CN112530163B - Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium - Google Patents

Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium Download PDF

Info

Publication number
CN112530163B
CN112530163B CN202011324403.3A CN202011324403A CN112530163B CN 112530163 B CN112530163 B CN 112530163B CN 202011324403 A CN202011324403 A CN 202011324403A CN 112530163 B CN112530163 B CN 112530163B
Authority
CN
China
Prior art keywords
predicted
road section
traffic
accident
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.)
Active
Application number
CN202011324403.3A
Other languages
Chinese (zh)
Other versions
CN112530163A (en
Inventor
李优康
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN202011324403.3A priority Critical patent/CN112530163B/en
Publication of CN112530163A publication Critical patent/CN112530163A/en
Application granted granted Critical
Publication of CN112530163B publication Critical patent/CN112530163B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications

Landscapes

  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The embodiment of the application discloses a traffic accident prediction method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring traffic data and target characteristic data of a road section to be predicted, wherein the target characteristic data is traffic data corresponding to the type of the traffic data influencing the occurrence probability of a traffic accident; acquiring the prediction probability of the traffic accident on the road section to be predicted according to the target characteristic data; and when the prediction probability is greater than the probability threshold value, obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the corresponding running speed of each position. The method can be applied to an intelligent traffic big data system and an automatic driving system, and can be used for obtaining the specific position of the predicted accident occurrence point when the road section to be predicted is predicted to be the road section with the possibility of occurrence of the road accident, so that the instantaneity of obtaining the predicted accident occurrence point is improved.

