CN118097968A - Road traffic safety assessment method - Google Patents
Road traffic safety assessment method Download PDFInfo
- Publication number
- CN118097968A CN118097968A CN202410481555.6A CN202410481555A CN118097968A CN 118097968 A CN118097968 A CN 118097968A CN 202410481555 A CN202410481555 A CN 202410481555A CN 118097968 A CN118097968 A CN 118097968A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- safety
- value
- sub
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 151
- 230000002159 abnormal effect Effects 0.000 claims abstract description 59
- 231100000279 safety data Toxicity 0.000 claims abstract description 53
- 238000010801 machine learning Methods 0.000 claims abstract description 39
- 230000001133 acceleration Effects 0.000 claims description 71
- 230000008569 process Effects 0.000 claims description 49
- 238000012549 training Methods 0.000 claims description 42
- 239000000203 mixture Substances 0.000 claims description 31
- 206010039203 Road traffic accident Diseases 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 14
- 239000013598 vector Substances 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 8
- 238000005457 optimization Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of traffic safety, and discloses a road traffic safety assessment method; the method comprises the steps of collecting real-time comprehensive safety data, predicting a real-time safety state value based on a machine learning model, judging whether to enter a road early warning mode, comparing the comprehensive safety data with a corresponding safety value, marking abnormal data, generating an excess ratio, and marking safety intervention data; compared with the prior art, the method and the system can accurately predict and evaluate the safety state of the road traffic, timely and accurately mark the abnormal data which has negative influence on the road traffic when the road traffic is in a high-risk state, accurately mark the data which needs to be subjected to safety intervention according to the specific numerical value of the abnormal data, provide the data for a traffic management department as a basis for the follow-up road traffic intervention, ensure that the road traffic makes a targeted intervention measure at the first time of the high-risk state, and further improve the safety state of the road traffic.
Description
Technical Field
The invention relates to the technical field of traffic safety, in particular to a road traffic safety assessment method.
Background
In an urban road traffic network, road traffic is an important component of an urban road network system, different types of traffic participants participate in urban operation through the road traffic, the safety of the road traffic directly affects the personal and property safety of all the traffic participants, and in order to reduce the probability of traffic accidents on the road traffic, a traffic management department needs to accurately evaluate the safety state of the road traffic, so that subsequent optimization processing is performed according to the actual safety state of the road traffic.
The patent application with the reference of publication number CN117809458A discloses a real-time assessment method and a real-time assessment system for traffic accident risks, which are based on high-precision vehicle track data, calculate road section traffic conflict conditions by using an expanded ranging collision algorithm, and calibrate the risks of road sections based on the number of traffic conflicts, the severity of the traffic conflicts and the number of conflicting vehicles in unit time of the road sections; considering the running state of the traffic flow of the road section, selecting indexes from three aspects of speed characteristics, lane changing behaviors and traffic flow to construct an interweaving region risk evaluation index set, finally establishing a road section risk identification model based on a binary logic model, and completing model parameter estimation by using a maximum likelihood method to realize real-time and accurate evaluation of the road section traffic accident risk probability;
The prior art has the following defects:
The existing road traffic safety assessment method collects real-time road condition information on road traffic through a camera monitoring device, assesses the safety state of the road traffic after manual analysis and automatic calculation, and is not beneficial to the safety construction of the road traffic because the type of the road condition information is not comprehensive enough, so that the assessment result of the road traffic safety state is limited, the accuracy of the assessment result is not high, and abnormal data cannot be marked quickly and accurately when the road traffic safety state is abnormal.
In view of the above, the present invention proposes a road traffic safety assessment method to solve the above-mentioned problems.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides the following technical scheme for achieving the purposes: a road traffic safety assessment method is applied to a traffic control center, and comprises the following steps:
S1: collecting historical training data of road traffic, wherein the historical training data comprises comprehensive safety data and safety state values;
S2: training a machine learning model for predicting a safety state value based on historical training data;
S3: collecting real-time comprehensive safety data, inputting the real-time comprehensive safety data into a machine learning model after training to predict a real-time safety state value, and judging whether to enter a road early warning mode; if the road early warning mode is entered, executing S4-S5; if the road early warning mode is not entered, repeating the step S3;
S4: comparing the comprehensive safety data with the corresponding safety value, and marking abnormal data based on the comparison result;
s5: and generating excess duty ratio values based on the abnormal data, and arranging the excess duty ratio values in a descending order to mark safety intervention data.
Further, the comprehensive safety data comprise road weather, vehicle acceleration values, vehicle density values, vehicle accident triggering rates and vehicle mixing change degrees;
The method for acquiring the vehicle acceleration value comprises the following steps:
Shooting a vehicle driving video on a road through a monitoring camera, and counting the number of motor vehicles in the vehicle driving video;
when the number of the first-appearing motor vehicles is larger than a preset number threshold value, marking all the motor vehicles in the vehicle driving video as target vehicles to obtain Target vehicles;
by measuring the speed by means of a speed measuring radar Obtaining/>, the speed of each target vehicle when entering the vehicle driving videoIndividual entry speeds and record/>The moment of the entry speed is obtained/>Starting time;
by measuring the speed by means of a speed measuring radar The speed of each target vehicle when the target vehicle exits the vehicle driving video is obtained/>Individual drive-out speeds, and record/>The moment of the individual drive-out speed is obtained/>A termination time;
Will be The respective driving-out speeds are respectively equal to/>After difference comparison of the respective entering speeds, the/>A difference in speed;
the expression of the speed difference is:
；
In the method, in the process of the invention, For/>Velocity difference,/>For/>Individual travel-out speed,/>For/>A plurality of entry speeds;
Will be The termination time is respectively equal to/>After difference comparison of the initial moments, the/>The driving time;
The expression of the travel time is:
；
In the method, in the process of the invention, For/>Travel time,/>For/>Time of termination,/>For/>Starting time;
Will be The difference of the speeds is respectively equal to/>After comparing the running time, obtain/>Sub-accelerations;
The expression of the sub acceleration is:
；
In the method, in the process of the invention, For/>Sub-accelerations;
removing sub-accelerations less than a predetermined acceleration threshold, leaving The sub acceleration is accumulated and then averaged to obtain a vehicle acceleration value;
the expression of the vehicle acceleration value is:
；
In the method, in the process of the invention, For vehicle acceleration value,/>For/>Sub-accelerations.
