[0119] The road section division module 2 is used to divide the road section to be tested into multiple intervals to obtain multiple interval sections.

[0120] The trajectory data division module 3 is used to divide the comprehensive vehicle trajectory data into vehicle trajectory data corresponding to each section according to the section.

[0121] The index value calculation module 4 is used to calculate the risk evaluation index value of each risk evaluation index in each section according to the vehicle trajectory data of each section.

[0122] The weight calculation module 5 is used to calculate the weight of each risk assessment indicator according to the risk assessment indicator value of each risk assessment indicator in different interval sections.

[0123] The risk assessment module 6 is used to calculate the risk assessment result of each section according to the weight of each risk assessment indicator and the risk assessment indicator value of each risk assessment indicator in different sections.

[0124] In some optional implementations, the weight calculation module 5 includes:

[0125] The weight calculation unit is used to calculate the risk assessment index value of each risk assessment index in different interval sections by using the entropy weight method to obtain the weight of each risk assessment index.

[0126] The quantity judgment unit is used to judge the quantity of risk assessment indicators and the quantity of interval sections.

[0127] The indicator screening unit is used to arrange the risk assessment indicators from large to small according to the weight, and determine the first n risk assessment indicators as the evaluation indicators used to calculate the risk assessment results, where n is the number of interval sections.

[0128] In some optional implementations, the weight calculation unit includes:

[0129] The original matrix establishing subunit is used to establish the original matrix according to the risk evaluation index value of each risk evaluation index in each section.

[0130] The index value calculation unit for each interval road section is used to standardize the original matrix to obtain the index value of each interval road section; each interval road section index value corresponds to a risk assessment index in an interval road section.

[0131] The information entropy calculation subunit is used to calculate the corresponding weight according to the index value of each interval road section, and calculate each information entropy according to the weight; each information entropy corresponds to a risk evaluation index.

[0132] The weight calculation subunit is used to obtain the risk assessment index weight based on information entropy calculation.

[0133] The road risk prediction device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0134] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0135] The embodiment of the present invention also provides a computer device having the above Figure 4 The road risk prediction device shown.

[0136] See also Figure 5 , Figure 5 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 5 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 10 is taken as an example.

[0137] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0138] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0139] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the use of a computer device based on the presentation of a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0140] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0141] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0142] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0143] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A road risk prediction method, characterized in that: include: Obtain comprehensive vehicle trajectory data of the road section to be tested; Dividing the road section to be tested into a plurality of intervals to obtain a plurality of interval sections, including: dividing the road section to be tested into interval sections with different functions according to the functional attributes of the road section in the road section to be tested; According to the interval road section, the comprehensive vehicle trajectory data is divided into vehicle trajectory data corresponding to each interval road section; According to the vehicle trajectory data of each section, the risk assessment index value of each risk assessment index in each section is calculated respectively; the risk assessment index refers to an index that can be used to evaluate the risk situation of the section, including: traffic flow characteristic index and traffic safety alternative assessment index; According to the risk assessment index value of each risk assessment index in different interval sections, the weight of each risk assessment index is calculated, including: using the entropy weight method to calculate the risk assessment index value of each risk assessment index in different interval sections to obtain the weight of each risk assessment index, including: establishing an original matrix according to the risk assessment index value of each risk assessment index in each interval section; standardizing the original matrix to obtain the index value of each interval section; each interval section index value corresponds to a risk assessment index in an interval section; calculating the corresponding proportion according to the index value of each interval section, and calculating each information entropy according to the proportion; each information entropy corresponds to a risk assessment index; obtaining the risk assessment index weight according to the information entropy calculation; if the number of the risk assessment indicators is greater than the number of the interval sections, arranging the risk assessment indicators from large to small according to the weight, and determining the first n risk assessment indicators as the evaluation indicators used to calculate the risk assessment results, n is the number of interval sections; using the entropy weight method to calculate the risk assessment index values ​​of the evaluation indicators used to calculate the risk assessment results in different interval sections to obtain the weight of each risk assessment index; According to the weight of each risk assessment indicator and the risk assessment indicator value of each risk assessment indicator in different interval sections, the risk assessment result of each interval section is calculated.

2. The method according to claim 1, characterized in that include: Calculate the comprehensive risk assessment index value corresponding to each risk assessment index according to the comprehensive vehicle trajectory data; The comprehensive risk assessment result of the road section to be tested is calculated according to the comprehensive risk assessment index value and the weight of each risk assessment index.

3. The method according to claim 1, characterized in that The step of dividing the road section to be tested into a plurality of intervals to obtain a plurality of interval sections includes: The road section to be tested is divided into a plurality of interval sections of equal length by adopting an equal-length division method.

4. The method according to claim 1, characterized in that: The step of dividing the road section to be tested into a plurality of intervals to obtain a plurality of interval sections includes: The road section to be tested is divided into multiple interval sections according to the functional attributes of different sections in the road section to be tested.

5. The method according to claim 1, characterized in that The step of obtaining comprehensive vehicle trajectory data of the road section to be tested includes: Acquire a video image of the road section to be tested; The comprehensive vehicle driving trajectory is extracted according to the video image.

6. A road risk prediction device, characterized in that: The method for predicting road risks according to any one of claims 1 to 5 comprises: A trajectory data acquisition module is used to obtain comprehensive vehicle trajectory data of the road section to be tested; A road section division module, used for dividing the road section to be tested into a plurality of intervals to obtain a plurality of interval sections, including: dividing the road section to be tested into interval sections with different functions according to the functional attributes of the road section in the road section to be tested; A trajectory data division module, used for dividing the comprehensive vehicle trajectory data into vehicle trajectory data corresponding to each section according to the section; An index value calculation module is used to calculate the risk evaluation index value of each risk evaluation index in each section according to the vehicle trajectory data of each section; A weight calculation module is used to calculate the weight of each risk assessment indicator according to the risk assessment indicator value of each risk assessment indicator in different sections; The weight calculation module includes: a weight calculation unit, which is used to calculate the risk evaluation index value of each risk evaluation index in different interval sections using an entropy weight method to obtain the weight of each risk evaluation index; A quantity determination unit, used to determine the quantity of the risk assessment indicators and the quantity of the interval sections; An indicator screening unit, used to arrange the risk assessment indicators from large to small according to the weight, and determine the first n risk assessment indicators as the assessment indicators for calculating the risk assessment result, where n is the number of interval sections; The weight calculation unit includes: an original matrix establishment subunit, which is used to establish an original matrix according to the risk evaluation index value of each risk evaluation index in each section; Each section index value calculation unit is used to perform standardization processing on the original matrix to obtain each section index value; each section index value corresponds to a risk assessment index in one section; An information entropy calculation subunit is used to calculate the corresponding weight of each section index value according to the index value, and calculate each information entropy according to the weight; each information entropy corresponds to a risk assessment index; A weight calculation subunit, used to calculate the risk assessment index weight according to the information entropy; The risk assessment module is used to calculate the risk assessment results of each section according to the weight of each risk assessment indicator and the risk assessment indicator value of each risk assessment indicator in different section sections.

7. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the road risk prediction method according to any one of claims 1 to 5 by executing the computer instructions.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the road risk prediction method according to any one of claims 1 to 5.

Images