Description

Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium
Technical Field
The present disclosure relates to the field of internet technologies, and in particular, to a traffic accident prediction method, apparatus, electronic device, and storage medium.
Background
In recent years, with the increase of vehicles, more and more users travel by taking a taxi or driving a car, traffic congestion is getting more and more serious, and causes of traffic congestion are various factors, such as early and late peaks, traffic accidents, climate, road attribute factors or too many traffic lights. In the related art, when it is required to know whether a traffic accident occurs on a certain congested road segment, the method generally depends on accident information reported by users passing through the congested road segment, but most users do not report the accident information when passing through the congested road segment where the accident occurs. Therefore, when a traffic accident occurs on a congested road segment in the related art, there is a problem that the real-time performance of acquiring accident information is poor.
Disclosure of Invention
In view of the above, embodiments of the present application provide a traffic accident prediction method, apparatus, electronic device and storage medium to improve the above problems.
In a first aspect, an embodiment of the present application provides a traffic accident prediction method, where the method includes: acquiring traffic data and target characteristic data of a road section to be predicted, wherein the target characteristic data is traffic data corresponding to the type of the traffic data influencing the occurrence probability of a traffic accident, and the type of the traffic data comprises at least one of road grade type of the road section to be predicted, length type of the road section to be predicted and driving speed type of different positions in the road section to be predicted; acquiring the prediction probability of the traffic accident of the road section to be predicted according to the target characteristic data; and when the prediction probability is greater than a probability threshold value, obtaining a predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the corresponding running speed of each position.
In a second aspect, an embodiment of the present application further provides a traffic accident prediction method, where the method includes: sending an accident prediction request including information of a road section to be predicted to a server, and indicating the server to return a predicted accident occurrence point obtained according to the traffic accident prediction method; receiving a predicted accident occurrence point of the road section to be predicted, which is returned by the server; and displaying the information of the road section to be predicted and the predicted accident occurrence point.
In a third aspect, an embodiment of the present application provides a traffic accident prediction apparatus, including: the device comprises a data acquisition module, a feature extraction module, a probability prediction module and an occurrence point prediction module. The data acquisition module is used for acquiring traffic data and target characteristic data of a road section to be predicted, wherein the target characteristic data is traffic data corresponding to the type of the traffic data influencing the occurrence probability of a traffic accident, and the type of the traffic data comprises at least one of the road grade type of the road section to be predicted, the length type of the road section to be predicted and the driving speed type of different positions in the road section to be predicted; the probability prediction module is used for acquiring the prediction probability of the traffic accident on the road section to be predicted according to the target characteristic data; and the accident occurrence point prediction module is used for obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted and the corresponding running speed of each position included in the traffic data when the prediction probability is greater than the probability threshold.
In a fourth aspect, an embodiment of the present application provides an electronic device, including a processor and a memory; one or more programs are stored in the memory and configured to be executed by the processor to implement the methods described above.
In a fifth aspect, the present application provides a computer-readable storage medium, in which a program code is stored, wherein the program code performs the above-mentioned method when executed by a processor.
In a sixth aspect, the present application provides a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the above-described method.
According to the traffic accident prediction method, the traffic accident prediction device, the electronic equipment and the storage medium, the traffic data and the target characteristic data of the road section to be predicted are obtained, the target characteristic data are the traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident, the prediction probability of the traffic accident occurring on the road section to be predicted is obtained according to the target characteristic data, and when the prediction probability is larger than a probability threshold value, the predicted accident occurring point of the road section to be predicted is obtained according to a plurality of different positions of the road section to be predicted and the corresponding running speed of each position included in the traffic data. Therefore, whether the traffic accident happens to the road section to be predicted is predicted according to the traffic data of the road section to be predicted, the predicted accident occurrence point of the road section to be predicted is predicted after the traffic accident happens, and therefore the instantaneity of obtaining the predicted accident occurrence point is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 shows a block diagram of a system architecture provided in an embodiment of the present application.
Fig. 2 is a flowchart illustrating a traffic accident prediction method according to an embodiment of the present application
FIG. 3 is a flow chart illustrating a traffic accident prediction method according to another embodiment of the present application;
FIG. 4 is a flow chart illustrating a traffic accident prediction method according to yet another embodiment of the present application;
FIG. 5 is a flow chart illustrating a traffic accident prediction method according to yet another embodiment of the present application;
FIG. 6 is a schematic diagram showing the position of a predicted accident occurrence point in a road section to be predicted in the embodiment of the application;
FIG. 7 shows another schematic diagram of the position of the accident occurrence point in the road section to be predicted in the embodiment of the application;
FIG. 8 is a flow chart illustrating a traffic accident prediction method according to yet another embodiment of the present application;
FIG. 9 is a schematic diagram illustrating a prompt message displayed by a terminal in an embodiment of the present application;
FIG. 10 is a schematic diagram illustrating another prompt message displayed by the terminal in the embodiment of the present application;
fig. 11 is a flow chart illustrating a traffic accident prediction method according to yet another embodiment of the present application;
FIG. 12 is a schematic diagram showing a terminal display including a prompt message and a confirmation control in an embodiment of the present application;
FIG. 13 is another diagram illustrating a terminal display including a prompt and confirmation control in an embodiment of the application;
fig. 14 is a timing diagram illustrating a traffic accident prediction method provided in an embodiment of the present application;
fig. 15 is a block diagram showing a traffic accident prediction apparatus according to an embodiment of the present application;
fig. 16 is a block diagram showing a traffic accident prediction apparatus according to another embodiment of the present application;
fig. 17 is a block diagram showing an electronic device for executing a traffic accident prediction method according to an embodiment of the present application;
fig. 18 illustrates a storage unit for storing or carrying program codes for implementing a traffic accident prediction method according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
With the increasing number of motor vehicles and the increasing number of traffic accidents, a large number of people die of the traffic accidents every year, and the traffic accidents cause great emotional and economic cost to people. In the related art, when a congested road segment with a traffic accident is authenticated or reported, the congested road segment is displayed on a navigation interface, otherwise, the navigation interface only displays the congested road segment. However, reporting and authentication of accident information of a traffic accident generally depends on user operations, and many users do not report awareness or habits of the traffic accident. Therefore, the real-time performance of reporting and authenticating accident information of a traffic accident is poor, which may cause that other users may not timely change a driving path when not knowing that a traffic accident occurs on a congested road section after the traffic accident occurs on the congested road section, thereby causing more serious congestion, thereby affecting accident rescue efficiency and causing more serious loss. Therefore, the technical problem to be solved urgently is to improve the instantaneity of obtaining accident information.
The inventor provides a traffic accident prediction method, a device, an electronic device and a storage medium through serious research, in the method, traffic data and target characteristic data of a road section to be predicted are obtained, the target characteristic data are traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident, the prediction probability of the traffic accident occurring on the road section to be predicted is obtained according to the target characteristic data, and when the prediction probability is larger than a probability threshold value, a predicted accident occurring point of the road section to be predicted is obtained according to a plurality of different positions of the road section to be predicted and the driving speed corresponding to each position included in the traffic data. Therefore, when the road section to be predicted is predicted to be the road section possibly having the traffic accident, the specific position of the predicted accident occurrence point is obtained, and the instantaneity of obtaining the predicted accident occurrence point is improved.
It should be understood that the method can be applied to an intelligent traffic big data system and an automatic driving system, so that when the road section to be predicted is predicted to be the road section possibly suffering from a traffic accident, the specific position of the predicted accident occurrence point is obtained, the real-time performance of obtaining the predicted accident occurrence point is improved, meanwhile, the intelligent traffic big data system can be enabled to broadcast the position of the predicted accident occurrence point in time, other users can plan the driving route again according to the broadcast information in time, the automatic driving system can automatically plan a new driving route according to the position of the predicted accident occurrence point, and therefore more serious congestion of the position where the predicted accident occurrence point is located is avoided, the accident rescue efficiency is improved, and more serious loss is avoided.
Embodiments of the present application will be described in detail below with reference to the accompanying drawings.
The scheme provided by the embodiment of the application can be realized by the system architecture shown in fig. 1:
the system architecture comprises a server 10, a terminal 20 and a cloud platform 30, wherein the terminal 20 may comprise a vehicle-mounted terminal or a mobile terminal, and the mobile terminal may be a mobile phone, a tablet computer, and the like.
The cloud platform 30 may store traffic data of different road segments, and may be a cloud server providing basic cloud computing services such as cloud services, a cloud database, cloud computing, a cloud function, cloud storage, a web service, cloud communication, middleware services, a domain name service, a security service, a CDN, and a big data and artificial intelligence platform.
The server 10 may obtain traffic data and target feature data of a road segment to be predicted from traffic data of different road segments stored by the cloud platform 30 upon receiving information of the road segment to be predicted sent by the terminal 20. And then, acquiring the prediction probability of the traffic accident on the road section to be predicted according to the target characteristic data, and when the prediction probability is greater than a probability threshold value, obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted, which are included in the traffic data, and the driving speed corresponding to each position. Therefore, when the road section to be predicted is predicted to be the road section possibly having the road accident, the specific position of the predicted accident occurrence point is obtained, and the real-time performance of obtaining the predicted accident occurrence point is improved.
The terminal 20 may display the road information including the road section of the road section to be predicted and the prompt information of the predicted accident occurrence point, so that the user can plan the path in time according to the specific position of the predicted accident occurrence point and avoid congestion in time, and in addition, the user can plan the path in time and avoid congestion while relieving traffic congestion of the road section to be predicted, so that the rescuer can rapidly arrive at the predicted accident occurrence point for rescue, thereby improving rescue efficiency and reducing casualties and property loss.
Referring to fig. 2, fig. 2 is a flowchart illustrating a traffic accident prediction method according to an embodiment of the present application, where the embodiment describes a server-side processing flow, and the method includes:
s110: and acquiring traffic data and target characteristic data of the road section to be predicted, wherein the target characteristic data is the traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident.
The manner of obtaining the traffic data of the road section to be predicted may be to obtain the traffic data of the road section to be predicted in response to a traffic accident prediction instruction initiated by a client or a server.
As a mode, the traffic data of the road segment to be predicted is acquired in response to a traffic accident prediction instruction initiated by the server, which may specifically be: the method comprises the steps that the server acquires traffic data of different road sections in road network data at preset time intervals, whether a congested road section exists or not is confirmed according to the traffic data of each road section, when the congested road section exists, the congested road section or sub-road sections obtained by dividing the congested road section are used as road sections to be predicted, and traffic data of the road sections to be predicted are acquired.
As a mode, the traffic data of the road section to be predicted is acquired in response to a traffic accident prediction instruction initiated by a terminal, and the method specifically includes: when the terminal obtains the starting position and the end position based on the operation of the user, an accident prediction instruction including the starting position and the end position is generated and sent to the server. When the server receives the accident prediction instruction, the congested road section between the initial position and the terminal position is obtained according to the traffic data of the road section between the initial position and the terminal position in the accident prediction instruction, the congested road section or sub-road sections obtained by dividing the congested road section are used as the road section to be predicted, and the traffic data of the road section to be predicted is obtained. The user operation may include a voice operation, a touch operation, a text input operation, or the like.
When the road section to be predicted is a sub-road section obtained by dividing the congested road section, the corresponding dividing mode may be that the sub-road section obtained by dividing the congested road section is divided according to signs such as traffic lights, intersections, schools and the like, or that the sub-road section obtained by dividing the congested road section is obtained according to a preset length (such as 200 meters, 500 meters, 1000 meters and the like).
The traffic data of the road section to be predicted may include attribute parameters of the road section to be predicted, traffic data corresponding to the road section to be predicted at the prediction time, historical traffic data of the road section to be predicted, and the like.
The attribute parameters of the road section to be predicted may include a type parameter of whether the road section to be predicted is a closed road section type, a road grade parameter of the road section to be predicted, a length of the road section to be predicted, and the like.
It should be understood that the closed road section is a road section which is blocked by traffic and needs to be bypassed, the non-closed road section is a road smooth road section, and different type parameters are respectively corresponding to the road section to be predicted as a closed type and a non-closed type. The road grades are divided into 5 grades of expressways, first-grade roads, second-grade roads, third-grade roads and fourth-grade roads according to the use tasks, functions and adaptive traffic volumes, and different road grade parameters are correspondingly arranged when the road sections to be predicted are different road grades.
The traffic data corresponding to the road section to be predicted at the prediction moment comprises the following steps: the method comprises the steps of calculating the driving speed of the road section to be predicted at the different positions in the prediction moment, the meteorological data corresponding to the road section to be predicted at the prediction moment, the number of vehicles in the road section to be predicted at the prediction moment, the driving speed of the road section to be predicted at the different positions in the upstream and downstream road sections of the road section to be predicted at the prediction moment, and the like, wherein the upstream and downstream road sections comprise the upstream road section and the downstream road section.
The meteorological data corresponding to the road section to be predicted at the prediction time may include parameters such as temperature, humidity and visibility. The driving speeds at different positions may be the driving speeds acquired by speed measuring devices arranged at different positions, or the driving speeds acquired by vehicle-mounted terminals or mobile terminals at different positions.
The traffic data corresponding to the road segment to be predicted at the prediction moment may further include: and obtaining the congestion length of the upstream road section of the road section to be predicted at the prediction time and the slow traveling distance of the upstream road section of the road section to be predicted at the prediction time according to the traveling speeds at different positions in the upstream road section of the road section to be predicted at the prediction time. And obtaining the congestion length of the downstream road section of the road section to be predicted at the prediction time and the slow traveling distance of the downstream road section of the road section to be predicted at the prediction time according to the traveling speeds at different positions in the downstream road section of the road section to be predicted at the prediction time.