Further, the method for acquiring the vehicle density value comprises the following steps:
taking the time length corresponding to a preset density period as a standard, and intercepting the time length from the vehicle driving video The same-sized and rectangular images are obtained/>A sub-image;
Measuring the length of the sub-image and the width of the sub-image through a scale, and calculating the area of the sub-image through a rectangular area formula;
Sequentially counting The number of vehicles in the motor vehicle lane in the individual sub-images is obtained/>A respective vehicle value;
Will be The individual vehicle values are compared with the areas of the sub-images respectively to obtain/>A sub-density value;
The expression of the subdensity value is:
；
In the method, in the process of the invention, For/>Sub-density value,/>For/>Individual vehicle value,/>Is the area of the sub-image;
Removing the maximum and minimum values of the sub-density values, leaving The sub-density values are accumulated and averaged to obtain a vehicle density value;
The expression of the vehicle density value is:
；
In the method, in the process of the invention, Is the vehicle density value,/>For/>Sub-density values.
Further, the method for acquiring the vehicle accident triggering rate comprises the following steps:
A1: identification by monitoring camera License plates of the target vehicles and screening out and/>, from a traffic management databaseTraffic accidents corresponding to license plates of the target vehicles;
A2: recording the occurrence time of traffic accidents, marking the traffic accidents with the occurrence time within a preset accident counting period as effective accidents, and counting the number of the effective accidents;
A3: comparing the number of effective accidents with the number of target vehicles to obtain a sub-trigger rate;
a4: repeating execution Steps of sub A1-A3, obtaining/>A sub-trigger rate;
The expression of the sub-trigger rate is:
；
In the method, in the process of the invention, For/>Individual sub-trigger rate,/>For/>The number of valid incidents;
a5: marking the effective triggering rate of the sub-triggering rate larger than a preset sub-triggering threshold value to obtain Effective trigger rate, and will/>The effective triggering rates are accumulated and averaged to obtain the triggering rate of the vehicle accident;
The expression of the vehicle accident triggering rate is:
；
In the method, in the process of the invention, For the accident triggering rate of vehicles,/>For/>An effective trigger rate.
Further, the method for acquiring the vehicle mixing variation degree comprises the following steps:
b1: identifying a motor vehicle lane and a non-motor vehicle lane in the sub-image through a computer vision technology, and marking the motor vehicle lane adjacent to the non-motor vehicle lane as a target lane;
B2: at the time T1, counting the number of motor vehicles and the number of non-motor vehicles in a target lane, and marking the number as the initial number of motor vehicles and the initial number of non-motor vehicles respectively;
b3: comparing the initial quantity of the non-motor vehicles with the initial quantity of the motor vehicles to obtain initial mixing degree;
The expression of the initial mix is:
；
In the method, in the process of the invention, For initial mix,/>For the initial quantity of non-motor vehicles,/>Is the initial number of motor vehicles;
B4: at the time T2, counting the number of motor vehicles and the number of non-motor vehicles in a target lane, and marking the number as the end number of motor vehicles and the end number of non-motor vehicles respectively;
B5: comparing the end number of the non-motor vehicles with the end number of the motor vehicles to obtain an end mixing degree;
The expression for ending the degree of mixing is:
；
In the method, in the process of the invention, To end the mixing degree,/>For the end number of non-motor vehicles,/>For the end number of motor vehicles;
B6: comparing the ending mixture with the initial mixture to obtain a sub-mixture;
The expression of the sub-mix is:
；
In the method, in the process of the invention, Is the sub-mix degree;
B7: repeating execution Secondary B1-B6 procedure, obtaining/>Individual sub-mix, will/>The individual sub-mixing degrees are accumulated and averaged to obtain the vehicle mixing variation degree;
The expression of the vehicle mixture variation is:
；
In the method, in the process of the invention, For the degree of change of the vehicle mixture,/>For/>Degree of sub-mixing.
Further, the training method of the machine learning model for predicting the safety state value comprises the following steps:
Sequentially assigning road weather, assigning sunny day as AA, cloudy day as BB, rainy day as CC, ice and snow day as DD and strong wind day as EE;
the comprehensive safety data are converted into a corresponding group of characteristic vectors, the characteristic vectors are used as input of a machine learning model, the safety state value corresponding to each group of comprehensive safety data is used as output of the machine learning model, the safety state value is used as a prediction target, the sum of prediction errors of all training data is minimized to be used as a training target, and the machine learning model is trained until the sum of the prediction errors reaches convergence, and training is stopped.
Further, the security state values include a high risk state and a low risk state;
when the output of the machine learning model is 1, the real-time safety state value is in a high risk state;
When the output of the machine learning model is 0, the real-time safety state value is in a low risk state;
the judging method for entering the road early warning mode comprises the following steps:
when the real-time safety state value is in a high risk state, judging to enter a road early warning mode;
And when the real-time safety state value is in a low risk state, judging that the road early warning mode is not entered.
Further, the method for marking the abnormal data comprises the following steps:
Comparing the vehicle acceleration value with a preset acceleration safety value, and marking the vehicle acceleration value as abnormal data when the vehicle acceleration value is larger than the preset acceleration safety value;
comparing the vehicle density value with a preset density safety value, and marking the vehicle density value as abnormal data when the vehicle density value is larger than the preset density safety value;
Comparing the vehicle accident triggering rate with a preset triggering safety value, and marking the vehicle accident triggering rate as abnormal data when the vehicle accident triggering rate is larger than the preset triggering safety value;
and comparing the vehicle mixing variation with a preset mixing safety value, and marking the vehicle mixing variation as abnormal data when the vehicle mixing variation is larger than the preset mixing safety value.
Further, the excess duty ratio comprises a first duty ratio, a second duty ratio, a third duty ratio and a fourth duty ratio;
the method for generating the first duty ratio, the second duty ratio, the third duty ratio and the fourth duty ratio comprises the following steps:
When the abnormal data is a vehicle acceleration value, comparing the vehicle acceleration value with a preset acceleration safety value, and then comparing the vehicle acceleration value with the vehicle acceleration value to obtain a first duty ratio;
The expression of the first occupancy value is:
；
In the method, in the process of the invention, For the first ratio,/>The acceleration safety value is preset;
When the abnormal data is a vehicle density value, comparing the vehicle density value with a preset density safety value, and then comparing the vehicle density value with the vehicle density value to obtain a second duty ratio;
The expression of the second ratio is:
；
In the method, in the process of the invention, For the second ratio,/>Is a preset density safety value;
When the abnormal data is the vehicle accident triggering rate, comparing the vehicle accident triggering rate with a preset triggering safety value, and then comparing the vehicle accident triggering rate with the vehicle accident triggering rate to obtain a third occupation ratio;
the expression of the third ratio is:
；
In the method, in the process of the invention, For a third ratio,/>The triggering safety value is a preset triggering safety value;
when the abnormal data is the vehicle mixing variation, comparing the vehicle mixing variation with a preset mixing safety value, and then comparing the vehicle mixing variation with the vehicle mixing variation to obtain a fourth duty ratio;
The fourth duty cycle is expressed as:
；
In the method, in the process of the invention, For a fourth duty cycle,/>Is a preset trigger safety value.
Further, the marking method of the safety intervention data comprises the following steps:
when the quantity of the excess ratio is unique, marking the comprehensive safety data corresponding to the excess ratio as safety intervention data;
when the number of the excess ratio is not the same, arranging the excess ratio in descending order from large to small;
When the maximum value of the excess ratio is the same, marking the comprehensive safety data corresponding to the excess ratio of the first ranking as safety intervention data;
when the maximum value of the excess ratio is not the same, the first bit is ranked in parallel Corresponding to the ratio of excessThe individual integrated security data are marked as security intervention data.
The road traffic safety assessment method has the technical effects and advantages that:
the method comprises the steps of collecting historical training data of road traffic, wherein the historical training data comprises comprehensive safety data and safety state values, training a machine learning model for predicting the safety state values based on the historical training data, collecting real-time comprehensive safety data, inputting the real-time comprehensive safety data into the trained machine learning model to predict the real-time safety state values, judging whether to enter a road early warning mode according to the real-time safety state values, comparing the comprehensive safety data with the corresponding safety values, marking abnormal data based on a comparison result, generating an excess ratio based on the abnormal data, and arranging the excess ratio in a descending order to mark safety intervention data; compared with the prior art, the method has the advantages that the comprehensive safety data influencing the safety state of the road traffic is collected, the safety state of the road traffic can be accurately predicted and estimated based on the machine learning model, when the high risk state of the road traffic occurs, abnormal data which negatively influences the road traffic are marked timely and accurately, the data needing safety intervention are marked accurately according to the specific numerical value of the abnormal data, and the first time of the high risk state of the road traffic is ensured to be provided with targeted intervention measures, so that the safety state of the road traffic is improved.
Drawings
Fig. 1 is a schematic flow chart of a road traffic safety assessment method provided in embodiment 1 of the present invention;
fig. 2 is a schematic diagram of a road traffic safety assessment system according to embodiment 2 of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1: referring to fig. 1, the method for evaluating road traffic safety according to the present embodiment is applied to a traffic control center, and includes:
S1: collecting historical training data of road traffic, wherein the historical training data comprises comprehensive safety data and safety state values;
The comprehensive safety data is diversified data which can cause positive and negative effects on road traffic safety, and the road traffic safety state of the road traffic in a specific time period can be evaluated relatively accurately by acquiring the comprehensive safety data, so that accurate and timely data support is provided for traffic management departments;
the comprehensive safety data comprise road weather, vehicle acceleration values, vehicle density values, vehicle accident triggering rates and vehicle mixing change degrees;
Road gas refers to environmental meteorological data of the position of road traffic to be evaluated, and road meteorological data comprise sunny days, overcast days, rainy days, ice and snow days, strong wind days and the like; different road weather can cause different influences on road traffic participants, and by way of example, the line of sight of the road traffic participants is good in sunny days, accidents are not easy to happen, and the phenomenon of wheel slip easily occurs on the road in ice and snow days, and accidents are easy to happen; the worse the road weather is, the worse the safety state of the road traffic is;
The vehicle acceleration value refers to the increasing range of the vehicle speed of the vehicle on the road in a specific time period, when the vehicle acceleration value is larger, the vehicle speed on the road is larger, and when the vehicle encounters an emergency, the time required for a driver to tread a brake pedal until the vehicle is completely stationary is also larger, so that the safety state of road traffic is worse;
The method for acquiring the vehicle acceleration value comprises the following steps:
Shooting a vehicle driving video on a road through a monitoring camera, and counting the number of motor vehicles in the vehicle driving video;
when the number of the first-appearing motor vehicles is larger than a preset number threshold value, marking all the motor vehicles in the vehicle driving video as target vehicles to obtain Target vehicles; the preset number threshold value refers to the lowest value of the number of the motor vehicles which can be marked as the target vehicle, so that the motor vehicles with enough number in the vehicle driving video can meet the subsequent calculation requirement; the preset quantity threshold value is obtained by collecting a large number of historical motor vehicles marked as the target vehicles and averaging the number;
by measuring the speed by means of a speed measuring radar Obtaining/>, the speed of each target vehicle when entering the vehicle driving videoIndividual entry speeds and record/>The moment of the entry speed is obtained/>Starting time;
by measuring the speed by means of a speed measuring radar The speed of each target vehicle when the target vehicle exits the vehicle driving video is obtained/>Individual drive-out speeds, and record/>The moment of the individual drive-out speed is obtained/>A termination time;
Will be The respective driving-out speeds are respectively equal to/>After difference comparison of the respective entering speeds, the/>A difference in speed;
the expression of the speed difference is:
；
In the method, in the process of the invention, For/>Velocity difference,/>For/>Individual travel-out speed,/>For/>A plurality of entry speeds;
Will be The termination time is respectively equal to/>After difference comparison of the initial moments, the/>The driving time;
The expression of the travel time is:
；
In the method, in the process of the invention, For/>Travel time,/>For/>Time of termination,/>For/>Starting time;
Will be The difference of the speeds is respectively equal to/>After comparing the running time, obtain/>Sub-accelerations;
The expression of the sub acceleration is:
；
In the method, in the process of the invention, For/>Sub-accelerations;
removing sub-accelerations less than a predetermined acceleration threshold, leaving The sub acceleration is accumulated and then averaged to obtain a vehicle acceleration value; the preset acceleration threshold value is the minimum value of the sub-acceleration, so that the numerical value of the sub-acceleration participating in calculation can be limited, the sub-acceleration which has extremely small numerical value and basically does not influence the vehicle acceleration value is removed, the calculated amount is reduced, and the calculation rate is improved; the preset acceleration threshold is obtained by collecting the minimum value of the sub acceleration values when a large number of histories participate in the calculation of the acceleration values of the vehicle;
the expression of the vehicle acceleration value is:
；
In the method, in the process of the invention, For vehicle acceleration value,/>For/>Sub-accelerations;