The upstream road section refers to a road section which is adjacent to the starting point of the road section to be predicted and has the length within the range of the first length threshold value. The downstream road section is a road section which is adjacent to the end point of the road section to be predicted and has the length within the second length threshold value range. The creep distance refers to the length of a road section of the running speed of the vehicle between a first preset speed threshold and a second preset speed threshold, wherein the first preset speed threshold is smaller than the second speed threshold.
The historical traffic data of the road section to be predicted comprises a plurality of historical moments and traffic data corresponding to each historical moment. Aiming at any target historical time, the traffic data corresponding to the target historical time comprises: weather data of the predicted road section at the target history time, a parameter of whether congestion occurs in the predicted road section at the target history time, a parameter of whether an accident occurs in the predicted road section at the target history time, a traveling speed at different positions in the road section to be predicted at the target history time, the number of vehicles in the road section to be predicted at the target history time, and a traveling speed at different positions in the road section upstream and downstream of the road section to be predicted at the target history time, and the like.
The historical traffic data of the road section to be predicted can further comprise historical congestion frequency of the road section to be predicted, historical accident occurrence frequency of the road section to be predicted, congestion frequency of the road section to be predicted when a traffic accident occurs at the same historical moment as the prediction moment, and accident occurrence frequency of the road section to be predicted corresponding to the upstream and downstream road sections of the road section to be predicted at the same historical moment as the prediction moment.
The historical congestion frequency of the road section to be predicted can be obtained according to whether congestion occurs in the road section to be predicted at different historical moments. The historical accident occurrence frequency of the road section to be predicted can be obtained according to whether traffic accidents occur on the road section to be predicted at different historical moments. The congestion frequency of the road section to be predicted when the traffic accident occurs at the same historical time as the prediction time can be obtained according to whether the road section to be predicted is congested and whether the traffic accident occurs at the same historical time as the prediction time. The frequency of accidents of the road section to be predicted corresponding to the upstream and downstream road sections at the same historical time as the prediction time can be obtained according to whether the traffic accidents occur on the upstream and downstream road sections of the road section to be predicted at the same historical time as the prediction time.
The types of traffic data affecting the probability of occurrence of a traffic accident may be obtained in various manners, for example, the types may be obtained by analyzing and comparing a large amount of sample traffic data, or may be obtained by setting by research personnel according to experience.
Specifically, when the type of traffic data affecting the probability of occurrence of a traffic accident is obtained by analyzing and comparing sample traffic data, the sample traffic data should include a plurality of positive sample traffic data when a traffic accident occurs and a plurality of negative sample traffic data when no traffic accident occurs. And determining the type of the traffic data influencing the occurrence probability of the traffic accident according to the deviation ratio calculation result by calculating the deviation ratio of the data belonging to the same type in the positive sample traffic data and the negative sample traffic data.
As one way, the category of the traffic data may include at least one of a road class category of the section to be predicted, a length category of the section to be predicted, and a travel speed category at different positions in the section to be predicted. It should be understood that the categories of traffic data may also include more.
As still another way, the type of the traffic data affecting the occurrence probability of the traffic accident may include at least one of a road class type of a road section to be predicted, whether the road section to be predicted is a closed road type, a length type of the road section to be predicted, a cruising distance type of an upstream section of the road section to be predicted, a cruising distance type of a downstream section of the road section to be predicted, a road condition type of the road section to be predicted, a historical clear frequency type of the road section to be predicted, a historical cruising frequency type of the road section to be predicted, a congestion frequency type of the road section to be predicted, a traveling speed type at different positions in the road section to be predicted, a historical accident occurrence frequency type of the upstream and downstream sections of the road section to be predicted, a type of the road section to be predicted corresponding to a frequency of an accident occurring at an upstream section and a downstream section thereof at the same historical time as the time of prediction, and the like.
S120: and acquiring the prediction probability of the traffic accident on the road section to be predicted according to the target characteristic data.
The above-mentioned manner of obtaining the prediction probability of the traffic accident occurring on the road section to be predicted according to the target characteristic data may be various.
As one mode, the target feature data may be subjected to weight calculation to obtain a prediction probability of a traffic accident occurring on the road section to be predicted.
In the method, each kind of target characteristic data is respectively configured with a weight coefficient, the weight corresponding to each kind of target characteristic data is searched from a preset weight relation table, and the prediction probability of the traffic accident on the road section to be predicted is obtained through calculation according to the weight corresponding to each kind of target characteristic data and the weight coefficient. The preset weight relation table can be stored in a server or a cloud platform, and comprises a plurality of kinds of characteristic data and weights respectively corresponding to the characteristic data at different values.
As another mode, the preset model may be utilized to input the target feature data into the preset model, so as to obtain the prediction probability of the traffic accident occurring on the road segment to be predicted.
The preset model can be obtained through sample traffic data training. The preset model may be a neural network model, such as a decision tree model or a regression model.
S130: and when the prediction probability is greater than the probability threshold value, obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the respective corresponding running speeds of each position.
The probability threshold may be preset, for example, 0.5, 0.7, or 0.8, and may be obtained according to a predicted time and a probability setting rule corresponding to the predicted time. The distance between any two adjacent positions of the plurality of different positions may be the same.
The manner of obtaining the predicted accident occurrence point of the road section to be predicted according to the plurality of different positions of the road section to be predicted included in the traffic data and the driving speed corresponding to each position can be various.
As one mode, it is considered that the speed change is large before and after the vehicle passes the accident occurrence point. That is, the speed may gradually decrease as the vehicle travels to approach the accident occurrence point, and the speed may rapidly increase after passing the accident occurrence point. Therefore, the predicted accident occurrence point obtained according to the plurality of positions in the road section to be predicted and the running speed corresponding to each position may be: obtaining a speed fitting curve according to the speeds of different positions in the road section to be predicted, obtaining a target curve section with gradually increased speed from the speed fitting curve, obtaining a target position with a speed change rate larger than a preset change rate threshold value from the target curve section, and determining a predicted accident occurrence point according to the target position (for example, taking the position, closest to the starting point of the road section to be predicted, in the target position as the predicted accident occurrence point).
As another mode, when the road segment to be predicted is a congested road segment, the predicted accident occurrence point of the road segment to be predicted may be obtained according to the length of the road segment to be predicted, a plurality of different positions of the road segment to be predicted, and the respective corresponding traveling speeds of each of the positions.
Specifically, within a certain time period from the accident occurrence time to the prediction time, as the time length between the prediction time and the accident occurrence time increases, the length of the congested road section gradually increases, and the distance between the accident occurrence point and the ending point of the congested road section also gradually increases. Therefore, the specific step of obtaining the predicted accident occurrence point of the to-be-predicted road section according to the length of the congested road section, the plurality of different positions of the to-be-predicted road section, and the respective corresponding running speeds of each position may be: and obtaining a first distance between the predicted accident occurrence point and the end point of the road section to be predicted according to the corresponding relation between the preset road section length and the distance value and the length of the road section to be predicted, wherein the corresponding relation between the preset road section length and the distance value comprises a plurality of congestion road section lengths and a distance corresponding to each congestion road section length.
Obtaining a speed fitting curve according to different positions in the road section to be predicted and the speed corresponding to each position, obtaining a target curve section with gradually increased speed from the speed fitting curve, obtaining a target position with a speed change rate larger than a preset change rate threshold value from the target curve section, obtaining a second distance between the target position and the end point of the road section to be predicted, weighting the first distance and the second distance to obtain a target distance, and determining the predicted accident occurrence point according to the target distance and the distance between the different positions in the road section to be predicted and the end point of the road section to be predicted.
As another mode, the speed change rate of the road section to be predicted between any two adjacent positions can be obtained according to a plurality of different positions of the road section to be predicted and the respective corresponding driving speeds of each position, a target speed change rate can be obtained according to the speed change rate between any two adjacent positions, and the predicted accident occurrence point can be determined according to the position corresponding to the target speed change rate.
The target speed change rate may be a speed change rate greater than a change rate threshold value among the plurality of speed change rates, or may be a maximum speed change rate among the plurality of speed change rates. The predicted accident occurrence point may be a position closest to the start point of the road segment to be predicted in the positions corresponding to the target speed change rates, or a midpoint of a connection line between two positions having the largest distance in at least two positions corresponding to the target speed change rates, as the predicted accident occurrence point.
As another mode, the acceleration of the road section to be predicted between any two positions can be obtained according to the multiple positions of the road section to be predicted and the corresponding running speed of each position, and a target acceleration can be obtained according to the acceleration between any two positions, so that the predicted accident occurrence point is determined according to the position corresponding to the target acceleration.
The target acceleration may be an acceleration greater than an acceleration threshold among the plurality of accelerations, or may be a maximum acceleration among the plurality of accelerations. The predicted accident occurrence point may be a position closest to the start point of the road segment to be predicted in the positions corresponding to the target acceleration as the predicted accident occurrence point, or a midpoint of a connection line between two positions having the largest distance in at least two positions corresponding to the target acceleration as the predicted accident occurrence point.
The traffic accident prediction method provided by this embodiment obtains traffic data and target feature data of a road segment to be predicted, where the target feature data is traffic data corresponding to a type of the traffic data that affects a probability of occurrence of a traffic accident, obtains a prediction probability of occurrence of the traffic accident on the road segment to be predicted according to the target feature data, and obtains a predicted accident occurrence point of the road segment to be predicted according to a plurality of different positions of the road segment to be predicted included in the traffic data and a respective corresponding driving speed of each of the positions when the prediction probability is greater than a probability threshold. The method can be applied to an intelligent traffic big data system and an automatic driving system, and can be used for obtaining the specific position of the predicted accident occurrence point when the road section to be predicted is predicted to be the road section with the possibility of occurrence of the road accident, so that the instantaneity of obtaining the predicted accident occurrence point is improved.
In addition, the instantaneity of acquiring the accident occurrence point to be predicted is improved, so that a user can plan a path in time according to the predicted accident occurrence point and avoid congestion, traffic congestion of a road section to be predicted can be relieved, rescuers can rapidly arrive at the predicted accident occurrence point for rescue, rescue efficiency is improved, and casualties and property loss are reduced.
As a mode, when the road section to be predicted includes a plurality of sub-road sections obtained by dividing a congested road section according to signs such as traffic lights, intersections, schools and the like or preset lengths, the correspondingly acquired traffic data of the road section to be predicted includes traffic data of each sub-road section, and the acquired target feature data of the road section to be predicted includes target feature data of each sub-road section.
In order to improve the reliability of the obtained predicted accident occurrence point, in this way, the prediction probability of the traffic accident occurrence of each sub-road section can be obtained according to the target characteristic data of each sub-road section; and then, taking the sub-road section corresponding to the maximum prediction probability in the prediction probabilities respectively corresponding to the sub-road sections as a new road section to be predicted, and taking the maximum prediction probability as the prediction probability of the traffic accident of the road section to be predicted. And finally, when the prediction probability is larger than the probability threshold value, obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the corresponding running speed of each position.
By adopting the method, the congested road section can be divided into a plurality of sub-road sections under the condition that the congested road section is longer, so that the target sub-road section with the highest prediction probability is determined, namely, the target sub-road section is the sub-road section which is most likely to have an accident in the congested road section, and therefore the predicted accident occurrence point in the target sub-road section is obtained according to the traffic data corresponding to the target sub-road section, and the reliability of the obtained predicted accident occurrence point is improved. Therefore, the rescue personnel can further rapidly arrive at the predicted accident occurrence point for rescue, and the rescue efficiency is improved to reduce casualties and property loss.
Referring to fig. 3, fig. 3 is a schematic flow chart of a traffic accident prediction method according to another embodiment of the present application, which describes a server-side processing flow, and the method includes:
step S210: the method comprises the steps of obtaining sample traffic data, wherein the sample traffic data comprise positive sample traffic data and negative sample traffic data, each sample traffic data corresponds to a data obtaining moment, the positive sample traffic data are sample traffic data with traffic accidents in a first time length range from the data obtaining moment of the positive sample traffic data, and the negative sample traffic data are sample traffic data without traffic accidents in a second time length range from the data obtaining moment of the negative sample traffic data.
Wherein, the first preset time period may be 10 minutes, 20 minutes or 1 hour, etc., and the second preset time period may be 2 hours, 3 hours or 3 hours, etc.
The sample traffic data may include a length of the sample section, a road grade parameter of the sample section, a type parameter of whether the sample section is of a closed section type at the sample acquisition time, a driving speed at different positions in the sample section at the sample acquisition time, meteorological data corresponding to the sample section at the sample acquisition time, whether the sample section corresponds to a traffic accident, a driving speed at different positions in upstream and downstream sections of the sample section, and the like.
In order to ensure that there is no sample traffic data with high similarity in the acquired sample traffic data, and thus ensure the reliability of the type of traffic data that affects the probability of occurrence of a traffic accident and is obtained by using the sample traffic data, in one aspect, the step S210 may include: and selecting a sample path from the plurality of paths, wherein the traffic flow within a third preset time range is greater than a traffic flow threshold, the road grade belongs to a preset grade, and the sample traffic data is obtained from historical traffic data corresponding to the sample path.