The vehicle density value refers to the density degree of the number of vehicles on the road in a specific time period, when the vehicle density value is larger, the number of vehicles on the road traffic is larger, the safety distance between the vehicles is smaller, the distance reserved for the vehicles to avoid emergency is shorter, and the safety state of the road traffic is worse;
The method for acquiring the vehicle density value comprises the following steps:
taking the time length corresponding to a preset density period as a standard, and intercepting the time length from the vehicle driving video The same-sized and rectangular images are obtained/>A sub-image; the preset density period is used for intercepting the interval duration of two adjacent sub-images and distinguishing the two adjacent sub-images in time span so as to avoid the number of motor vehicles in the two adjacent sub-images to be in the same state and ensure the independence of the data in each sub-image; the preset density period is set according to a proportion based on the duration corresponding to the vehicle driving video; the duration corresponding to one preset density period is one tenth of the duration corresponding to the vehicle driving video;
Measuring the length of the sub-image and the width of the sub-image through a scale, and calculating the area of the sub-image through a rectangular area formula;
Sequentially counting The number of vehicles in the motor vehicle lane in the individual sub-images is obtained/>A respective vehicle value;
Will be The individual vehicle values are compared with the areas of the sub-images respectively to obtain/>A sub-density value;
The expression of the subdensity value is:
；
In the method, in the process of the invention, For/>Sub-density value,/>For/>Individual vehicle value,/>Is the area of the sub-image;
Removing the maximum and minimum values of the sub-density values, leaving The sub-density values are accumulated and averaged to obtain a vehicle density value;
The expression of the vehicle density value is:
；
In the method, in the process of the invention, Is the vehicle density value,/>For/>A sub-density value;
The vehicle accident triggering rate refers to the probability of occurrence of traffic accidents during the passing period of vehicles on road traffic, and when the vehicle accident triggering rate is larger, the more times of occurrence of traffic accidents during the passing period of vehicles on road traffic are indicated, the greater the probability of occurrence of traffic accidents on current road traffic is, and the worse the safety state of road traffic is;
The method for acquiring the vehicle accident triggering rate comprises the following steps:
A1: identification by monitoring camera License plates of the target vehicles and screening out and/>, from a traffic management databaseTraffic accidents corresponding to license plates of the target vehicles;
A2: recording the occurrence time of traffic accidents, marking the traffic accidents with the occurrence time within a preset accident counting period as effective accidents, and counting the number of the effective accidents; the preset accident counting period is used for limiting the occurrence time of the effective accident so as to ensure that the occurrence time of the effective accident is in a specified time range, thereby representing the accident state of the current target vehicle as far as possible; the preset accident counting period can be set manually; illustratively, a preset accident statistic period is 6 months;
A3: comparing the number of effective accidents with the number of target vehicles to obtain a sub-trigger rate;
a4: repeating execution Steps of sub A1-A3, obtaining/>A sub-trigger rate;
The expression of the sub-trigger rate is:
；
In the method, in the process of the invention, For/>Individual sub-trigger rate,/>For/>The number of valid incidents;
a5: marking the effective triggering rate of the sub-triggering rate larger than a preset sub-triggering threshold value to obtain Effective trigger rate, and will/>The effective triggering rates are accumulated and averaged to obtain the triggering rate of the vehicle accident; the preset sub-trigger threshold is the lowest limit of the sub-trigger rate for participating in the calculation of the vehicle accident trigger rate so as to ensure that the value of the effective trigger rate reaches the calculation requirement and ensure the accurate value of the vehicle accident trigger rate; the preset sub-trigger threshold is obtained through coefficient optimization after collecting the minimum value of the effective trigger rate of a large number of historical participated vehicle accident trigger rates;
The expression of the vehicle accident triggering rate is:
；
In the method, in the process of the invention, For the accident triggering rate of vehicles,/>For/>The effective triggering rate;
The vehicle mixing change degree refers to the change degree of the mixing ratio between the motor vehicle and the non-motor vehicle in the traffic participant on the road traffic, and when the vehicle mixing change degree is larger, the greater the mixing ratio between the motor vehicle and the non-motor vehicle in the traffic participant is, the more complex the type of the traffic participant is, and the worse the safety state of the road traffic is;
The method for acquiring the vehicle mixing variation degree comprises the following steps:
b1: identifying a motor vehicle lane and a non-motor vehicle lane in the sub-image through a computer vision technology, and marking the motor vehicle lane adjacent to the non-motor vehicle lane as a target lane;
B2: at the time T1, counting the number of motor vehicles and the number of non-motor vehicles in a target lane, and marking the number as the initial number of motor vehicles and the initial number of non-motor vehicles respectively;
b3: comparing the initial quantity of the non-motor vehicles with the initial quantity of the motor vehicles to obtain initial mixing degree;
The expression of the initial mix is:
；
In the method, in the process of the invention, For initial mix,/>For the initial quantity of non-motor vehicles,/>Is the initial number of motor vehicles;
B4: at the time T2, counting the number of motor vehicles and the number of non-motor vehicles in a target lane, and marking the number as the end number of motor vehicles and the end number of non-motor vehicles respectively; the time T2 is the time next to the time T1, and the time between the time T2 and the time T1 is enough for the number of the motor vehicles and the number of the non-motor vehicles in the target lane to be changed sufficiently;
B5: comparing the end number of the non-motor vehicles with the end number of the motor vehicles to obtain an end mixing degree;
The expression for ending the degree of mixing is:
；
In the method, in the process of the invention, To end the mixing degree,/>For the end number of non-motor vehicles,/>For the end number of motor vehicles;
B6: comparing the ending mixture with the initial mixture to obtain a sub-mixture;
The expression of the sub-mix is:
；
In the method, in the process of the invention, Is the sub-mix degree;
B7: repeating execution Secondary B1-B6 procedure, obtaining/>Individual sub-mix, will/>The individual sub-mixing degrees are accumulated and averaged to obtain the vehicle mixing variation degree;
The expression of the vehicle mixture variation is:
；
In the method, in the process of the invention, For the degree of change of the vehicle mixture,/>For/>Individual sub-mixes;
The safety state value refers to the safety state of road traffic in a specific time period and a specific road area by traffic participants on the road traffic; the security state values include a high risk state and a low risk state; the safety state value is obtained by collecting road weather, vehicle acceleration value, vehicle density value, vehicle accident triggering rate and vehicle mixing change degree corresponding to the high risk state and the low risk state in an experimental environment.