The preset levels may include a highway, a first level, and a second level, among others. The third preset time period may be one day, one week, one month, or the like. It should be understood that the corresponding traffic threshold is larger when the third predetermined period is longer, for example, the traffic threshold may be 10000 when the third predetermined period is one day, and the traffic threshold may be 50000 when the third predetermined period is one week.
In addition, in order to further ensure that repeated sample traffic data does not exist in the obtained sample traffic data, the reliability of the type of the obtained traffic data which influences the occurrence probability of the traffic accident is higher. In this embodiment, when at least two sample traffic data belonging to the same sample road segment exist in the acquired sample traffic data, the time length between the data acquisition time of any two sample traffic data in the at least two sample traffic data is greater than a fourth preset time length threshold.
To ensure the reliability of traffic accidents in the positive samples taken. The positive sample traffic data can be historical traffic data with traffic accidents reported by users, and can also be historical traffic data with traffic accidents when the number of user authentications reaches a preset threshold and the authentication result is the preset threshold.
The method aims to avoid the problem that the reliability of target characteristic data obtained by calculating the positive sample and the negative sample is influenced due to overlarge data quantity deviation of the traffic data of the positive sample and the negative sample. In one embodiment, the number of positive samples and the number of negative samples tend to be the same, or the ratio of the number of positive samples to the number of negative samples is within a preset ratio threshold.
Step S220: and analyzing and comparing the positive sample traffic data and the negative sample traffic data to obtain the types of the traffic data which influence the occurrence probability of the traffic accident.
The type of the traffic data affecting the occurrence probability of the traffic accident may include at least one of a road grade type of the road segment to be predicted, a length type of the road segment to be predicted, and a driving speed type at different positions in the road segment to be predicted.
As one way, the step S220 may be to calculate a deviation ratio of the traffic data of the same kind in the positive sample traffic data and the negative sample traffic data to obtain a deviation ratio of each kind of traffic data; and taking the type of the data corresponding to the traffic data with the deviation ratio larger than the second ratio threshold value as the type of the traffic data influencing the occurrence probability of the traffic accident.
Alternatively, in step S220, a deviation ratio of the same type of data in the positive sample traffic data and the negative sample traffic data may be calculated to obtain a deviation ratio of each type of data, the types may be sorted in an order of decreasing deviation ratio of each type of data, and the type of data corresponding to the deviation ratio with the previously set sorting order value may be selected as the type of traffic data affecting the traffic accident occurrence probability.
Step S230: and acquiring traffic data and target characteristic data of the road section to be predicted, wherein the target characteristic data is the traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident.
Step S240: and acquiring the prediction probability of the traffic accident on the road section to be predicted according to the target characteristic data.
Step S250: and when the prediction probability is greater than the probability threshold value, obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the respective corresponding running speeds of each position.
The traffic accident prediction method provided by this embodiment determines the type of traffic data affecting the accident occurrence probability according to the acquired sample traffic data, acquires the traffic data of the road segment to be predicted and the target feature data, where the target feature data is the traffic data corresponding to the type of the traffic data affecting the traffic accident occurrence probability, acquires the prediction probability of the traffic accident occurrence on the road segment to be predicted according to the target feature data, and obtains the predicted accident occurrence point of the road segment to be predicted according to a plurality of different positions of the road segment to be predicted included in the traffic data and the respective corresponding driving speeds of each of the positions when the prediction probability is greater than a probability threshold. The method can be applied to an intelligent traffic big data system and an automatic driving system, and can be used for obtaining the specific position of the predicted accident occurrence point when the road section to be predicted is predicted to be the road section with the possibility of occurrence of the road accident, so that the instantaneity of obtaining the predicted accident occurrence point is improved.
Referring to fig. 4, fig. 4 is a schematic flow chart of a traffic accident prediction method according to another embodiment of the present application, which describes a server-side processing flow, and the method includes:
step S310: the method comprises the steps of obtaining sample traffic data, wherein the sample traffic data comprise positive sample traffic data and negative sample traffic data, the time when a traffic accident occurs exists within a first preset time range from the data obtaining time of the positive sample traffic data, and the time when the traffic accident does not occur within a second preset time range from the data obtaining time of the negative sample traffic data.
Step S320: and analyzing and comparing the positive sample traffic data and the negative sample traffic data to obtain the types of the traffic data which influence the occurrence probability of the traffic accident.
Step S330: and acquiring traffic data and target characteristic data of the road section to be predicted, wherein the target characteristic data is the traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident.
The type of the traffic data affecting the occurrence probability of the traffic accident may include at least one of a road grade type of the road segment to be predicted, a length type of the road segment to be predicted, and a driving speed type at different positions in the road segment to be predicted.
Step S340: and inputting the target characteristic data into a preset model to obtain the prediction probability of the traffic accident on the road section to be predicted, and establishing the preset model based on the sample traffic data.
In one mode, the target feature data is multiple types, and the step S340 may include: and performing characteristic construction on the various target characteristic data to obtain construction characteristic data, and inputting the construction characteristic data into a preset model to obtain the prediction probability of the traffic accident on the road section to be predicted.
The characteristic construction of the multiple target characteristic data to obtain the construction characteristic data may be: and grouping the multiple target characteristic data to obtain multiple groups of target characteristic data, and performing weight calculation on at least one target characteristic data included in each group of target characteristic data to obtain multiple structural characteristic data. The feature constructing the target feature data to obtain the constructed feature data may further be: and respectively carrying out normalization processing on the plurality of target characteristic data to obtain structural characteristic data respectively corresponding to each target characteristic data.
As one mode, grouping a plurality of target feature data and then obtaining a plurality of structural feature data by adopting weight calculation includes: and performing weight calculation on whether the road section to be predicted in the target characteristic data is a closed road section type parameter at the prediction time, the road grade parameter of the road section to be predicted and the length of the road section to be predicted to obtain attribute construction characteristic data of the road section to be predicted at the prediction time. And performing weight calculation on the upstream crawling distance of the road section to be predicted and the downstream crawling distance of the road section to be predicted in the target characteristic data to obtain the spatial congestion construction characteristic data of the road section to be predicted at the prediction time. And performing weight calculation on the historical smooth frequency of the road section to be predicted, the historical slow-moving frequency of the road section to be predicted and the historical congestion frequency of the road section to be predicted in the target characteristic data to obtain road condition speed construction characteristic data of the road section to be predicted. And performing weight calculation on the frequency of accidents of each road section in the upstream and downstream road sections of the road section to be predicted in the target characteristic data to obtain road condition construction characteristic data of the upstream and downstream road sections of the road section to be predicted. And performing weight calculation on the frequency of accidents occurring at the same historical time as the prediction time of each road section in the upstream and downstream road sections of the road section to be predicted in the target characteristic data to obtain frequency construction characteristic data of the accidents occurring at the same historical time as the prediction time of the upstream and downstream road sections of the road section to be predicted. And performing weight calculation on the frequency of congestion when an accident occurs at the historical time when each road section in the upstream and downstream road sections of the road section to be predicted is the same as the prediction time in the plurality of target characteristic data to obtain frequency construction characteristic data of the congestion when the accident occurs at the historical time when the upstream and downstream road sections of the road section to be predicted are the same as the prediction time.
Accordingly, the structural feature data obtained in this manner includes: the method comprises the following steps of constructing feature data according to attributes of a road section to be predicted at a prediction time, constructing feature data of space congestion of the road section to be predicted at the prediction time, constructing feature data of road speed of the road section to be predicted, constructing feature data of road conditions of an upstream road section and a downstream road section of the road section to be predicted, constructing feature data of frequency of accidents occurring at the same historical time of the upstream road section and the downstream road section of the road section to be predicted as well as the prediction time, and constructing feature data of frequency of congestion occurring at the same historical time of the upstream road section and the downstream road section of the road section to be predicted as the prediction time.
And inputting the acquired construction characteristic data into a preset model to obtain the prediction probability of the traffic accident on the road section to be predicted.
The preset model may be a regression model, which may be constructed by:
and acquiring sample characteristic data corresponding to the type of the traffic data influencing the traffic accident occurrence probability in each sample traffic data, and performing characteristic construction on the sample characteristic data to obtain sample construction characteristic data. Establishing a multiple linear regression equation based on sample construction characteristic data corresponding to each sample traffic data, carrying out equation coefficient significance test on the multiple linear regression equation to obtain a coefficient significance test result, and finally determining a regression model according to the significance test result.
Specifically, the number of factors (sample configuration feature data) causing the occurrence of a road traffic accident is generally not less than two. Thus, feature data (x) can be constructed using multiple linear regression analysis, assuming random variable y and a plurality of samples that lead to the occurrence of a road traffic accident1,x2,…,xm) Linear correlation, then the multiple linear regression formula can be obtained as:
Figure BDA0002793881380000151
wherein, b0Is a constant coefficient, biAnd m is a partial regression coefficient of the equation and is the number of the sample construction characteristic data.
When the significance of each partial regression coefficient is tested by using the partial regression coefficient test principle, the statistic F is selectedi，
Figure BDA0002793881380000152
Detecting to obtain the detection result of partial regression coefficient, wherein Vi(i ═ 1, 2, …, m) is the partial regression sum of squares, Q is the residual sum of squares, m is the number of counts of the sample formation features, n is the number of sets of linear regression equations, when F isiExceeds a critical value F1-a(1, n-m-1), in the negative hypothesis, x is describediAnd the function on the equation y is not significant, so that when the detection result difference of each partial regression coefficient exceeds a critical value, the partial regression coefficient is adjusted to obtain a corrected partial regression coefficient so as to complete the test, and the following preset regression model is obtained:
Figure BDA0002793881380000153
wherein, betaiIs a constant coefficient, betaiIs the corrected partial regression coefficient.
After the regression model is built, the constructed feature data obtained according to the target feature data is input into the regression model, and then the probability of traffic accidents on the road section to be predicted can be obtained.
When the structural characteristics are that the target characteristic data are normalized, after the establishment of a regression model is completed, each target characteristic data is normalized, regression calculation is carried out on each processed target characteristic data by adopting a partial regression coefficient corresponding to each target characteristic data, and the probability of traffic accidents on the road section to be predicted is obtained.
As one mode, the partial regression coefficient corresponding to the target feature data is a corrected partial regression coefficient, and the mode of performing regression calculation on each processed target feature data and the partial regression coefficient corresponding to each target feature data may be: and calculating the processed target characteristic data and the corrected partial regression coefficients corresponding to the target characteristic data by adopting a preset calculation formula to obtain the probability of the traffic accident on the road section to be predicted.
Wherein the preset calculationIs of the formula
Figure BDA0002793881380000161
xiFor the ith normalized target feature data, betaiCorrected partial regression coefficients, a and beta, corresponding to the ith normalized target feature data0Respectively, are constant coefficients.
Step S350: and when the prediction probability is greater than the probability threshold value, obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the respective corresponding running speeds of each position.
The traffic accident prediction method provided by this embodiment determines the type of traffic data affecting the accident occurrence probability according to the acquired sample traffic data, acquires the traffic data of the road section to be predicted and the target feature data corresponding to the type of the traffic data affecting the traffic accident occurrence probability, inputs the target feature data into a preset model obtained by using the sample traffic data to obtain the prediction probability of the traffic accident occurring on the road section to be predicted, and obtains the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the respective corresponding driving speed of each position when the prediction probability is greater than a probability threshold. The method can be applied to an intelligent traffic big data system and an automatic driving system, and can be used for obtaining the specific position of the predicted accident occurrence point when the road section to be predicted is predicted to be the road section with the possibility of occurrence of the road accident, so that the instantaneity of obtaining the predicted accident occurrence point is improved.
Referring to fig. 5, fig. 5 is a schematic flow chart of a traffic accident prediction method according to another embodiment of the present application, which describes a server-side processing flow, and the method includes:
step S410: and acquiring traffic data and target characteristic data of the road section to be predicted, wherein the target characteristic data is the traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident.
The type of the traffic data comprises at least one of a road grade type of the road section to be predicted, a length type of the road section to be predicted and a driving speed type at different positions in the road section to be predicted.
Step S420: and acquiring the prediction probability of the traffic accident on the road section to be predicted according to the target characteristic data.
Step S430: and obtaining the speed change rate of any two adjacent positions according to the corresponding running speed of each position, and obtaining the target speed change rate according to the speed change rate of any two adjacent positions.
As one way, the target speed change rate may be the maximum speed change rate among the speed change rates of any two adjacent positions.
Step S440: it is determined whether the target speed rate of change is less than or equal to a rate of change threshold.
Step S450: and when the target speed change rate is less than or equal to the change rate threshold value, acquiring the predicted accident occurrence point of the road section to be predicted according to the road type, the road grade and the length of the road section to be predicted.
Referring to fig. 6, as one way, a path length L1 between a predicted accident occurrence point and a start point of a to-be-predicted link may be obtained by L1/L0, where a1 × Rt + a2 × Rg + a3 × L0, where L0 is the length of the to-be-predicted link, a1, a2, and a3 are constant coefficients, Rt is a road type parameter of the to-be-predicted link, and Rg is a road grade parameter of the to-be-predicted link, and according to the path length L1, the predicted accident occurrence point corresponding to the path length is searched from a correspondence table, where a plurality of positions and the path length between each position and the start point of the to-be-predicted link are stored in the correspondence table.
Step S460: and when the target speed change rate is greater than the change rate threshold value, obtaining a speed fitting curve of the road section to be predicted according to the driving speed corresponding to each position in the road section to be predicted.
Step S470: and acquiring the position of the minimum speed value in the speed fitting curve corresponding to the road section to be predicted as a target position, acquiring the path length from the target position to the starting point of the road section to be predicted, and calculating the ratio of the path length to the length of the road section to be predicted.