S2: training a machine learning model for predicting a safety state value based on historical training data;
the training method of the machine learning model for predicting the safety state value comprises the following steps:
sequentially assigning values to road weather; illustratively, a sunny day is assigned to AA, a cloudy day is assigned to BB, a rainy day is assigned to CC, an ice and snow day is assigned to DD, and a windy day is assigned to EE;
Converting the comprehensive safety data into a corresponding group of feature vectors, taking the feature vectors as input of a machine learning model, taking a safety state value corresponding to each group of comprehensive safety data as output of the machine learning model, taking the safety state value as a prediction target, taking the sum of prediction errors of all training data as a training target, and training the machine learning model until the sum of the prediction errors reaches convergence;
illustratively, the machine learning model is any one of a CNN neural network model or AlexNet;
The calculation formula of the prediction error is as follows:
；
In the method, in the process of the invention, For prediction error,/>Group number for feature vector; /(I)For/>Predicted state value corresponding to group feature vector,/>For/>The actual state value corresponding to the group training data;
In the machine learning model, the feature vector is comprehensive safety data, the state value is a safety state value, other model parameters of the machine learning model, the target loss value, the optimization algorithm, the training set test set verification set proportion, the loss function optimization and the like are all realized through actual engineering, and the model is obtained after experimental tuning is continuously carried out.
S3: collecting real-time comprehensive safety data, inputting the real-time comprehensive safety data into a machine learning model after training to predict a real-time safety state value, and judging whether to enter a road early warning mode;
Collecting real-time comprehensive safety data of road traffic, and outputting a safety state value corresponding to the real-time comprehensive safety data through a trained machine learning model;
the security state values include a high risk state and a low risk state;
When the output of the machine learning model is 1, the road traffic safety state is poor, and the real-time safety state value is in a high-risk state; when the output of the machine learning model is 0, the safety state of road traffic is indicated to be good, and the real-time safety state value is in a low-risk state;
When a real-time safety state value is predicted, judging whether a road early warning mode is required to be sent according to the specific risk corresponding to the safety state value, wherein the road early warning mode is a mode which is sent to a traffic management department by a traffic control center and can be used for relieving and coping with the road safety risk when a high risk state occurs, so that the establishment of subsequent response specific measures and instructions is facilitated;
the judging method for entering the road early warning mode comprises the following steps:
When the real-time safety state value is in a high-risk state, the safety state of road traffic is poor, and the probability of traffic accidents of traffic participants is high, judging that the road traffic enters a road early warning mode;
When the real-time safety state value is in a low-risk state, the safety state of road traffic is good, and the probability of traffic accidents of traffic participants is low, judging that the road early warning mode is not entered.
S4: comparing the comprehensive safety data with the corresponding safety value, and marking abnormal data based on the comparison result;
The abnormal data is data which is abnormal after one or more of the comprehensive safety data exceeds the corresponding safety value when the road early warning mode is entered, and when the abnormal data appears, the abnormal data indicates that some specific data in the comprehensive safety data has larger fluctuation, thereby negatively affecting the safety state of road traffic, and therefore the abnormal data needs to be marked;
the method for marking the abnormal data comprises the following steps:
Comparing the vehicle acceleration value with a preset acceleration safety value; the maximum value of the vehicle acceleration value when the preset acceleration safety value safety state value is in a low risk state is used for limiting the upper limit of the vehicle acceleration value, so that the vehicle acceleration value is ensured not to exceed the safety range;
when the vehicle acceleration value is larger than a preset acceleration safety value, the vehicle acceleration value is indicated to exceed the corresponding maximum value when in a low risk state, and the vehicle acceleration value can negatively influence the safety state of road traffic at the moment, and the vehicle acceleration value is marked as abnormal data;
comparing the vehicle density value with a preset density safety value; the preset density safety value is the maximum value of the vehicle density value when the safety state value is in a low risk state, and is used for limiting the upper limit of the vehicle density value, so that the vehicle density value is ensured not to exceed the safety range;
when the vehicle density value is larger than a preset density safety value, the vehicle density value is indicated to exceed the corresponding maximum value when in a low risk state, and the vehicle density value can negatively influence the safety state of road traffic at the moment, and the vehicle density value is marked as abnormal data;
comparing the triggering rate of the vehicle accident with a preset triggering safety value; the preset triggering safety value is the maximum value of the vehicle accident triggering rate when the safety state value is in a low risk state, and is used for limiting the upper limit of the vehicle accident triggering rate, so that the vehicle accident triggering rate is ensured not to exceed the safety range;
When the vehicle accident triggering rate is larger than a preset triggering safety value, the vehicle accident triggering rate is indicated to exceed the corresponding maximum value when in a low risk state, and the vehicle accident triggering rate can negatively influence the safety state of road traffic at the moment, and the vehicle accident triggering rate is marked as abnormal data;
comparing the vehicle mixing change degree with a preset mixing safety value; the preset mixed safety value is the maximum value of the mixed variation degree of the vehicle when the safety state value is in a low risk state, and is used for limiting the upper limit of the mixed variation degree of the vehicle, so that the mixed variation degree of the vehicle is ensured not to exceed the safety range;
When the vehicle mixing change degree is larger than a preset mixing safety value, the vehicle accident triggering rate is indicated to exceed the corresponding maximum value when the vehicle accident triggering rate is in a low risk state, at the moment, the vehicle mixing change degree can negatively influence the safety state of road traffic, and the vehicle mixing change degree is marked as abnormal data.
S5: generating excess ratio values based on the abnormal data, and arranging the excess ratio values in a descending order to mark safety intervention data;
after the abnormal data are marked, the specific numerical value of the abnormal data is required to be analyzed, so that the abnormal data and the corresponding safety value are required to be subjected to difference comparison, and the value after difference is recorded as an excess ratio, so that the excess amplitude of each abnormal data can be accurately reflected;
The excess duty ratio comprises a first duty ratio, a second duty ratio, a third duty ratio and a fourth duty ratio;
the method for generating the first duty ratio, the second duty ratio, the third duty ratio and the fourth duty ratio comprises the following steps:
When the abnormal data is a vehicle acceleration value, comparing the vehicle acceleration value with a preset acceleration safety value, and then comparing the vehicle acceleration value with the vehicle acceleration value to obtain a first duty ratio;
The expression of the first occupancy value is:
；
In the method, in the process of the invention, For the first ratio,/>The acceleration safety value is preset;
When the abnormal data is a vehicle density value, comparing the vehicle density value with a preset density safety value, and then comparing the vehicle density value with the vehicle density value to obtain a second duty ratio;
The expression of the second ratio is:
；
In the method, in the process of the invention, For the second ratio,/>Is a preset density safety value;
When the abnormal data is the vehicle accident triggering rate, comparing the vehicle accident triggering rate with a preset triggering safety value, and then comparing the vehicle accident triggering rate with the vehicle accident triggering rate to obtain a third occupation ratio;
the expression of the third ratio is:
；
In the method, in the process of the invention, For a third ratio,/>The triggering safety value is a preset triggering safety value;
when the abnormal data is the vehicle mixing variation, comparing the vehicle mixing variation with a preset mixing safety value, and then comparing the vehicle mixing variation with the vehicle mixing variation to obtain a fourth duty ratio;
The fourth duty cycle is expressed as:
；
In the method, in the process of the invention, For a fourth duty cycle,/>The triggering safety value is a preset triggering safety value;
After the excess ratio is obtained, analyzing real-time comprehensive safety data according to the size of the excess ratio, so as to mark safety intervention data, wherein the safety intervention data is used as a basis for safety intervention treatment of road traffic by a traffic management department, so that the probability of occurrence of accidents of the road traffic is reduced, the safety of the road traffic is improved, and the excess ratio is required to be arranged in a certain order before the safety intervention data is marked, so that the safety intervention data can be accurately marked in the arranged excess ratio;
the marking method of the safety intervention data comprises the following steps:
when the number of the excess ratio is unique, only one comprehensive safety data can be marked as safety intervention data, and the comprehensive safety data corresponding to the excess ratio is marked as safety intervention data;
When the number of the excess ratio is not the same, the number of the comprehensive safety data which can be marked as the safety intervention data is more than one, the size of the excess ratio is required to be compared, and the excess ratio is arranged in descending order from large to small;
When the maximum value of the excess ratio is the same, marking the comprehensive safety data corresponding to the excess ratio of the first ranking as safety intervention data;
when the maximum value of the excess ratio is not the same, the first bit is ranked in parallel Corresponding to the ratio of excessThe individual integrated security data are marked as security intervention data.
It should be noted that, after the safety intervention data are marked, the traffic control center can make a safety intervention instruction corresponding to the safety intervention data according to the difference of the safety intervention data, and send the safety intervention instruction to the traffic management department, so that the traffic management department can apply the made safety intervention instruction specifically and execute the safety intervention instruction through road traffic management equipment or personnel finally, thereby ensuring that the road traffic is in a high-safety state and greatly reducing the occurrence probability of traffic accidents.
In this embodiment, by collecting historical training data of road traffic, the historical training data includes comprehensive safety data and safety state values, training a machine learning model for predicting the safety state values based on the historical training data, collecting real-time comprehensive safety data, inputting the real-time comprehensive safety data into the machine learning model for predicting the real-time safety state values after training, judging whether to enter a road early warning mode according to the real-time safety state values, comparing the comprehensive safety data with the corresponding safety values, marking abnormal data based on the comparison result, generating excess ratio values based on the abnormal data, arranging the excess ratio values in a descending order, and marking safety intervention data; compared with the prior art, the method has the advantages that the comprehensive safety data influencing the safety state of the road traffic is collected, the safety state of the road traffic can be accurately predicted and estimated based on the machine learning model, when the high risk state of the road traffic occurs, abnormal data which negatively influences the road traffic are marked timely and accurately, the data needing safety intervention are marked accurately according to the specific numerical value of the abnormal data, and the first time of the high risk state of the road traffic is ensured to be provided with targeted intervention measures, so that the safety state of the road traffic is improved.
Example 2: referring to fig. 2, a part of the description of embodiment 1 is not described in detail in this embodiment, and a road traffic safety evaluation system is provided, which is applied to a traffic control center and is used for implementing a road traffic safety evaluation method, and includes a historical data acquisition module, a machine learning model module, a real-time prediction judgment module, an abnormal data marking module and an intervention data marking module, wherein the modules are connected through a wired or wireless network;
the historical data acquisition module is used for acquiring historical training data of road traffic, wherein the historical training data comprises comprehensive safety data and safety state values;
The machine learning model module is used for training a machine learning model for predicting a safety state value based on historical training data;
the real-time prediction judging module is used for collecting real-time comprehensive safety data, inputting the real-time comprehensive safety data into the trained machine learning model to predict a real-time safety state value, and judging whether to enter a road early warning mode or not;
the abnormal data marking module is used for comparing the comprehensive safety data with the corresponding safety value and marking the abnormal data based on the comparison result;
The intervention data marking module is used for generating excess ratio values based on the abnormal data, and marking the safety intervention data by arranging the excess ratio values in a descending order.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention.
Claims (9)
1. A road traffic safety assessment method applied to a traffic control center, comprising:
S1: collecting historical training data of road traffic, wherein the historical training data comprises comprehensive safety data and safety state values, the comprehensive safety data comprises road weather, vehicle acceleration values, vehicle density values, vehicle accident triggering rate and vehicle mixing change degree, and the safety state values comprise high-risk states and low-risk states;
S2: training a machine learning model for predicting a safety state value based on historical training data;
S3: collecting real-time comprehensive safety data, inputting the real-time comprehensive safety data into a machine learning model after training to predict a real-time safety state value, and judging whether to enter a road early warning mode; if the road early warning mode is entered, executing S4-S5; if the road early warning mode is not entered, repeating the step S3;
S4: comparing the comprehensive safety data with the corresponding safety value, and marking abnormal data based on the comparison result;
S5: generating excess ratio values based on the abnormal data, and arranging the excess ratio values in a descending order to mark safety intervention data;
The excess duty ratio comprises a first duty ratio, a second duty ratio, a third duty ratio and a fourth duty ratio;
the marking method of the safety intervention data comprises the following steps:
when the quantity of the excess ratio is unique, marking the comprehensive safety data corresponding to the excess ratio as safety intervention data;
when the number of the excess ratio is not the same, arranging the excess ratio in descending order from large to small;
When the maximum value of the excess ratio is the same, marking the comprehensive safety data corresponding to the excess ratio of the first ranking as safety intervention data;
when the maximum value of the excess ratio is not the same, the first bit is ranked in parallel Corresponding to the ratio of excessThe individual integrated security data are marked as security intervention data.