Step S480: and obtaining the slope of a tangent line tangent to the speed fitting curve according to the ratio, and taking the position of a tangent point of the tangent line tangent to the fitting curve, which corresponds to the road section to be predicted, as a predicted accident occurrence point.
Referring to fig. 7, as one way, a coordinate system is established with the starting point of the road segment to be predicted as the origin of coordinates, the length L0 of the road segment to be predicted as the abscissa, and the driving speed as the ordinate. When the target speed change rate is greater than the change rate threshold, according to the driving speed corresponding to each position in the road section to be predicted, a speed fitting curve L1 of the road section to be predicted in fig. 7 is obtained, the distance from the position corresponding to the minimum speed value in the fitting curve to the starting point of the road section to be predicted is a distance L2, a ratio can be obtained according to the distance L2 and the length L0 of the road section to be predicted, a tangent line tangent to the speed fitting curve obtained according to the ratio is shown as L2 in fig. 8, an abscissa corresponding to a tangent point where the fitting curve L1 is tangent to a tangent line L2 is the distance L1 from the starting point of the road section to be predicted to the accident occurrence point, and the tangent point is the predicted accident occurrence point.
Wherein the slope of the tangent line tangent to the velocity-fitted curve derived from the ratio may be: when the ratio is less than or equal to a first ratio threshold, taking the ratio as the slope of a tangent line tangent to the speed fitting curve; and when the ratio is larger than the first ratio threshold, averaging the ratio threshold and the ratio, and taking the average as the slope of a tangent line tangent to the fitting curve. Wherein the first ratio threshold may be, but is not limited to, 0.8, 0.85, or 0.9.
According to the traffic accident prediction method provided by the embodiment, the traffic data and the target characteristic data of the road section to be predicted are obtained, the prediction probability of the traffic accident of the road section to be predicted is obtained according to the target characteristic data, when the prediction probability is larger than a probability threshold value, the fact that all vehicles possibly exist in a stop state in the road section to be predicted, namely, the speed corresponding to each position tends to be zero is considered, therefore, the predicted accident occurrence point of the road section to be predicted is obtained according to the road type, the road grade and the length of the road section to be predicted when the speed change of all vehicles is small, and the accident occurrence point is determined by using the fitting curve when the speed change of the road section to be predicted is large, so that the real-time performance of obtaining the predicted accident occurrence point is improved, and the accuracy of the predicted accident occurrence point is also improved.
Referring to fig. 8, fig. 8 is a schematic flow chart of a traffic accident prediction method according to another embodiment of the present application, which describes a server-side processing flow, and the method includes:
step S510: and acquiring traffic data and target characteristic data of the road section to be predicted, wherein the target characteristic data is the traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident.
The type of the traffic data comprises at least one of a road grade type of the road section to be predicted, a length type of the road section to be predicted and a driving speed type at different positions in the road section to be predicted.
Step S520: and acquiring the prediction probability of the traffic accident on the road section to be predicted according to the target characteristic data.
Step S530: and when the prediction probability is greater than the probability threshold value, obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the respective corresponding running speeds of each position.
Step S540: and receiving the target road section information sent by the terminal.
The method for the terminal to send the target road section information may be road section information of a congested road section between a starting position and an ending position obtained based on an operation of a user, or road section information of a congested road section selected based on an operation of the user, where the operation of the user may be a voice control operation or a touch control operation.
Step S550: and if the target road section corresponding to the target road section information belongs to the road section to be predicted and the target road section corresponds to the predicted accident occurrence point, generating prompt information comprising the predicted accident occurrence point of the target road section and the road name of the target road section, and sending the prompt information to the terminal.
By one approach, prompt information may be generated that includes the source of the accident as a predicted accident, the type of accident as a traffic accident, and the predicted point of occurrence of the accident. Therefore, the predicted accident occurrence point in the road section to be predicted is explained for the user, and is not the event actually reported by the user. And when the terminal receives the prompt information, displaying the prompt information, so that a user can plan a driving route to avoid congestion conveniently.
Specifically, referring to fig. 9, when receiving the prompt message, the terminal displays "forecast accident" in a manner of a title and "XX position of XX road segment" in a manner of a subtitle on the display interface, so as to show the source of the prompt message, the name of the road segment to be forecasted, and the specific position of the accident occurrence point in the road segment to be forecasted to the user.
To implement updating or authenticating the predicted accident occurrence point, in one embodiment, the traffic accident prediction method further includes:
and the receiving terminal receives confirmation information fed back based on the prompt information and authenticates the predicted accident occurrence point according to the confirmation information.
As one mode, when the confirmation information includes the accident prediction abnormality information, the predicted accident occurrence point corresponding to the target link may be deleted.
It should be understood that when the confirmation information includes the accident prediction abnormal information, the characterization accident prediction result is inaccurate, that is, no accident occurs, and when the predicted accident occurrence point corresponding to the target road segment is deleted, the situation that the user detours due to inaccurate prediction result of the road segment to be predicted when other terminals acquire the corresponding road segment to be predicted subsequently can be avoided.
As another embodiment, when the confirmation information includes accident prediction accuracy information, authentication times with accurate prediction can be added to the predicted accident occurrence point; and adding the number of authentication times of the predicted abnormality for the predicted accident occurrence point when the confirmation information includes the accident prediction abnormality information.
By performing authentication for the predicted accident occurrence point according to the confirmation information, when target road section information including the road section to be predicted and sent by other terminals is subsequently received, prompt information including the fact that the topic is a suspected accident, and the side topic is an authentication result, the predicted accident occurrence point of the target road section and the road name of the target road section can be fed back.
As shown in fig. 10, the authentication result includes the number of authentications for prediction accuracy and the number of authentications for prediction abnormality. When the number of authentication times for predicting the accident occurrence point accurately is 5 times and the number of authentication times for predicting the abnormality is 2 times, the "predicted accident" is displayed in the display interface in the form of a title, "XX position of XX link is displayed in the form of a subtitle," useful number 5 corresponding thereto is displayed in the form of a floating window, and useless number 2 corresponding thereto is displayed in the form of a floating window.
The traffic accident prediction method provided by this embodiment obtains the predicted probability of a traffic accident occurring on a road segment to be predicted according to target characteristic data by obtaining traffic data of the road segment to be predicted and target characteristic data, obtains a predicted accident occurring point of the road segment to be predicted according to a plurality of different positions of the road segment to be predicted included in the traffic data and a driving speed corresponding to each of the positions when the predicted probability is greater than a probability threshold, and sends prompt information including the predicted accident occurring point to a terminal. The method can be applied to an intelligent traffic big data system and an automatic driving system, and can be used for obtaining the specific position of the predicted accident occurrence point when the road section to be predicted is predicted to be the road section with the possibility of occurrence of the road accident, so that the instantaneity of obtaining the predicted accident occurrence point is improved.
Referring to fig. 11, fig. 11 is a schematic flow chart of a traffic accident prediction method according to the present embodiment, which describes a processing flow at a terminal side, and the method includes:
step S610: and sending an accident prediction request comprising information of the road section to be predicted to the server, and instructing the server to return the predicted accident occurrence point obtained according to the traffic accident prediction method in any of the embodiments.
The manner of sending the accident prediction request of the road section information to be predicted to the server may be that the accident prediction request of the road section information to be predicted is sent to the server based on the operation of the user on the terminal.
The user may operate the terminal in various manners, such as voice control operation, text input operation, touch control operation, and the like.
As one way, the terminal may recognize the voice information of the user to obtain a start position and an end position in the voice information, and the start position or the end position may be used as the information of the road section to be predicted. Or obtaining a congested road section between the initial position and the terminal position according to the map big data, taking the congested road section as a road section to be predicted, and obtaining information of the road section to be predicted.
As another mode, the terminal may obtain a start position and an end position based on a text input operation of the user, and use the start position or the end position as the information of the road section to be predicted. Or obtaining a congested road section between the initial position and the terminal position according to the map big data, taking the congested road section as a road section to be predicted, and obtaining information of the road section to be predicted.
As another way, a congested road segment may be selected as a road segment to be predicted based on a touch operation of a user in a map application, and information of the road segment to be predicted may be obtained.
When receiving the accident prediction request, the server executes the method steps in the traffic accident prediction method in the above embodiment, and the specific method steps may refer to the specific description in the traffic accident prediction method in the above embodiment, which is not described in detail in this embodiment.
Step S620: and receiving the predicted accident occurrence point of the road section to be predicted returned by the server.
Step S630: and displaying the information of the road section to be predicted and the predicted accident occurrence point.
The road section information to be predicted and the predicted accident occurrence point can be displayed in a display frame mode in the map application, or can be displayed in an information notification column of the terminal.
As one way, the way of displaying the information of the road section to be predicted and the information of the predicted accident occurrence point may be to display the suspected accident or the predicted accident in a form of a title in the map application, and display the information of the road section to be predicted and the predicted accident occurrence point in a form of a subheading, so as to explain to the user that the accident occurrence point is predicted and not actually reported by the user, and specifically, refer to fig. 10 and the related description of the foregoing embodiment to fig. 10.
In order to realize the authentication of the predicted accident occurrence point so that the subsequent user can detour according to the authenticated predicted accident occurrence point or drive according to the original planned path, the method further comprises the following steps:
step S640: and determining the distance between the terminal and the predicted accident occurrence point according to the positioning information of the terminal and the predicted accident occurrence point.
It will be appreciated that the distance between the terminal and the predicted accident point is the length of the travel path between the location of the terminal and the predicted accident point.
Step S650: and when the distance is smaller than the distance threshold value, displaying prompt information including the road name of the road section to be predicted and the predicted accident occurrence point and a confirmation control.
When the distance between the terminal and the predicted accident occurrence point is smaller than the distance threshold, the prompt information and the confirmation control are displayed, so that the user operating the confirmation control is ensured to be the user passing through the predicted accident occurrence point, and the reliability of the confirmation information fed back based on the confirmation control is improved.
The number of the confirmation controls is two, and the confirmation controls are respectively used for confirming that the predicted accident occurrence point is accurate in prediction and confirming that the predicted accident occurrence point is abnormal in prediction. The control can be displayed in a touch button mode or a frame check mode.
Step S660: and responding to the operation of the confirmation control, and sending the corresponding confirmation information to the server.
As shown in fig. 12, as one mode, the control may be displayed in a manner of a touch button, for example, the terminal may display a first window and a second window on the navigation user interface, where the first window displays first confirmation information, the second window displays second confirmation information, the first confirmation information or the second confirmation information is obtained in response to a touch operation on the first window or the second window, and the first confirmation information or the second confirmation information is sent to the server, where the first confirmation information is used to represent that the predicted accident occurrence point is accurate, and the second confirmation information is used to represent that the predicted accident occurrence point is inaccurate.
As shown in fig. 13, as another mode, the control may be displayed in a check box manner, for example, the terminal may display a first check box and a second check box on the navigation user interface, where the first check box corresponds to first confirmation information, the second check box corresponds to second confirmation information, the first confirmation information or the second confirmation information is obtained in response to a touch operation on the first check box or the second check box, and the first confirmation information or the second confirmation information is sent to the server, where the first confirmation information is used to represent that the predicted accident occurrence point is accurate, and the second confirmation information is used to represent that the predicted accident occurrence point is inaccurate.
And feeding back the confirmation information to the server so that the server authenticates the predicted accident occurrence point based on the confirmation information fed back by the prompt information at the receiving terminal. In the authentication process, the server can delete the predicted accident occurrence point corresponding to the target road section when the confirmation information comprises the accident prediction abnormal information, so that the condition that a user bypasses due to inaccurate prediction results of the road section to be predicted when other terminals obtain the corresponding road section to be predicted subsequently is avoided. The server can also add an authentication result to the predicted accident occurrence point according to the confirmation information, and feed back prompt information comprising an authentication identifier, the predicted accident occurrence point of the target road section and the road name of the target road section when subsequently receiving target road section information comprising the road section to be predicted, which is sent by other terminals, so that the terminal displays the prompt information shown in fig. 11, and a user can conveniently avoid a driving path in time to avoid congestion when the predicted accident occurrence point is accurate.
In the traffic accident prediction method provided by this embodiment, an accident prediction request including information of a road segment to be predicted is sent to a server, the server is instructed to obtain traffic data and target feature data of the road segment to be predicted according to the information of the road segment to be predicted, the target feature data is traffic data corresponding to a type of the traffic data affecting a probability of occurrence of a traffic accident, so as to obtain a prediction probability of occurrence of a traffic accident on the road segment to be predicted according to the target feature data, and when the prediction probability is greater than a probability threshold, a predicted accident occurrence point of the road segment to be predicted is obtained and fed back according to a plurality of different positions of the road segment to be predicted included in the traffic data and a driving speed corresponding to each position, so that a terminal receives and displays the predicted accident occurrence point fed back by the server, and a user can avoid congestion in time according to the predicted accident occurrence point, therefore, traffic jam of the road section to be predicted is relieved, and rescue workers can quickly arrive at the predicted accident occurrence point for rescue, so that rescue efficiency is improved, and casualties and property loss are reduced.
Referring to fig. 14, fig. 14 is a schematic flow chart of a traffic accident prediction method according to an embodiment of the present application, where the embodiment describes a processing flow for jointly implementing traffic accident prediction on a road section to be predicted on a server side and a terminal side, and the method includes:
step S710: and the terminal sends an accident prediction request comprising information of the road section to be predicted to the server.
Step S720: and when receiving the accident prediction request, the server acquires traffic data and target characteristic data corresponding to the road section to be predicted according to the information of the road section to be predicted.
The target characteristic data is traffic data corresponding to the type of the traffic data influencing the occurrence probability of the traffic accident.
Step S730: the server acquires the prediction probability of the traffic accident of the road section to be predicted according to the target characteristic data, and when the prediction probability is larger than a probability threshold value, the server acquires the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted and the corresponding running speed of each position included in the traffic data.
Step S740: the server sends prompt information including the predicted accident occurrence point to the terminal.
Step S750: and the terminal displays the prompt message when receiving the prompt message.
Step S760: the terminal acquires the positioning information and the predicted accident occurrence point according to the terminal, and displays the road name of the road section to be predicted, the prompt information of the predicted accident occurrence point and the confirmation control when the distance between the terminal and the predicted accident occurrence point is determined to be smaller than the distance threshold.
Step S770: and the terminal responds to the operation of the confirmation control and sends the corresponding confirmation information to the server.
Step S780: and when receiving the confirmation information, the server authenticates the predicted accident occurrence point according to the confirmation information.
According to the traffic accident prediction method provided by the embodiment, the traffic data and the target characteristic data of the road section to be predicted are obtained, the prediction probability of the traffic accident of the road section to be predicted is obtained according to the target characteristic data, and when the prediction probability is larger than a probability threshold value, the predicted accident occurrence point of the road section to be predicted is obtained according to a plurality of different positions of the road section to be predicted, which are included in the traffic data, and the corresponding running speed of each position. The method and the device realize that the specific position of the predicted accident occurrence point is obtained when the road section to be predicted is predicted to be the road section with the possibility of road accidents, thereby improving the instantaneity of obtaining the predicted accident occurrence point. In addition, the terminal displays information including the predicted accident occurrence point and the road section to be predicted, so that a user can plan a path in time according to the predicted accident occurrence point and avoid congestion, and meanwhile, traffic congestion of the road section to be predicted can be relieved, and rescuers can rapidly arrive at the predicted accident occurrence point to rescue, so that rescue efficiency is improved, and casualties and property loss are reduced.
Referring to fig. 15, fig. 15 is a block diagram of a traffic accident prediction apparatus 700 according to another embodiment of the present application, which describes modules included in a server side, where the apparatus 700 includes:
the data obtaining module 720 is configured to obtain traffic data of a road segment to be predicted and target feature data, where the target feature data is traffic data corresponding to a type of the traffic data that affects occurrence probability of a traffic accident.
Wherein the traffic data type includes at least one of a road grade type of the road section to be predicted, a length type of the road section to be predicted, and a driving speed type at different positions in the road section to be predicted
And the probability prediction module 740 is configured to obtain the prediction probability of the traffic accident occurring on the road section to be predicted according to the target feature data.
As one mode, the probability prediction module 760 is specifically configured to input the target feature data into a preset model to obtain a predicted probability of a traffic accident occurring on a road segment to be predicted, where the preset model is established based on sample traffic data.
And the accident occurrence point prediction module 760 is used for obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted and the corresponding running speed of each position, wherein the different positions are included in the traffic data, when the prediction probability is greater than the probability threshold value.
As one mode, the accident occurrence point prediction module 760 is specifically configured to obtain speed change rates of any two adjacent positions according to respective corresponding driving speeds of each position, obtain a target speed change rate according to the speed change rates of any two adjacent positions, and obtain a predicted accident occurrence point of a road section to be predicted according to a road type, a road grade, and a length of the road section to be predicted when the target speed change rate is less than or equal to a change rate threshold.
When the target speed change rate is less than or equal to the change rate threshold, the occurrence point prediction module may specifically be: obtaining a path length L1 between a predicted accident occurrence point and a starting point of a road section to be predicted through L1/L0, namely a1 × Rt + a2 × Rg + a3 × L0, wherein L0 is the length of the road section to be predicted, a1, a2 and a3 are constant coefficients respectively, Rt is a road type parameter of the road section to be predicted, and Rg is a road grade parameter of the road section to be predicted; according to the path length L1, the predicted accident occurrence point corresponding to the path length is searched from the correspondence table in which a plurality of positions and the path length from each position to the start point of the link to be predicted are stored.
When the target speed change rate is greater than the change rate threshold, the accident occurrence point prediction module 760 is specifically configured to obtain a speed fitting curve of the road section to be predicted according to the respective corresponding driving speed of each position in the road section to be predicted; acquiring the position of the minimum speed value in the speed fitting curve corresponding to the road section to be predicted as a target position, acquiring the path length from the target position to the initial point of the road section to be predicted, and calculating the ratio of the path length to the length of the road section to be predicted; and obtaining the slope of a tangent line tangent to the speed fitting curve according to the ratio, and taking the position of a tangent point of the tangent line tangent to the fitting curve, which corresponds to the road section to be predicted, as a predicted accident occurrence point.
An accident occurrence point prediction module 760, specifically configured to take the ratio as a slope of a tangent line tangent to the speed-fitting curve when the ratio is less than or equal to a first ratio threshold; and when the ratio is larger than the first ratio threshold, averaging the ratio threshold and the ratio, and taking the average as the slope of a tangent line tangent to the fitting curve.
As a mode, when the acquired traffic data of the road segment to be predicted includes traffic data corresponding to a plurality of sub-road segments, the probability prediction module 740 is specifically configured to obtain a prediction probability of a traffic accident occurring in each sub-road segment based on the target feature data of each sub-road segment; and correspondingly, the accident occurrence point prediction module 760 obtains the predicted accident occurrence point of the road section to be predicted according to the traffic data of the road section to be predicted when the prediction probability is greater than the probability threshold value.
By one approach, the apparatus 700 further comprises:
the system comprises a sample acquisition module 710 for acquiring sample traffic data, wherein the sample traffic data comprises positive sample traffic data and negative sample traffic data, the time of occurrence of a traffic accident exists within a first preset time range from the data acquisition time of the positive sample traffic data, and the time of occurrence of a traffic accident does not exist within a second preset time range from the data acquisition time of the negative sample traffic data; and analyzing and comparing the positive sample traffic data and the negative sample traffic data to obtain the types of the traffic data which influence the occurrence probability of the traffic accident.
In this way, the sample obtaining module 710 is specifically configured to select, from the multiple paths, a sample path in which the traffic flow within the third preset duration range is greater than the traffic flow threshold and the road grade belongs to the preset grade; and acquiring sample traffic data from historical traffic data corresponding to the sample path.
The sample acquisition module 710 is specifically configured to perform deviation rate calculation on the same type of data in the positive sample traffic data and the negative sample traffic data to obtain a deviation rate of each type of data; and taking the data type corresponding to the data with the deviation rate larger than the second ratio threshold value as the type of the traffic data influencing the traffic accident occurrence probability.
By one approach, the apparatus 700 further comprises:
the information generating module 770 is configured to receive the target road segment information sent by the terminal, generate a prompt message including the predicted accident occurrence point of the target road segment and the road name of the target road segment if the target road segment corresponding to the target road segment information belongs to the road segment to be predicted and the target road segment corresponds to the predicted accident occurrence point, and send the prompt message to the terminal.
And the data deleting module 780 is configured to delete the predicted accident occurrence point corresponding to the target road segment when the confirmation information fed back by the receiving terminal based on the prompt information includes the accident prediction abnormal information.
The traffic accident prediction device provided by the embodiment comprises a data acquisition module, a probability prediction module and an accident occurrence point prediction module, wherein the data acquisition module is used for acquiring traffic data and target characteristic data of a road section to be predicted, the probability prediction module is used for acquiring the prediction probability of a traffic accident occurring on the road section to be predicted according to the target characteristic data, and the accident occurrence point prediction module is used for acquiring the prediction accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted and the corresponding running speed of each position when the prediction probability is greater than a probability threshold value. The method and the device realize that the specific position of the predicted accident occurrence point is obtained when the road section to be predicted is predicted to be the road section with the possibility of road accidents, thereby improving the instantaneity of obtaining the predicted accident occurrence point.
Referring to fig. 16, fig. 16 is a block diagram of a traffic accident prediction apparatus 800 according to another embodiment of the present application, where the apparatus 800 includes:
a sending module 810, configured to send an accident prediction request including information of a road segment to be predicted to the server, and instruct the server to return the predicted accident occurrence point obtained according to the traffic accident prediction apparatus in the foregoing embodiment.
And the receiving module 820 is used for receiving the predicted accident occurrence point of the road section to be predicted, which is returned by the server.
And the first display module 830 is configured to display information of the road section to be predicted and a predicted accident occurrence point.
As one mode, the traffic accident prediction apparatus 800 further includes:
the distance obtaining module 840 is configured to determine a distance between the terminal and the predicted accident occurrence point according to the positioning information of the terminal and the predicted accident occurrence point.
And a second display module 850, configured to display, when the distance is smaller than the distance threshold, a prompt message including a road name of the road segment to be predicted and a predicted accident occurrence point, and a confirmation control.
And the information sending module 860 is used for responding to the operation of the confirmation control and sending the corresponding confirmation information to the server.
The traffic accident prediction apparatus 800 according to this embodiment includes a sending module, a receiving module, and a first display module, where the sending module is configured to send an accident prediction request including information of a road segment to be predicted to a server, instruct a module in the server to execute a corresponding function, and return a predicted accident occurrence point, the receiving module is configured to receive the predicted accident occurrence point, and the first display module is configured to display the information of the road segment to be predicted and the predicted accident occurrence point. The method and the device realize that the specific position of the predicted accident occurrence point is obtained when the road section to be predicted is predicted to be the road section with the possibility of road accidents, thereby improving the instantaneity of obtaining the predicted accident occurrence point. Therefore, the user can timely avoid congestion according to the predicted accident occurrence point, the traffic congestion of the road section to be predicted is relieved, the rescue workers can quickly arrive at the predicted accident occurrence point for rescue, and the rescue efficiency is improved to reduce casualties and property loss.
It should be noted that the device embodiment and the method embodiment in the present application correspond to each other, and specific principles in the device embodiment may refer to the contents in the method embodiment, which is not described herein again.
An electronic device provided in an embodiment of the present application will be described below with reference to fig. 17.
Referring to fig. 17, based on the traffic accident prediction method, another electronic device 100 including a processor 102 that can execute the traffic accident prediction method is provided in an embodiment of the present application, where the electronic device 100 may be a server, a smart phone, a tablet computer, a vehicle-mounted terminal, or a portable computer. The electronic device 100 also includes a memory 104. The memory 104 stores programs that can execute the content of the foregoing embodiments, and the processor 102 can execute the programs stored in the memory 104.
Processor 102 may include, among other things, one or more cores for processing data and a message matrix unit. The processor 102 interfaces with various components throughout the electronic device 100 using various interfaces and circuitry to perform various functions of the electronic device 100 and process data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 104 and invoking data stored in the memory 104. Alternatively, the processor 102 may be implemented in hardware using at least one of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The processor 102 may integrate one or more of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing display content; the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 102, but may be implemented by a communication chip.
The Memory 104 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). The memory 104 may be used to store instructions, programs, code sets, or instruction sets.
The electronic device 100 may also include a network module and a screen.
The network module is used for receiving and sending electromagnetic waves, and realizing the interconversion of the electromagnetic waves and the electric signals, so as to communicate with a communication network or other equipment, for example, audio playing equipment. The network module may include various existing circuit elements for performing these functions, such as an antenna, a radio frequency transceiver, a digital signal processor, an encryption/decryption chip, a Subscriber Identity Module (SIM) card, memory, and so forth. The network module may communicate with various networks such as the internet, an intranet, a wireless network, or with other devices via a wireless network. The wireless network may comprise a cellular telephone network, a wireless local area network, or a metropolitan area network. For example, the network module may exchange information with the base station.
The screen may display content and may also be used to respond to touch operations.
It should be noted that the electronic device 100 may also protect more devices in order to realize more functions.
Referring to fig. 18, a block diagram of a computer-readable storage medium according to an embodiment of the present application is shown. The computer readable medium 900 has stored therein a program code that can be called by a processor to execute the method described in the above method embodiments.
The computer-readable storage medium 900 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. Optionally, computer-readable storage medium 910 includes non-volatile computer-readable storage medium. The computer readable storage medium 900 has storage space for program code 910 to perform any of the method steps of the method described above. The program code can be read from or written to one or more computer program products. The program code 910 may be compressed, for example, in a suitable form.
Embodiments of the present application also provide a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the method described in the various alternative implementations described above.
In summary, according to the traffic accident prediction method, the traffic accident prediction device, the electronic device, and the storage medium provided by the embodiments of the present application, the traffic data and the target feature data of the road segment to be predicted are obtained, the prediction probability of the traffic accident occurring on the road segment to be predicted is obtained according to the target feature data, and when the prediction probability is greater than the probability threshold, the predicted accident occurring point of the road segment to be predicted is obtained according to a plurality of different positions of the road segment to be predicted included in the traffic data and the respective corresponding driving speeds of each of the positions. The method and the device realize that the specific position of the predicted accident occurrence point is obtained when the road section to be predicted is predicted to be the road section with the possibility of road accidents, thereby improving the instantaneity of obtaining the predicted accident occurrence point. Therefore, the user can timely avoid congestion according to the predicted accident occurrence point, the traffic congestion of the road section to be predicted is relieved, the rescue workers can quickly arrive at the predicted accident occurrence point for rescue, and the rescue efficiency is improved to reduce casualties and property loss.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims (13)

1. A traffic accident prediction method, comprising:
acquiring traffic data and target characteristic data of a road section to be predicted, wherein the target characteristic data is traffic data corresponding to the type of the traffic data influencing the occurrence probability of a traffic accident, and the type of the traffic data comprises at least one of road grade type of the road section to be predicted, length type of the road section to be predicted and driving speed type of different positions in the road section to be predicted;
respectively carrying out normalization processing on each target characteristic data, and carrying out regression calculation on each processed target characteristic data and a partial regression coefficient corresponding to each target characteristic data to obtain the prediction probability of the traffic accident of the road section to be predicted;
and when the prediction probability is greater than a probability threshold value, obtaining a predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted included in the traffic data and the corresponding running speed of each position.
2. The method of claim 1, further comprising:
obtaining sample traffic data, wherein the sample traffic data comprises positive sample traffic data and negative sample traffic data, each sample traffic data corresponds to a data obtaining moment, the positive sample traffic data is sample traffic data with traffic accidents in a first time length range from the data obtaining moment of the positive sample traffic data, and the negative sample traffic data is sample traffic data without traffic accidents in a second time length range from the data obtaining moment of the negative sample traffic data;
and analyzing and comparing the positive sample traffic data and the negative sample traffic data to obtain the type of the traffic data influencing the traffic accident occurrence probability.
3. The method of claim 2, wherein the obtaining sample traffic data comprises:
selecting a sample path from the plurality of paths, wherein the traffic flow within a third preset time range is greater than a traffic flow threshold value, and the road grade belongs to a preset grade;
and acquiring sample traffic data from historical traffic data corresponding to the sample path.
4. The method according to claim 1, wherein the obtaining the predicted accident occurrence point of the road section to be predicted according to the plurality of different positions and the respective corresponding running speeds of each position comprises:
obtaining the speed change rate of any two adjacent positions according to the corresponding running speed of each position, and obtaining a target speed change rate according to the speed change rates of any two adjacent positions;
and when the target speed change rate is smaller than or equal to a change rate threshold value, acquiring a predicted accident occurrence point of the road section to be predicted according to the road type, the road grade and the length of the road section to be predicted.
5. The method according to claim 4, wherein the step of obtaining the predicted accident occurrence point of the road section to be predicted according to the plurality of different positions and the corresponding running speed of each position further comprises the following steps:
when the target speed change rate is larger than the change rate threshold value, obtaining a speed fitting curve of the road section to be predicted according to the running speed corresponding to each position in the road section to be predicted;
acquiring the position of the minimum speed value in the speed fitting curve corresponding to the road section to be predicted as a target position, acquiring the path length from the target position to the initial point of the road section to be predicted, and calculating the ratio of the path length to the length of the road section to be predicted;
and obtaining the slope of a tangent line tangent to the speed fitting curve according to the ratio, and taking the position of a tangent point of the tangent line tangent to the fitting curve, which corresponds to the road section to be predicted, as the predicted accident occurrence point.
6. The method of claim 5, wherein deriving a slope of a tangent to the fitted curve from the ratio comprises:
when the ratio is less than or equal to a first ratio threshold, taking the ratio as the slope of a tangent line tangent to the speed fitting curve;
and when the ratio is larger than a first ratio threshold, averaging the ratio threshold and the ratio, and taking the average as the slope of a tangent line tangent to the fitting curve.
7. The method according to any one of claims 1 to 6, wherein when the acquired traffic data of the road segment to be predicted includes traffic data corresponding to a plurality of sub-road segments respectively, the acquiring the predicted probability of the traffic accident occurring on the road segment to be predicted according to the target feature data comprises:
acquiring the predicted probability of the traffic accident of each sub-road section according to the target characteristic data of each sub-road section;
and taking the sub-road section corresponding to the maximum prediction probability in the prediction probabilities respectively corresponding to the sub-road sections as a new road section to be predicted, and taking the maximum prediction probability as the prediction probability of the traffic accident of the road section to be predicted.
8. The method of claim 1, further comprising:
receiving target road section information sent by a terminal;
and if the target road section corresponding to the target road section information belongs to the road section to be predicted and the target road section corresponds to the predicted accident occurrence point, generating prompt information comprising the predicted accident occurrence point of the target road section and the road name of the target road section, and sending the prompt information to the terminal.
9. The method of claim 8, further comprising:
and receiving confirmation information fed back by the terminal based on the prompt information, and deleting the predicted accident occurrence point corresponding to the target road section when the confirmation information comprises accident prediction abnormal information.
10. A traffic accident prediction method, comprising:
sending an accident prediction request including information of a road section to be predicted to a server, and instructing the server to return a predicted accident occurrence point obtained according to the method of any one of claims 1 to 9;
receiving a predicted accident occurrence point of the road section to be predicted, which is returned by the server;
and displaying the information of the road section to be predicted and the predicted accident occurrence point.
11. A traffic accident prediction apparatus, comprising:
the data acquisition module is used for acquiring traffic data and target characteristic data of a road section to be predicted, wherein the target characteristic data is traffic data corresponding to the type of the traffic data influencing the occurrence probability of a traffic accident, and the type of the traffic data comprises at least one of the road grade type of the road section to be predicted, the length type of the road section to be predicted and the driving speed type of different positions in the road section to be predicted;
the probability prediction module is used for respectively carrying out normalization processing on each target characteristic data, carrying out regression calculation on each processed target characteristic data and a partial regression coefficient corresponding to each target characteristic data, and obtaining the prediction probability of the traffic accident on the road section to be predicted;
and the accident occurrence point prediction module is used for obtaining the predicted accident occurrence point of the road section to be predicted according to a plurality of different positions of the road section to be predicted and the corresponding running speed of each position included in the traffic data when the prediction probability is greater than the probability threshold.
12. An electronic device, comprising:
one or more processors;
a memory;
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the method of any of claims 1-9 or claim 10.
13. A computer-readable storage medium, characterized in that the computer-readable storage medium stores program code that can be invoked by a processor to perform the method according to any one of claims 1 to 9 or claim 10.
CN202011324403.3A 2020-11-23 2020-11-23 Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium Active CN112530163B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011324403.3A CN112530163B (en) 2020-11-23 2020-11-23 Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011324403.3A CN112530163B (en) 2020-11-23 2020-11-23 Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium

Publications (2)

Publication Number Publication Date
CN112530163A CN112530163A (en) 2021-03-19
CN112530163B true CN112530163B (en) 2022-03-01

Family

ID=74992931

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011324403.3A Active CN112530163B (en) 2020-11-23 2020-11-23 Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium

Country Status (1)

Country Link
CN (1) CN112530163B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022226689A1 (en) * 2021-04-25 2022-11-03 华为技术有限公司 Data management method and apparatus, and terminal device
CN113256969B (en) * 2021-04-30 2022-08-16 山东金宇信息科技集团有限公司 Traffic accident early warning method, device and medium for expressway
CN113780641A (en) * 2021-08-31 2021-12-10 同济大学 Accident prediction method and device based on transfer learning
CN116189134A (en) * 2023-04-26 2023-05-30 宜宾闪马智通科技有限公司 Region identification method and device based on image identification and radar
CN116663934A (en) * 2023-06-14 2023-08-29 中瑞科技术有限公司 Traffic event processing method and system integrating monitoring data

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19806687A1 (en) * 1998-02-18 1999-08-26 Daimler Chrysler Ag Method of preventing collision of vehicle with obstruction
JP2004070588A (en) * 2002-08-05 2004-03-04 Sumitomo Electric Ind Ltd Occurrence point specifying device for unforeseen event
CN105243840A (en) * 2015-09-30 2016-01-13 重庆云途交通科技有限公司 V2I-based adaptive accident identification method and system
CN105279572A (en) * 2015-09-16 2016-01-27 北京城建设计发展集团股份有限公司 City track traffic passenger flow density index calculating and releasing system
CN105469603A (en) * 2015-12-30 2016-04-06 青岛海信网络科技股份有限公司 Traffic congestion source analysis method and traffic congestion source analysis device
CN105701579A (en) * 2016-03-08 2016-06-22 北京工业大学 Prediction method for predicting traffic accidents on basic section of dual-lane secondary road in plateau area
CN106023585A (en) * 2015-03-25 2016-10-12 丰田自动车株式会社 Congestion information generation device and congestion information generation method
CN106971537A (en) * 2017-04-20 2017-07-21 山东高速信息工程有限公司 For the congestion in road Forecasting Methodology and system of accident
CN107045794A (en) * 2017-01-16 2017-08-15 百度在线网络技术（北京）有限公司 Road conditions processing method and processing device
CN110264711A (en) * 2019-05-29 2019-09-20 北京世纪高通科技有限公司 A kind of traffic accident method of determining probability and device
CN110264735A (en) * 2019-06-28 2019-09-20 佛山科学技术学院 A kind of traffic congestion forecasting system, method and storage medium based on big data
KR102033858B1 (en) * 2018-07-10 2019-10-17 주식회사 퀀텀게이트 Prediction system for traffic accident
CN111724590A (en) * 2020-06-03 2020-09-29 重庆大学 Highway abnormal event occurrence time estimation method based on travel time correction

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH11283159A (en) * 1998-03-30 1999-10-15 Hidekatsu Hiroya Present vehicle information transmitting and traveling system
JP4175312B2 (en) * 2004-09-17 2008-11-05 株式会社日立製作所 Traffic information prediction device
US9569960B2 (en) * 2015-02-24 2017-02-14 Here Global B.V. Method and apparatus for providing traffic jam detection and prediction
JP2017084268A (en) * 2015-10-30 2017-05-18 株式会社東芝 Accident occurrence forecast system and accident occurrence forecast method
CN110751311B (en) * 2019-09-05 2022-09-13 北京交通大学 Data extraction and real-time prediction method for sporadic traffic jam duration

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19806687A1 (en) * 1998-02-18 1999-08-26 Daimler Chrysler Ag Method of preventing collision of vehicle with obstruction
JP2004070588A (en) * 2002-08-05 2004-03-04 Sumitomo Electric Ind Ltd Occurrence point specifying device for unforeseen event
CN106023585A (en) * 2015-03-25 2016-10-12 丰田自动车株式会社 Congestion information generation device and congestion information generation method
CN105279572A (en) * 2015-09-16 2016-01-27 北京城建设计发展集团股份有限公司 City track traffic passenger flow density index calculating and releasing system
CN105243840A (en) * 2015-09-30 2016-01-13 重庆云途交通科技有限公司 V2I-based adaptive accident identification method and system
CN105469603A (en) * 2015-12-30 2016-04-06 青岛海信网络科技股份有限公司 Traffic congestion source analysis method and traffic congestion source analysis device
CN105701579A (en) * 2016-03-08 2016-06-22 北京工业大学 Prediction method for predicting traffic accidents on basic section of dual-lane secondary road in plateau area
CN107045794A (en) * 2017-01-16 2017-08-15 百度在线网络技术（北京）有限公司 Road conditions processing method and processing device
CN106971537A (en) * 2017-04-20 2017-07-21 山东高速信息工程有限公司 For the congestion in road Forecasting Methodology and system of accident
KR102033858B1 (en) * 2018-07-10 2019-10-17 주식회사 퀀텀게이트 Prediction system for traffic accident
CN110264711A (en) * 2019-05-29 2019-09-20 北京世纪高通科技有限公司 A kind of traffic accident method of determining probability and device
CN110264735A (en) * 2019-06-28 2019-09-20 佛山科学技术学院 A kind of traffic congestion forecasting system, method and storage medium based on big data
CN111724590A (en) * 2020-06-03 2020-09-29 重庆大学 Highway abnormal event occurrence time estimation method based on travel time correction

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于经验贝叶斯法的道路事故预测分析研究;李政等;《交通信息与安全》;20110820(第04期);全文 *
基于非线性回归和BP神经网络的交通事故时空影响预测模型;朱博雅等;《公路工程》;20181213(第06期);全文 *

Also Published As

Publication number Publication date
CN112530163A (en) 2021-03-19

Similar Documents

Publication Publication Date Title
CN112530163B (en) Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium
US11891071B2 (en) Vehicle to vehicle telematics
US11397993B2 (en) Electronic logging and track identification system for mobile telematics devices, and corresponding method thereof
CN109801491B (en) Intelligent navigation method, device and equipment based on risk assessment and storage medium
CN109076093B (en) System and method for controlling sensor-based data acquisition and signal processing in a vehicle
US12008896B2 (en) System and method for detection and quantification of irregular traffic congestion
CN112863172B (en) Highway traffic running state judgment method, early warning method, device and terminal
KR102352666B1 (en) System and Method for Predicting Traffic Accident Risk
US20180276988A1 (en) Automated traffic data validation
US20190186929A1 (en) Route searching device, route searching system, and computer program
EP4191443A1 (en) Vehicle data processing method and apparatus, computer device and storage medium
CN113155173B (en) Perception performance evaluation method and device, electronic device and storage medium
CN112434260B (en) Road traffic state detection method, device, storage medium and terminal
CN111144446A (en) Driver identity recognition method and system based on space-time grid
CN112434075A (en) ETC portal frame based traffic anomaly detection method and device, storage medium and terminal
US20140031061A1 (en) Systems And Methods For Monitoring Device And Vehicle
CN114495305A (en) Method and system for road differential charging, road side base station and storage medium
US11449832B2 (en) Systems and methods for obtaining data annotations
Peng et al. Usage-based insurance system based on carrier-cloud-client
CN112781607A (en) Navigation apparatus and method of providing personalized map service thereof
US11856061B2 (en) Using contextual information for vehicle trip loss risk assessment scoring
US20150348407A1 (en) Recommendation Engine Based on a Representation of the Local Environment Augmented by Citizen Sensor Reports
CN114627643A (en) Expressway accident risk prediction method, device, equipment and medium
TW202216510A (en) Method, electronic device, and system for predicting future overspeeding
CN117894181B (en) Global traffic abnormal condition integrated monitoring method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 40041019

Country of ref document: HK

GR01 Patent grant
GR01 Patent grant