2. The road traffic safety evaluation method according to claim 1, wherein the vehicle acceleration value acquisition method comprises:
Shooting a vehicle driving video on a road through a monitoring camera, and counting the number of motor vehicles in the vehicle driving video;
when the number of the first-appearing motor vehicles is larger than a preset number threshold value, marking all the motor vehicles in the vehicle driving video as target vehicles to obtain Target vehicles;
by measuring the speed by means of a speed measuring radar Obtaining/>, the speed of each target vehicle when entering the vehicle driving videoIndividual entry speeds and record/>The moment of the entry speed is obtained/>Starting time;
by measuring the speed by means of a speed measuring radar The speed of each target vehicle when the target vehicle exits the vehicle driving video is obtained/>Individual drive-out speeds, and record/>The moment of the individual drive-out speed is obtained/>A termination time;
Will be The respective driving-out speeds are respectively equal to/>After difference comparison of the respective entering speeds, the/>A difference in speed;
the expression of the speed difference is:
；
In the method, in the process of the invention, For/>Velocity difference,/>For/>Individual travel-out speed,/>For/>A plurality of entry speeds;
Will be The termination time is respectively equal to/>After difference comparison of the initial moments, the/>The driving time;
The expression of the travel time is:
；
In the method, in the process of the invention, For/>Travel time,/>For/>Time of termination,/>For/>Starting time;
Will be The difference of the speeds is respectively equal to/>After comparing the running time, obtain/>Sub-accelerations;
The expression of the sub acceleration is:
；
In the method, in the process of the invention, For/>Sub-accelerations;
removing sub-accelerations less than a predetermined acceleration threshold, leaving The sub acceleration is accumulated and then averaged to obtain a vehicle acceleration value;
the expression of the vehicle acceleration value is:
；
In the method, in the process of the invention, For vehicle acceleration value,/>For/>Sub-accelerations.
3. The road traffic safety evaluation method according to claim 2, wherein the vehicle density value acquisition method comprises:
taking the time length corresponding to a preset density period as a standard, and intercepting the time length from the vehicle driving video The same-sized and rectangular images are obtained/>A sub-image;
Measuring the length of the sub-image and the width of the sub-image through a scale, and calculating the area of the sub-image through a rectangular area formula;
Sequentially counting The number of vehicles in the motor vehicle lane in the individual sub-images is obtained/>A respective vehicle value;
Will be The individual vehicle values are compared with the areas of the sub-images respectively to obtain/>A sub-density value;
The expression of the subdensity value is:
；
In the method, in the process of the invention, For/>Sub-density value,/>For/>Individual vehicle value,/>Is the area of the sub-image;
Removing the maximum and minimum values of the sub-density values, leaving The sub-density values are accumulated and averaged to obtain a vehicle density value;
The expression of the vehicle density value is:
；
In the method, in the process of the invention, Is the vehicle density value,/>For/>Sub-density values.
4. A road traffic safety assessment method according to claim 3, wherein said method of obtaining a vehicle accident triggering rate comprises:
A1: identification by monitoring camera License plates of the target vehicles and screening out and/>, from a traffic management databaseTraffic accidents corresponding to license plates of the target vehicles;
A2: recording the occurrence time of traffic accidents, marking the traffic accidents with the occurrence time within a preset accident counting period as effective accidents, and counting the number of the effective accidents;
A3: comparing the number of effective accidents with the number of target vehicles to obtain a sub-trigger rate;
a4: repeating execution Steps of sub A1-A3, obtaining/>A sub-trigger rate;
The expression of the sub-trigger rate is:
；
In the method, in the process of the invention, For/>Individual sub-trigger rate,/>For/>The number of valid incidents;
a5: marking the effective triggering rate of the sub-triggering rate larger than a preset sub-triggering threshold value to obtain Effective trigger rate, and will/>The effective triggering rates are accumulated and averaged to obtain the triggering rate of the vehicle accident;
The expression of the vehicle accident triggering rate is:
；
In the method, in the process of the invention, For the accident triggering rate of vehicles,/>For/>An effective trigger rate.
5. The road traffic safety evaluation method according to claim 4, wherein the vehicle mixture variability obtaining method comprises:
b1: identifying a motor vehicle lane and a non-motor vehicle lane in the sub-image through a computer vision technology, and marking the motor vehicle lane adjacent to the non-motor vehicle lane as a target lane;
B2: at the time T1, counting the number of motor vehicles and the number of non-motor vehicles in a target lane, and marking the number as the initial number of motor vehicles and the initial number of non-motor vehicles respectively;
b3: comparing the initial quantity of the non-motor vehicles with the initial quantity of the motor vehicles to obtain initial mixing degree;
The expression of the initial mix is:
；
In the method, in the process of the invention, For initial mix,/>For the initial quantity of non-motor vehicles,/>Is the initial number of motor vehicles;
B4: at the time T2, counting the number of motor vehicles and the number of non-motor vehicles in a target lane, and marking the number as the end number of motor vehicles and the end number of non-motor vehicles respectively;
B5: comparing the end number of the non-motor vehicles with the end number of the motor vehicles to obtain an end mixing degree;
The expression for ending the degree of mixing is:
；
In the method, in the process of the invention, To end the mixing degree,/>For the end number of non-motor vehicles,/>For the end number of motor vehicles;
B6: comparing the ending mixture with the initial mixture to obtain a sub-mixture;
The expression of the sub-mix is:
；
In the method, in the process of the invention, Is the sub-mix degree;
B7: repeating execution Secondary B1-B6 procedure, obtaining/>Individual sub-mix, will/>The individual sub-mixing degrees are accumulated and averaged to obtain the vehicle mixing variation degree;
The expression of the vehicle mixture variation is:
；
In the method, in the process of the invention, For the degree of change of the vehicle mixture,/>For/>Degree of sub-mixing.
6. The method according to claim 5, wherein the training method of the machine learning model for predicting the safety state value comprises:
Sequentially assigning road weather, assigning sunny day as AA, cloudy day as BB, rainy day as CC, ice and snow day as DD and strong wind day as EE;
the comprehensive safety data are converted into a corresponding group of characteristic vectors, the characteristic vectors are used as input of a machine learning model, the safety state value corresponding to each group of comprehensive safety data is used as output of the machine learning model, the safety state value is used as a prediction target, the sum of prediction errors of all training data is minimized to be used as a training target, and the machine learning model is trained until the sum of the prediction errors reaches convergence, and training is stopped.
7. The method according to claim 6, wherein the real-time safety state value is a high risk state when the output of the machine learning model is 1;
When the output of the machine learning model is 0, the real-time safety state value is in a low risk state;
the judging method for entering the road early warning mode comprises the following steps:
when the real-time safety state value is in a high risk state, judging to enter a road early warning mode;
And when the real-time safety state value is in a low risk state, judging that the road early warning mode is not entered.
8. The road traffic safety assessment method according to claim 7, wherein the marking method of the abnormal data comprises:
Comparing the vehicle acceleration value with a preset acceleration safety value, and marking the vehicle acceleration value as abnormal data when the vehicle acceleration value is larger than the preset acceleration safety value;
comparing the vehicle density value with a preset density safety value, and marking the vehicle density value as abnormal data when the vehicle density value is larger than the preset density safety value;
Comparing the vehicle accident triggering rate with a preset triggering safety value, and marking the vehicle accident triggering rate as abnormal data when the vehicle accident triggering rate is larger than the preset triggering safety value;
and comparing the vehicle mixing variation with a preset mixing safety value, and marking the vehicle mixing variation as abnormal data when the vehicle mixing variation is larger than the preset mixing safety value.
9. The method for evaluating road traffic safety according to claim 8, wherein the generating method of the first, second, third and fourth duty ratios comprises:
When the abnormal data is a vehicle acceleration value, comparing the vehicle acceleration value with a preset acceleration safety value, and then comparing the vehicle acceleration value with the vehicle acceleration value to obtain a first duty ratio;
The expression of the first occupancy value is:
；
In the method, in the process of the invention, For the first ratio,/>The acceleration safety value is preset;
When the abnormal data is a vehicle density value, comparing the vehicle density value with a preset density safety value, and then comparing the vehicle density value with the vehicle density value to obtain a second duty ratio;
The expression of the second ratio is:
；
In the method, in the process of the invention, For the second ratio,/>Is a preset density safety value;
When the abnormal data is the vehicle accident triggering rate, comparing the vehicle accident triggering rate with a preset triggering safety value, and then comparing the vehicle accident triggering rate with the vehicle accident triggering rate to obtain a third occupation ratio;
the expression of the third ratio is:
；
In the method, in the process of the invention, For a third ratio,/>The triggering safety value is a preset triggering safety value;
when the abnormal data is the vehicle mixing variation, comparing the vehicle mixing variation with a preset mixing safety value, and then comparing the vehicle mixing variation with the vehicle mixing variation to obtain a fourth duty ratio;
The fourth duty cycle is expressed as:
；
In the method, in the process of the invention, For a fourth duty cycle,/>Is a preset trigger safety value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410481555.6A CN118097968A (en) | 2024-04-22 | 2024-04-22 | Road traffic safety assessment method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410481555.6A CN118097968A (en) | 2024-04-22 | 2024-04-22 | Road traffic safety assessment method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118097968A true CN118097968A (en) | 2024-05-28 |
Family
ID=91142366
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410481555.6A Pending CN118097968A (en) | 2024-04-22 | 2024-04-22 | Road traffic safety assessment method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118097968A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103258427A (en) * | 2013-04-24 | 2013-08-21 | 北京工业大学 | Urban expressway traffic real-time monitoring system and method based on information physical network |
| CN108335487A (en) * | 2018-03-20 | 2018-07-27 | 北方工业大学 | Road traffic state prediction system based on traffic state time sequence |
| CN113178069A (en) * | 2021-04-20 | 2021-07-27 | 宁波德尔菲信息科技有限公司 | Traffic early warning system based on big data machine learning technology |
| CN116524720A (en) * | 2023-06-19 | 2023-08-01 | 安徽省通信产业服务有限公司 | 5G technology-based integrated intelligent traffic management control system for Internet of vehicles |
| CN117238126A (en) * | 2023-08-29 | 2023-12-15 | 东南大学 | Traffic accident risk assessment method under continuous flow road scene |
| CN117746626A (en) * | 2023-12-18 | 2024-03-22 | 河南省智慧交通研究院有限公司 | Intelligent traffic management method and system based on traffic flow |
-
2024
- 2024-04-22 CN CN202410481555.6A patent/CN118097968A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103258427A (en) * | 2013-04-24 | 2013-08-21 | 北京工业大学 | Urban expressway traffic real-time monitoring system and method based on information physical network |
| CN108335487A (en) * | 2018-03-20 | 2018-07-27 | 北方工业大学 | Road traffic state prediction system based on traffic state time sequence |
| CN113178069A (en) * | 2021-04-20 | 2021-07-27 | 宁波德尔菲信息科技有限公司 | Traffic early warning system based on big data machine learning technology |
| CN116524720A (en) * | 2023-06-19 | 2023-08-01 | 安徽省通信产业服务有限公司 | 5G technology-based integrated intelligent traffic management control system for Internet of vehicles |
| CN117238126A (en) * | 2023-08-29 | 2023-12-15 | 东南大学 | Traffic accident risk assessment method under continuous flow road scene |
| CN117746626A (en) * | 2023-12-18 | 2024-03-22 | 河南省智慧交通研究院有限公司 | Intelligent traffic management method and system based on traffic flow |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102136190B (en) | Dispatching management system and method for event emergency response of urban bus passenger transport | |
| CN107085943B (en) | Short-term prediction method and system for road travel time | |
| CN111325978B (en) | Whole-process monitoring and warning system and method for abnormal behaviors of vehicles on expressway | |
| CN113436432A (en) | Method for predicting short-term traffic risk of road section by using road side observation data | |
| CN111951550B (en) | Traffic safety risk monitoring method and device, storage medium and computer equipment | |
| CN114783183A (en) | Monitoring method and system based on traffic situation algorithm | |
| CN113033840B (en) | Method and device for judging highway maintenance | |
| CN112382086A (en) | Application method of open road test evaluation system of automatic driving automobile | |
| CN116631186B (en) | Expressway traffic accident risk assessment method and system based on dangerous driving event data | |
| CN103425884A (en) | Subway transfer channel traffic state fuzzy prediction method | |
| CN116434523A (en) | Vehicle active safety control method and device based on constraint degree in information perception scene | |
| CN116168356A (en) | Vehicle damage judging method based on computer vision | |
| CN117746626A (en) | Intelligent traffic management method and system based on traffic flow | |
| CN118097968A (en) | Road traffic safety assessment method | |
| CN114627643B (en) | Highway accident risk prediction method, device, equipment and medium | |
| CN116946183A (en) | Commercial vehicle driving behavior prediction method considering driving capability and vehicle equipment | |
| CN116596126A (en) | Bus string prediction method and system | |
| CN113936469B (en) | Traffic information interaction system and method based on highway lane sensing equipment | |
| CN110705854A (en) | Driving level evaluation method and system | |
| CN117033162B (en) | Automatic driving virtual simulation test method and system based on scene generation | |
| CN117633489B (en) | AI-based vehicle safety analysis method and system | |
| CN113688687B (en) | Traffic jam state rapid identification and prediction method based on electric warning data | |
| CN113763708B (en) | Multi-type special lane control method based on traffic simulation | |
| CN116343484B (en) | Traffic accident identification method, terminal and storage medium | |
| CN117104255B (en) | Intelligent perception human-vehicle interaction system and method for intelligent driving vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |