Method, equipment, medium and device for identifying accident-prone areas

Abstract

The present invention discloses a method, equipment, medium and device for identifying accident-prone areas, comprising the following steps: a data acquisition and processing step for acquiring travel data and vehicle historical accident alarm data transmitted by an on-board terminal; collecting vehicle-related data, including travel data, vehicle historical accident alarm data, real-time accident alarm data and environmental data, and pre-processing these data; reflecting the vehicle's speed before the historical accident alarm according to the travel data; extracting the accident location according to the real-time accident alarm number, and associating it with the environmental conditions at that time to obtain accident-prone areas under different environmental conditions; setting an update cycle based on the same steps to add new accident-prone areas; associating the local environment of the day according to the travel data acquired in real time, and when the vehicle contacts the corresponding accident-prone area under the corresponding environmental level, being able to dynamically monitor the vehicle's driving condition, and issuing safety reminders to vehicles entering the accident-prone area, thereby reducing driving risks.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle traffic safety, and in particular to a method, equipment, medium and device for identifying accident-prone areas. Background Art

[0002] With the rapid development of the national economy, the number of cars in China has increased annually, reaching 290 million by June 2021. Furthermore, with the advancement of wireless communication and positioning technologies, many of these vehicles are equipped with onboard GPS devices, making cars increasingly intelligent and user-friendly. Smart cockpit services are becoming increasingly diverse and are even becoming a key competitive advantage for auto brands. Among these, smart services related to safety are receiving significant attention from both OEMs and users, as safety is a bottleneck limiting the advancement of smart cars. Therefore, OEMs urgently need to ensure safety while enriching smart cockpit services to gain a competitive advantage. Among these smart safety services, the detection and notification of accident-prone areas play a crucial role in improving driving safety.

[0003] Using connected vehicle technology, OEMs already have access to extensive vehicle driving information. Based on this information, they have developed numerous intelligent services, such as vehicle tracking, anomaly warnings, remote guidance, and location notifications. However, developing accident-prone location detection and notification services based on this data remains a pressing challenge. The difficulty of quickly implementing road safety reminders within OEM equipment hinders the ability to improve after-sales service. To address these issues, existing OEMs typically collaborate with insurance companies or directly rely on accident alert data, taking action only after a traffic accident occurs, without providing pre-accident safety reminders. Summary of the Invention

[0004] The present invention aims to avoid the deficiencies of the above-mentioned prior art and provides a method, equipment, medium and device for identifying accident-prone areas, which can dynamically monitor whether a vehicle passes through an accident-prone area, remind the owner of the vehicle passing by, and reduce the occurrence of accidents for the owner.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] An embodiment of the present invention provides a method for identifying accident-prone areas, comprising the following steps:

[0007] a data acquisition and processing step for acquiring travel data, historical vehicle accident alarm data, real-time accident alarm data, and daily environmental data of cities in various provinces collected by a weather website API transmitted by an on-board terminal, and pre-processing the travel data, historical vehicle accident alarm data, real-time accident alarm data, and daily environmental data of cities in various provinces collected by a weather website API;

[0008] A historical accident-prone area generation step, comprising: constructing a historical accident-prone area generation model by performing GEOHASH coding on the vehicle historical accident alarm data;

[0009] An accident-prone area updating step, constructing an accident-prone area updating model by performing GEOHASH encoding on the real-time accident alarm data;

[0010] The accident area monitoring step is to determine the accident-prone area by performing GEOHASH matching based on the travel data and environmental data transmitted by the on-board terminal. The monitoring model is used to monitor whether the vehicle enters the accident-prone area during driving. If it does, the driving speed of the vehicle entering the accident-prone area will be reminded in real time.

[0011] In the accident-prone area identification method of the present invention, the step of generating historical accident-prone areas includes:

[0012] S1. Based on the historical accident alarm data, the collision position at the time of the historical accident alarm is obtained. The longitude and latitude of the collision position at the time of the historical accident alarm are obtained by removing duplicates to obtain a collision point set L. m,k ={(lng 1,1 ,lat 1,1 ,time 1,1 ),(lng 1,2 ,lat 1,2 ,time 1,2 )…(lng m,k ,lat m,k ,time m,k )}, where m is the mth vehicle that has collided, k is the kth collision of the mth collision vehicle, lng and lat are the longitude and latitude of the collision point respectively, and time m,k is the time when the kth collision of the mth collision vehicle occurs; the set of pre-collision speeds S is obtained according to the collision point m,k =(s 1,1 ,s 1,2 …s m,k ), where s m,k , is the speed of the m-th collision vehicle before the k-th collision; where; the temperature t corresponding to the k-th collision point of the m-th vehicle is calculated by m,k Find the mean Thus, the temperature before the collision is obtained; the mean Let x be the temperature, and divide it into different levels according to the rules. The corresponding formula is: T1, T2…T N Respectively represent the threshold value of division; based on regularized matching, the climate and wind are combined to divide the levels, where x represents the climate and wind characteristics corresponding to this collision point, and the corresponding formula is: c rei Indicates the relevant climate conditions at level i, where i = 1, 2…M1, and the priority of the rules is from M1 to 1; w rei Indicates the relevant situation of wind force at level i, where i = 1, 2…M1, and the priority of the rules is also from M1 to 1;

[0013] S2, based on the GEOHASH algorithm, the collision point set L m,k Perform a-bit GEOHASH encoding to obtain the corresponding GEOHASH encoding dataset G m,k [a]=(g 1,1 [a],g 1,2 [a]…g m,k [a]), where G m,k The same value exists in [a];

[0014] S3. Acquire a set of environmental data C when the collision occurs according to the collision point. o,e , where the environmental data calculation formula is:

[0015] {(c 1,1 ,w 1,1 ,t 1,1 ,di 1,1 ,date1),

[0016] C o,e =(c 1,2 ,w 1,2 ,t 1,2 ,di 1,2 ,date2),

[0017] …

[0018] (c o,e ,w o,e ,t o,e ,di o,e ,date o )}

[0019] Where date is the date corresponding to the environmental data, o is the date of the oth day, and each day contains e cities; di represents the address of each province and city area, that is, di o,e is the address of the city area in the province e on the oth day; c represents the climate, c r,j is the specific climate of the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; w represents the wind force, w r,j is the specific wind speed in the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; t represents the temperature of the day, t r,jRepresents the specific temperature of the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; the environmental data set C o,e By comparing the address and time variables (d, date) with the collision point set L obtained in step S1 m,k The time and address (D, time) are spliced together, and the environment data corresponding to the collision is GEOHASH coded by the GEOHASH algorithm to obtain a collection L with environment and GEOHASH coding. m,k ; l m,k ={(lng 1,1 ,lat 1,1 ,g 1,1 [a],time 1,1 ,d 1,1 ,s 1,1 ,c 1,1 ,w 1,1 ,t 1,1 ),

[0020] (lng 1,2 ,lat 1,2 ,g 1,2 [a],time 1,2 ,d 1,2 ,s 1,2 ,c 1,2 ,w 1,2 ,t 1,2 )

[0021] …

[0022] (lng m,k ,lat m,k ,g m,k [a],time m,k ,d m,k ,s m,k ,c m,k ,w m,k ,t m,k )}

[0023] Obtain the road network data based on OSM to obtain the province, city, and region address D of the collision point. m,k =(d 1,1 ,d 1,2 …d m,k ), matching the obtained b-digit GEOHASH code of the longitude and latitude of the collision point with the b-digit GEOHASH code of the longitude and latitude before the collision point, if they are the same, the collision point is considered to be the address D on the road network data;

[0024] S4. By setting a threshold value Q of the number of accidents, it is determined whether the area is a historical accident-prone area when the number of accidents in different environmental combinations is q. When q ≥ Q, the area is a historical accident-prone area.

[0025] In the accident-prone area identification method of the present invention, the step of generating historical accident-prone areas further includes:

[0026] When the number of accidents reaches Q, the corresponding a-bit GEOHASH code is G0[a]. Since the collision points may be concentrated at the critical point of the coding area, the a-bit GEOHASH codes of the eight areas around the code are obtained, that is, a 3*3 adjacent range code combination is formed as the accident-prone area, which is expressed as follows: The information stored in the accident-prone areas here includes a bit of GEOHASH code corresponding to the set of all accident-prone areas in, is the mean of the pre-collision speeds of all collision points in the accident-prone area; the set of accident locations that do not reach the accident-prone area threshold is denoted as l v ; where l v =({g1[a],T1,C1,s1),(g2[a],T2,C2,s2)…(g v [a],T v ,C v ,s v )}, corresponding to GEOHASH code, temperature layer, climate layer and speed before the accident, respectively.

[0027] In the accident-prone area identification method of the present invention, the accident-prone area updating step includes:

[0028] S31, by l v Combined with the accident locations generated by real-time accident alarm data to form new accident-prone areas;

[0029] S32: Based on the real-time accident alarm data, obtain the collision position at the time of the real-time accident alarm, take the longitude and latitude of the collision position at the time of the real-time accident alarm, remove the duplicates, and obtain a collision point set L m,k ={(lng 1,1 ,lat 1,1 ,time 1,1 ),(lng 1,2 ,lat 1,2 ,time 1,2 )…(lng m, ,lat m,k ,time m,k)}, where m is the mth vehicle that has collided, k is the kth collision of the mth collision vehicle, lng and lat are the longitude and latitude of the collision point respectively, and time m,k is the time when the kth collision of the mth collision vehicle occurs; the set of pre-collision speeds S is obtained according to the collision point m,k =(s 1,1 ,s 1,2 …s m,k ), where s m,k , is the speed of the m-th collision vehicle before the k-th collision; where; the temperature t corresponding to the k-th collision point of the m-th vehicle is calculated by m,k Find the mean Thus, the temperature before the collision is obtained; the mean Let x be the temperature, and divide it into different levels according to the rules. The corresponding formula is: T1, T2…T N Respectively represent the threshold value of division; based on regularized matching, the climate and wind are combined to divide the levels, where x represents the climate and wind characteristics corresponding to this collision point, and the corresponding formula is: c rei Indicates the relevant climate conditions at level i, where i = 1, 2…M2, and the priority of the rules is from M2 to 1; w rei Indicates the relevant situation of wind force at level i, where i = 1, 2…M2, and the priority of the rules is also from M2 to 1;

[0030] S33, based on the GEOHASH algorithm, the collision point set L m,k Perform a-bit GEOHAS encoding to obtain the corresponding GEOHASH encoding dataset G m,k [a]=(g 1,1 [a],g 1,2 [a]…g m,k [a]), where G m,k The same value exists in [a];

[0031] S34, obtaining an environmental data set C when the collision occurs according to the collision point. o,e , where the environmental data calculation formula is:

[0032] {(c 1,1 ,w 1,1 ,t 1,1 ,di 1,1 ,date1),

[0033] C o,e =(c 1,2 ,w 1,2 ,t1,2 ,di 1,2 ,date1),

[0034] …

[0035] (c o,e ,w o,e ,t o,e ,di o,e ,date o )}

[0036] Where date is the date corresponding to the environmental data, o is the date of the oth day, and each day contains e cities; di represents the province, city, and region address, i.e. di o,e is the address of the city area in the province e on the oth day; c represents the climate, c r,j is the specific climate of the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; w represents the wind force, w r,j represents the specific wind speed in the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; t represents the temperature of the day, t r,j Represents the specific temperature of the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; the environmental data set C o,e By comparing the address and time variables (d, date) with the collision point set L obtained in step S32 m,k The time and address (D, time) are spliced together, and the environment data corresponding to the collision is GEOHASH coded by the GEOHASH algorithm to obtain a collection with environment and GEOHASH coding l f,h ={(lng 1,1 ,lat 1,1 ,g 1,1 [a],time 1,1 ,d 1,1 ,s 1,1 ,T 1,1 ,C 1,1 ),

[0037] (lng 1,2 ,lat 1,2 ,g 1,2 [a],time 1,2 ,d 1,2 ,s 1,2 ,T 1,2 ,C 1,2 )

[0038] …

[0039] (lng f,h ,lat f,h,h f,h [a],time f,h ,d f,h ,s f,h ,T f,h ,C f,h )}

[0040] Obtain road network data based on OSM to obtain the address D of each province and city area of the collision point m,k =(d 1,1 ,d 1,2 …d m,k ), matching the obtained b-digit GEOHASH code of the longitude and latitude of the collision point with the b-digit GEOHASH code of the longitude and latitude before the collision point, if they are the same, the collision point is considered to be the address D on the road network data;

[0041] S35. By setting a threshold value Q of the number of accidents, it is determined whether it is a new accident-prone area when the number of accidents in different environmental combinations is q; when the GEOHASH code of the new accident-prone area is the same as the GEOHASH code of the historical accident-prone area and the environmental classification is consistent, it is classified as an accident-prone area and the collision information is deleted, and only collision points that have not reached the accident-prone area are retained; the retained collision points are set as new accident-prone areas, thereby updating the accident-prone areas.

[0042] In the accident-prone area identification method of the present invention, the accident-prone area monitoring step includes:

[0043] S41, by obtaining the travel data B transmitted by the vehicle terminal M , where B M =(lng1,lat1,time1,s1) M , where M is the number of vehicles with all driving behaviors, lng1, lat1 are the longitude and latitude information collected by the vehicle in real time, time1 is the time when the data is generated, and s1 represents the driving speed of the vehicle; GEOHASH code and corresponding address information are obtained through real-time driving behavior data;

[0044] S42. Obtain the road network data based on OSM to obtain the environmental information of the day. By associating the environmental information with the driving information, obtain the real-time temperature and climate level, and thus obtain the trip data with the environment B M ; That is B M =(lng1,lat1,time1,s1,g1[a],T1,C1) M ;

[0045] S43. When a vehicle approaches or enters an accident-prone area under a corresponding environmental level, a real-time reminder of the accident-prone area is issued to the vehicle, and the vehicle is reminded to control its driving speed.

[0046] In the accident-prone area identification method of the present invention, step S41 further includes:

[0047] S410, obtain a GEOHASH code through the travel data, and associate it with the road network data obtained based on OSM in step S3 to obtain the address D of each province and city area of the collision point m,k =(d 1,1 ,d 1,2 …d m,k ), based on the b-bit GEOHASH code matching, the corresponding provincial city area address information is obtained; that is, based on the acquired road network data and the corresponding b-bit GEOHASH code, the b-bit GEOHASH before the collision point is matched. If they are the same, the collision point is considered to be an address on the road network.

[0048] According to another aspect of the present invention, there is also provided a device for identifying accident-prone areas, the device comprising:

[0049] The data acquisition and processing module is used to obtain the travel data, historical vehicle accident alarm data, real-time accident alarm data transmitted by the vehicle terminal, and the daily environmental data of cities in each province collected by the weather website API, and pre-process the travel data, historical vehicle accident alarm data, real-time accident alarm data, and daily environmental data of cities in each province collected by the weather website API;

[0050] The historical accident-prone areas generation module is used to perform GEOHASH matching on historical vehicle accident alarm data and generate historical accident-prone areas based on whether the number of accidents reaches a threshold;

[0051] The accident-prone area update module combines real-time accident alarm data with accident locations that do not reach the accident-prone area threshold to form a new accident location. By performing GEOHASH matching on the new alarm collision point data obtained from the real-time alarm data, the collision points that do not reach the accident-prone area are retained; the retained collision points are set as new accident-prone areas, thereby updating the accident-prone areas.

[0052] The accident area monitoring module is used to monitor whether the vehicle enters the accident-prone area while driving, issue a reminder to the vehicle entering the accident-prone area, and remind the vehicle to control the driving speed.

[0053] Another aspect of this embodiment further provides a computer device, including:

[0054] one or more processors;

[0055] a storage device for storing one or more programs,

[0056] When the one or more programs are executed by the one or more processors, the one or more processors implement the accident-prone area identification method as described in any embodiment of the present invention.

[0057] In another aspect of this embodiment, a computer readable storage medium is provided on which a computer program is stored.

[0058] When executed, the accident-prone area identification method as described in any embodiment of the present invention is implemented.

[0059] The above-disclosed method, device, medium and apparatus for identifying accident-prone areas have the following beneficial effects: by collecting vehicle-related data, travel data, historical vehicle accident alarm data, real-time accident alarm data and weather website API, daily environmental data of cities in various provinces are collected and pre-processed; the speed of the vehicle before the historical accident alarm is reflected based on the travel data. The accident location is extracted based on the real-time accident alarm number, and the environmental conditions at that time are associated to obtain accident-prone areas under different environmental conditions; based on the same steps, an update cycle is set to add new accident-prone areas; based on the real-time travel data, the local environment of the day is associated. When a vehicle contacts an accident-prone area of its corresponding classification under the corresponding environmental level, the vehicle's driving condition can be dynamically monitored, and safety reminders can be issued to vehicles entering the accident-prone area, thereby reducing driving risks. At the same time, the present invention can also enable the main engine manufacturer to grasp the situation of the owner's vehicle accident location in real time, and can accurately provide information about the vehicle accident location, so that the main engine manufacturer can issue reminders to customers to pay attention to speed and driving behavior when passing through dangerous sections. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flowchart of a method for identifying accident-prone areas provided by one embodiment of the present invention;

[0061] Figure 2 This is a block diagram of the working principle of a device for identifying accident-prone areas provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0063] Figure 1 FIG2 is a flow chart of a method for identifying accident-prone areas according to an embodiment of the present invention.

[0064] like Figure 1 As shown, the accident-prone area identification method provided by the present invention includes the following steps:

[0065] 101. Data acquisition and processing steps are used to obtain travel data transmitted by the vehicle terminal, the vehicle's historical accident alarm data, real-time accident alarm data, and daily environmental data of cities in various provinces collected by the meteorological website API, and pre-process the travel data transmitted by the vehicle terminal, the vehicle's historical accident alarm data, real-time accident alarm data, and daily environmental data of cities in various provinces collected by the meteorological website API; the travel data includes the driving speed before the collision in the vehicle's historical accident alarm data and real-time accident alarm data, as well as the vehicle's real-time operating status, and is used for real-time monitoring and reminders; the environmental data collected is daily environmental data. Trip data collected by vehicle sensors consists of two parts: the first is the time and location of all front-end collisions, rear-end collisions, and airbag deployments, all recorded in historical accident alerts; the second is the time and location of real-time accident alerts. Trip data consists of two parts: one collects the vehicle's speed at the time of the alert, based on both historical and real-time accident alerts; the other is real-time driving data, used for real-time monitoring and alerts; and finally, environmental data, which is collected daily. Historical environmental data is collected directly, while real-time environmental data is collected daily at midnight. Preprocessing primarily targets structured data and includes three steps: 1) Dimensional transformation, converting units such as travel distance, time, and latitude and longitude; 2) Outlier filtering, which primarily filters out two types of data: data that clearly exceeds physical limits (e.g., speeds exceeding 300 km / h) and data with transmission errors; and 3) Missing value filling, which is primarily achieved through mode and mean filling.

[0066] 102. A historical accident-prone area generation step is to construct a historical accident-prone area generation model by performing GEOHASH encoding on the vehicle historical accident alarm data; wherein the vehicle historical accident alarm data is completed once and does not undergo iteration.

[0067] 103. Accident-prone area update step: constructing an accident-prone area update model by performing GEOHASH coding on real-time accident alarm data;

[0068] 104. Accident area monitoring step: determine accident-prone areas by performing GEOHASH matching based on the travel data and environmental data transmitted by the vehicle terminal. Use the monitoring model to monitor whether the vehicle enters the accident-prone area during driving. If so, the vehicle's driving speed in the accident-prone area will be reminded in real time.

[0069] In the accident-prone area identification method of the present invention, step 102 includes the following sub-steps S1-S4:

[0070] In step S1, the collision position of the first historical accident alarm is extracted from the historical alarm accident data as the accident location. The historical accident alarms include front collision, airbag violent deployment, rear collision, etc. The longitude and latitude of these alarm events are taken and duplicated, thereby obtaining the collision point set L m,k ;L m,k ={(lng 1,1 ,lat 1,1 ,time 1,1 ),(lng 1,2 ,lat 1,2 ,time 1,2 )…(lng m,k ,lat m,k ,time m,k )}, where m is the mth vehicle that has collided, k is the kth collision of the mth collision vehicle, lng and lat are the longitude and latitude of the collision point respectively, and time m,k is the time when the kth collision of the mth collision vehicle occurs; the set of pre-collision speeds S is obtained according to the collision point m,k =(s 1,1 ,s 1,2 …s m,k ), where s m,k , is the speed of the m-th collision vehicle before the k-th collision; where; the temperature t corresponding to the k-th collision point of the m-th vehicle is calculated by m,k Find the mean Thus, the temperature before the collision is obtained; the mean Let x be the temperature, and divide it into different levels according to the rules. The corresponding formula is: T1, T2…T N Respectively represent the threshold value of division; based on regularized matching, the climate and wind are combined to divide the levels, where x represents the climate and wind characteristics corresponding to this collision point, and the corresponding formula is: c rei Indicates the relevant climate conditions at level i, where i = 1, 2…M1, and the priority of the rules is from M1 to 1; w rei Indicates the relevant situation of wind force at level i, where i = 1, 2…M1, and the priority of the rules is also from M to 1; for example: climate c reM1 Heavy rain or wind reM1 If the value is greater than 7, the level is M1 (bad), thus forming N*M1 combinations of environments.

[0071] Table 1 shows the environmental combinations

[0072] Temperature\Climate 1 2 … M 1 1,1 1,2 … 1,M 2 2,1 2,2 … 2,M … … … … … N N,1 N,2 … N,M

[0073] In step S2, based on the GEOHASH algorithm, the collision point set L m,k Perform a-bit GEOHASH encoding to obtain the corresponding GEOHASH encoding dataset G m,k [a]=(g 1,1 [a],g 1,2 [a]…g m,k [a]), where G m,k The same value exists in [a];

[0074] In step S3, the environmental data set C at the time of collision is obtained according to the collision point. o,e , where the environmental data calculation formula is:

[0075] {(c 1,1 ,w 1,1 ,t 1,1 ,di 1,1 ,date1),

[0076] C o,e =(c 1,2 ,w 1,2 ,t 1,2 ,di 1,2 ,date2),

[0077] …

[0078] (c o,e ,w o,e ,t o,e ,di o,e ,date o )}

[0079] Where date is the date corresponding to the environmental data, o is the date of the oth day, and each day contains e cities; di represents the address of each province and city area, that is, di o,e is the address of the city area in the province e on the oth day; c represents the climate, c r,j is the specific climate of the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; w represents the wind force, w r,j is the specific wind speed in the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; t represents the temperature of the day, t r,j Represents the specific temperature of the jth city area on the i-th day, where r = 1, 2...o; j = 1, 2...e; the environmental data set C o,e Through variables such as address and time (d, date) and the collision point set L obtained in step S1 m,kThe time and address (D, time) are spliced together, and the corresponding environmental data when the collision occurs is GEOHASH encoded by the GEOHASH algorithm to obtain a collection L with environmental and GEOHASH encoding. m,k

[0080] (lng 1,2 ,lat 1,2 ,g 1,2 [a],time 1,2 ,d 1,2 ,s 1,2 ,c 1,2 ,w 1,2 ,t 1,2 )

[0081] …

[0082] (lng m,k ,lat m,k ,g m,k [a],time m,k ,d m,k ,s m,k ,c m,k ,w m,k ,t m,k )}

[0083] Obtain road network data based on OSM to obtain the collision point province, city, and region address D m,k =(d 1,1 ,d 1,2 …d m,k ), match the obtained b-digit GEOHASH code of the collision point's latitude and longitude with the b-digit GEOHASH code of the latitude and longitude before the collision point. If they match, the collision point is considered to be address D in the road network data; collect daily environmental data for cities in each province, including climate, wind speed, temperature, address, and date. For example: (Climate: Thunderstorm, Wind speed: Level 2, Temperature:, Address: Hongshan District, Wuhan City, Hubei Province, Date: 20210301),

[0084] In step S4, by setting the accident occurrence threshold Q, it is determined whether the area is a historical accident-prone area when the number of accidents in different environmental combinations is q. When q≥Q, the area is a historical accident-prone area.

[0085] In step S1, historical vehicle alarm data is acquired to determine the accident location. Based on the vehicle's existing component sensors, the collision locations of the first alarms (front-end collisions, airbag deployments, and rear-end collisions) are used as the accident locations. Simultaneously, vehicle-mounted equipment collects the vehicle's driving speed data before the alarms at these accident locations, as well as the time, location, and environmental data of the alarms. The latitude and longitude points of the historical vehicle alarm data are then GEOHASH-encoded. A 7-bit GEOHASH encoding is performed based on the collision point set. A partial GEOHASH encoding dataset is shown in Table 2. The GEOHASH codes contain identical values, such as ws45f5y and wkje2qs.

[0086] Table 2 shows the GEOHASH coding examples corresponding to the collision points.

[0087] Collision point coding GEOHASH encoding A00000000001 ws45f5y A00000000002 ws45f5y A00000000003 tz3407t A00000000004 wkje2qs A00000000005 wkje2qs A00000000006 wwk029d A00000000007 ws07rvj A00000000008 wkje2qs A00000000009 wpd3gv1 … …

[0088] This system collects daily environmental data from cities across provinces, including weather and temperature data. For example, weather data includes thunderstorms, wind speed level 2, and temperatures between 10°C and 20°C. Based on the longitude and latitude of the alarm data, the system uses road network data to obtain information about the province's cities and regions, as well as the weather, wind speed, and temperature at the specific time of the accident. Based on the average temperature rule, temperatures are divided into three levels: high, medium, and low (high temperatures are greater than 35°C, low temperatures are less than 5°C, and medium temperatures are between 5°C and 35°C). Based on regularized matching, the climate is divided into three levels: good, average, and poor. The climate conditions including "wind force greater than level 6", "heavy rain", "heavy snow", "blizzard", "hail", "heavy haze", "sandstorm", "dense fog", "strong sandstorm", and "heavy rain" are classified as "bad"; the climate conditions including "wind force 4-5", "rain and snow", "mixed rain and snow", "heavy rain", "haze", "fog", "moderate rain", "lightning", and "thunderstorm" are classified as "normal"; the climate conditions including "wind force 1-3", "calm", "sunny", "cloudy", and "light rain" are classified as "good". This forms a 3*3 combination of environments; Table 3 shows the environmental combinations.

[0089] Weather\Temperature good generally bad high Good*High Generally * High Severe*High middle Good* General* Severe* Low Good*Low Generally * Low Harsh* Low

[0090] By setting a threshold, the GEOHASH 7-bit matching code of the collision point in step S1 and the corresponding GEOHASH 7-bit code of the surrounding 8 areas are obtained. An adjacent 3*3 combination of accident-prone areas is formed. When the number of accident sites of a certain environmental combination in a certain 3*3 adjacent area reaches the threshold, that is, when the number of accidents under the corresponding environment is 6, it is considered that this place is an accident-prone area based on a certain climate. This 3*3 adjacent area is used as the monitoring area for the accident site, and the GEOHASH 7-bit code and corresponding environmental layered data that do not reach the accident-prone area threshold are retained. It is used for updating and iterating accident-prone areas when new alarm data is generated; Table 4 shows the corresponding field combination of accident-prone areas.

[0091] accident-prone locations 8 surrounding GEOHASH areas Temperature grade Climate level Minimum speed (km / h) Number of accidents ws45f5y ws45f5y middle bad 86 7 wkje2qs ws45f5y middle generally 74 6 wpd3gv1 tz3407t high good 110 10 ws07rvj wkje2qs high bad 100 7 tz3407t wkje2qs Low generally 66 6 … …

[0092] Table 5 shows the field combinations corresponding to non-accident-prone locations.

[0093] Failure to reach accident-prone areas 8 surrounding GEOHASH areas Temperature grade Climate level Minimum speed ws45f5y ws45f5y middle bad 56 wkje2qs ws45f5y middle generally 64 wpd3gv1 tz3407t high good 100 ws07rvj wkje2qs high bad 90 tz3407t wkje2qs Low generally 48 … …

[0094] In the accident-prone area identification method of the present invention, step 102 further includes the following steps:

[0095] In step S5, when the number of accidents reaches Q, the corresponding a-bit GEOHASH code is G0[a]. Since the collision points may be concentrated at the critical point of the coding area, the a-bit GEOHASH codes of the eight areas around the code are obtained, that is, a 3*3 adjacent range code combination is formed as the accident-prone area, which is expressed as follows: The information stored in the accident-prone areas here includes a bit of GEOHASH code corresponding to the set of all accident-prone areas in, is the mean of the pre-collision speeds of all collision points in the accident-prone area; the set of accident locations that do not reach the accident-prone area threshold is denoted as l v ; where l v ={(g1[a],T1,C1,s1),(g2[a],T2,C2,s2)…(g v [a],T v ,C v ,s v )}, corresponding to GEOHASH code, temperature layer, climate layer and speed before the accident, respectively.

[0096] By setting the accident occurrence threshold Q, when the accident occurrence number q in different environmental combinations reaches the threshold Q, that is: q ≥ Q, this place is considered to be an accident-prone area, such as GEOHASH code and corresponding 8 areas. Table 6 shows the GEOHASH code with environmental combination and the corresponding 8 areas and corresponding environmental levels;

[0097] Table 6 shows the Geohash codes with environmental combinations and the corresponding 8 regions.

[0098]

[0099] In the accident-prone area identification method of the present invention, step 103 includes the following sub-steps S31-S35;

[0100] In step S31, a collection period is set, wherein the collection time is set for each operation. v Combined with the accident locations generated by real-time accident alarm data to form new accident-prone areas;

[0101] In step S32, based on the real-time accident alarm data, the collision position at the time of the real-time accident alarm is obtained, the longitude and latitude of the collision position at the time of the real-time accident alarm is obtained and the collision point set L is obtained by removing the duplicates. m,k ={(lng 1,1 ,lat 1,1 ,time 1,1 ),(lng 1,2 ,lat 1,2 ,time 1,2 )…(lng m,k ,lat m,k ,time m,k )}, where m is the mth vehicle that has collided, k is the kth collision of the mth collision vehicle, lng and lat are the longitude and latitude of the collision point respectively, and time m,k is the time when the kth collision of the mth collision vehicle occurs; the set of pre-collision speeds S is obtained according to the collision point m,k =(s 1,1 ,s 1,2 …s m,k ), where s m,k , is the speed of the m-th collision vehicle before the k-th collision; where; the temperature t corresponding to the k-th collision point of the m-th vehicle is calculated by m,k Find the mean Thus, the temperature before the collision is obtained; the mean Let x be the temperature, and divide it into different levels according to the rules. The corresponding formula is: T1, T2…T N Respectively represent the threshold value of division; based on regularized matching, the climate and wind are combined to divide the levels, where x represents the climate and wind characteristics corresponding to this collision point, and the corresponding formula is: c rei Indicates the relevant climate conditions at level i, where i = 1, 2…M2, and the priority of the rules is from M2 to 1; wrei Indicates the relevant situation of wind force at level i, where i = 1, 2…M2, and the priority of the rules is also from M2 to 1;

[0102] In step S33, based on the GEOHASH algorithm, the collision point set L m,k Perform a-bit GEOHAS encoding to obtain the corresponding GEOHASH encoding dataset G m,k [a]=(g 1,1 [a],g 1,2 [a]…g m,k [a]), where G m,k The same value exists in [a];

[0103] In step S34, the environmental data set C at the time of collision is obtained according to the collision point. o,e , where the environmental data calculation formula is:

[0104] {(c 1,1 ,w 1,1 ,t 1,1 ,di 1,1 ,date1),

[0105] C o,e =(c 1,2 ,w 1,2 ,t 1,2 ,di 1,2 ,date1),

[0106] …

[0107] (c o,e ,w o,e ,t o,e ,di o,e ,date o )}

[0108] Where date is the date corresponding to the environmental data, o is the date of the oth day, and each day contains e cities; di represents the province, city, and region address, i.e. di o,e is the address of the city area in the province e on the oth day; c represents the climate, c r,j is the specific climate of the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; w represents the wind force, w r,j represents the specific wind speed in the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; t represents the temperature of the day, t r,j Represents the specific temperature of the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; the environmental data set C o,eBy comparing the address and time variables (d, date) with the collision point set L obtained in step S32 m,k The time and address (D, time) are spliced together, and the corresponding environmental data when the collision occurs is GEOHASH encoded by the GEOHASH algorithm to obtain a collection with environmental and GEOHASH encoding l f,h ={(lng 1,1 ,lat 1,1 ,g 1,1 [a],time 1,1 ,d 1,1 ,s 1,1 ,T 1,1 ,C 1,1 ),

[0109] (lng 1,2 ,lat 1,2 ,g 1,2 [a],time 1,2 ,d 1,2 ,s 1,2 ,T 1,2 ,C 1,2 )

[0110] …

[0111] (lng f,h ,lat f,h ,g f,h [a],time f,h ,d f,h ,s f,h ,T f,h ,C f,h )}

[0112] Obtain road network data based on OSM to obtain the address D of each province and city area of the collision point m,k =(d 1,1 ,d 1,2 …d m,k ), match the obtained b-digit GEOHASH code of the longitude and latitude of the collision point with the b-digit GEOHASH code of the longitude and latitude before the collision point. If they are the same, the collision point is considered to be the address D on the road network data;

[0113] In step S35; by setting the accident occurrence threshold Q, it is determined whether it is a new accident-prone area when the number of accidents in different environmental combinations is q; when the GEOHASH code of the new accident-prone area is the same as the GEOHASH code of the historical accident-prone area, and the environmental classification is consistent, it is classified as an accident-prone area and the collision information is deleted, and only the collision points that have not reached the accident-prone area are retained; the retained collision points are set as new accident-prone areas, thereby updating the accident-prone areas.

[0114] In the accident-prone area identification method of the present invention, step 104 includes the following sub-steps S41-S43:

[0115] In step S41, the trip data B transmitted by the vehicle terminal is obtained. M , where B M =(lng1,lat1,time1,s1) M , where M is the number of vehicles with all driving behaviors, lng1, lat1 are the longitude and latitude information collected by the vehicle in real time, time1 is the time when the data is generated, and s1 represents the driving speed of the vehicle; GEOHASH code and corresponding address information are obtained through real-time driving behavior data;

[0116] In step S42, the road network data is obtained based on OSM to obtain the environmental information of the day, and the real-time temperature and climate level are obtained by associating the environmental information with the driving information, thereby obtaining the trip data B with the environment. M ; That is B M =(lng1,lat1,time1,s1,g1[a],T1,C1) M ;

[0117] In step S43, when a vehicle approaches or enters an accident-prone area under the corresponding environmental level, a real-time reminder of the accident-prone area is issued to the vehicle, and the vehicle is reminded to control the driving speed.

[0118] In the accident-prone area identification method of the present invention, step 104 further includes:

[0119] In step S410, obtain a bit GEOHASH code through the travel data, and associate the road network data obtained based on OSM in step S3 to obtain the collision point address D of each province and city area m,k =(d 1,1 ,d 1,2 …d m,k ), based on the b-bit GEOHASH code matching, the corresponding provincial city area address information is obtained; that is, based on the acquired road network data and the corresponding b-bit GEOHASH code, the b-bit GEOHASH before the collision point is matched. If they are the same, the collision point is considered to be an address on the road network.

[0120] like Figure 2 As shown, based on the same inventive concept, the present invention also provides an accident-prone area identification device, the device comprising:

[0121] The data acquisition and processing module is used to obtain the travel data, historical vehicle accident alarm data, real-time accident alarm data transmitted by the vehicle terminal, and the daily environmental data of cities in each province collected by the weather website API, and pre-process the travel data, historical vehicle accident alarm data, real-time accident alarm data, and daily environmental data of cities in each province collected by the weather website API;

[0122] The historical accident-prone areas generation module is used to perform GEOHASH matching on historical vehicle accident alarm data and generate historical accident-prone areas based on whether the number of accidents reaches a threshold;

[0123] The accident-prone area update module combines real-time accident alarm data with accident locations that do not reach the accident-prone area threshold to form a new accident location. By performing GEOHASH matching on the new alarm collision point data obtained from the real-time alarm data, the collision points that do not reach the accident-prone area are retained; the retained collision points are set as new accident-prone areas, thereby updating the accident-prone areas.

[0124] The accident area monitoring module is used to monitor whether the vehicle enters the accident-prone area while driving, issue a reminder to the vehicle entering the accident-prone area, and remind the vehicle to control the driving speed.

[0125] Another aspect of this embodiment further provides a computer device, including:

[0126] One or more processors; a storage device for storing one or more programs,

[0127] When one or more programs are executed by one or more processors, the one or more processors implement the accident-prone area identification method provided by any embodiment of the present invention.

[0128] In another aspect of this embodiment, a computer readable storage medium is provided on which a computer program is stored.

[0129] When executed, the accident-prone area identification method provided by any embodiment of the present invention is implemented.

[0130] The accident-prone area identification method, equipment, medium, and device provided by the present invention have at least the following advantages:

[0131] 1. Overcome the difficulty faced by car manufacturers in obtaining real-time information on accident-prone areas and driving safety reminders.

[0132] 2. The present invention effectively integrates the accident-prone area identification and early warning functions, effectively utilizes and integrates the OEM resources, and improves data utilization.

[0133] 3. Continuously update information on accident-prone areas and synchronize information.

[0134] Various operations of the embodiments are provided herein. In one embodiment, the one or operations described may constitute computer-readable instructions stored on one or computer-readable media, which, when executed by an electronic device, will cause a computing device to perform the operations. The order in which some or all operations are described should not be interpreted as implying that these operations must be order-dependent. Those skilled in the art will appreciate alternative sequencing with the benefit of this specification. Furthermore, it should be understood that not all operations must be present in each embodiment provided herein.

[0135] Furthermore, the word "preferred" as used herein is intended to serve as an example, instance, or illustration. Any aspect or design described herein as "preferred" is not necessarily to be construed as advantageous over other aspects or designs. Rather, the use of the word "preferred" is intended to present concepts in a concrete manner. As used in this application, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless otherwise specified or clear from the context, "X employs A or B" is intended to mean any of the naturally inclusive permutations. That is, if X employs A; X employs B; or X employs both A and B, then "X employs A or B" is satisfied in any of the aforementioned instances.

[0136] Moreover, although the present disclosure has been shown and described with respect to one or implementations, those skilled in the art will think of equivalent variations and modifications based on reading and understanding of this specification and the accompanying drawings. The present disclosure includes all such modifications and variations and is limited only by the scope of the appended claims. In particular, with respect to the various functions performed by the above-mentioned components (such as elements, resources, etc.), the terms used to describe such components are intended to correspond to any component (unless otherwise indicated) that performs the specified function of the component (such as it is functionally equivalent), even if it is structurally different from the disclosed structure that performs the function in the exemplary implementation of the present disclosure shown herein. In addition, although the specific features of the present disclosure have been disclosed with respect to only one of several implementations, such features can be combined with one or other features of other implementations that may be desired and advantageous for a given or specific application. Moreover, insofar as the terms "including", "having", "containing" or their variations are used in specific embodiments or claims, such terms are intended to be included in a manner similar to the term "comprising".

[0137] The functional units in the embodiments of the present invention may be integrated into a single processing module, each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated module may be implemented in the form of hardware or a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium. The aforementioned storage medium may be a read-only memory, a magnetic disk, or an optical disk, etc. The aforementioned devices or systems may execute the methods in the corresponding method embodiments.

[0138] In summary, although the present invention has been disclosed above with reference to preferred embodiments, the above preferred embodiments are not intended to limit the present invention. A person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be based on the scope defined in the claims.

Claims

1. A method for identifying accident-prone areas, characterized in that: The following steps are involved: a data acquisition and processing step for acquiring travel data, historical vehicle accident alarm data, real-time accident alarm data, and daily environmental data of cities in various provinces collected by a weather website API transmitted by an on-board terminal, and pre-processing the travel data, historical vehicle accident alarm data, real-time accident alarm data, and daily environmental data of cities in various provinces collected by a weather website API; A historical accident-prone area generation step, comprising: constructing a historical accident-prone area generation model by performing GEOHASH coding on the vehicle historical accident alarm data; An accident-prone area updating step, constructing an accident-prone area updating model by performing GEOHASH encoding on the real-time accident alarm data; The accident area monitoring step involves performing GEOHASH matching on the travel data and environmental data transmitted by the vehicle terminal to determine the accident-prone area. The monitoring model is used to monitor whether the vehicle enters the accident-prone area during driving. If so, the vehicle's driving speed in the accident-prone area is reminded in real time. According to the collision point, the environmental data set at the time of collision is obtained, and the environmental data set is spliced with the two associated keys of time and address of the obtained collision point set through the address and time variables. The environmental data corresponding to the collision is GEOHASH encoded by the GEOHASH algorithm to obtain a collection with environment and GEOHASH codes.

2. The accident-prone area identification method according to claim 1, characterized in that: The step of generating a historical accident-prone area includes: S1. Based on the historical accident alarm data, the collision position at the time of the historical accident alarm is obtained. The longitude and latitude of the collision position at the time of the historical accident alarm are obtained by removing duplicates to obtain a collision point set L. m,k ={(lng 1,1 ,lat 1,1 ,time 1,1 ),(lng 1,2 ,lat 1,2 ,time 1,2 )…(lng m,k ,lat m,k ,time m,k )}, where m is the mth vehicle that has collided, k is the kth collision of the mth collision vehicle, lng and lat are the longitude and latitude of the collision point respectively, and time m,k is the time when the kth collision of the mth collision vehicle occurs; the set of pre-collision speeds S is obtained according to the collision point m,k =(s 1,1 ,s 1,2 …s m,k ), where s m,k , is the speed of the m-th collision vehicle before the k-th collision; where; the temperature t corresponding to the k-th collision point of the m-th vehicle is calculated by m,k Find the mean Thus, the temperature before the collision is obtained; the mean Let x be the temperature, and divide it into different levels according to the rules. The corresponding formula is: T1, T2…T N Respectively represent the threshold value of division; based on regularized matching, the climate and wind are combined to divide the levels, where x1 represents the climate and wind characteristics corresponding to this collision point, and the corresponding formula is: c rei Indicates the relevant climate conditions at level i, where i = 1, 2…M1, and the priority of the rules is from M1 to 1; w rei Indicates the relevant situation of wind force at level i, where i = 1, 2…M1, and the priority of the rules is also from M1 to 1; S2, based on the GEOHASH algorithm, the collision point set L m,k Perform a-bit GEOHASH encoding to obtain the corresponding GEOHASH encoding dataset G m,k [a]=(g 1,1 [a],g 1,2 [a]…g m,k [a]), where G m,k The same value exists in [a]; S3. Acquire a set of environmental data C when the collision occurs according to the collision point. o,e , where the environmental data calculation formula is: {(c 1,1 ,w 1,1 ,t 1,1 ,di 1,1 ,date1), C o,e =(c 1,2 ,w 1,2 ,t 1,2 ,di 1,2 ,date2), … (c o,e ,w o,e ,t o,e ,di o,e ,date o )} Where date is the date corresponding to the environmental data, o is the date of the oth day, and each day contains e cities; di represents the address of each province and city area, that is, di o,e is the address of the city area in the province e on the oth day; c represents the climate, c r,j is the specific climate of the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; w represents the wind force, w r,j is the specific wind speed in the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; t represents the temperature of the day, t r,j Represents the specific temperature of the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; the environmental data set C o,e The address and time variables are compared with the collision point set L obtained in step S1. m,k The time and address (D, time) are spliced together, and the environment data corresponding to the collision is GEOHASH coded by the GEOHASH algorithm to obtain a collection L with environment and GEOHASH coding. m,k ={(lng 1,1 ,lat 1,1 ,g 1,1 [a],time 1,1 ,d 1,1 ,s 1,1 ,c 1,1 ,w 1,1 ,t 1,1 ), (lng 1,2 ,lat 1,2 ,g 1,2 [a],time 1,2 ,d 1,2 ,s 1,2 ,c 1,2 ,w 1,2 ,t 1,2 ) … (lng m,k ,lat m,k ,g m,k [a],time m,k ,d m,k ,s m,k ,c m,k ,w m,k ,t m,k )} Obtain the road network data based on OSM to obtain the province, city, and region address D of the collision point. m,k =(d 1,1 ,d 1,2 …d m,k ), matching the obtained b-digit GEOHASH code of the longitude and latitude of the collision point with the b-digit GEOHASH code of the longitude and latitude before the collision point, if they are the same, the collision point is considered to be the address D on the road network data; S4. By setting a threshold value Q of the number of accidents, it is determined whether the area is a historical accident-prone area when the number of accidents in different environmental combinations is q. When q≥Q, the area is a historical accident-prone area.

3. The accident-prone area identification method according to claim 2, characterized in that: The step of generating a historical accident-prone area further includes: When the number of accidents reaches Q, the corresponding a-bit GEOHASH code is G0[a]. Since the collision points may be concentrated at the critical point of the coding area, the a-bit GEOHASH codes of the eight areas around the code are obtained, that is, a 3*3 adjacent range code combination is formed as the accident-prone area, which is expressed as follows: The information stored in the accident-prone areas here includes a bit of GEOHASH code corresponding to the set of all accident-prone areas in, is the mean of the pre-collision speeds of all collision points in the accident-prone area; the set of accident locations that do not reach the accident-prone area threshold is denoted as l v ; where l v ={(g1[a],T1,C1,s1),(g2[a],T2,C2,s2)…(g v [a],T v ,C v ,s v )}, corresponding to GEOHASH code, temperature stratification, climate stratification and speed before the accident, respectively.

4. The accident-prone area identification method according to claim 3, characterized in that: The accident-prone area updating step includes: S31, by l v Combined with the accident locations generated by real-time accident alarm data to form new accident-prone areas; S32: Based on the real-time accident alarm data, obtain the collision position at the time of the real-time accident alarm, take the longitude and latitude of the collision position at the time of the real-time accident alarm, remove the duplicates, and obtain a collision point set L m,k ={(lng 1,1 ,lat 1,1 ,time 1,1 ),(lng 1,2 ,lat 1,2 ,time 1,2 )…(lng m,k ,lat m,k ,time m,k )}, where m is the mth vehicle that has collided, k is the kth collision of the mth collision vehicle, lng and lat are the longitude and latitude of the collision point respectively, and time m,k is the time when the kth collision of the mth collision vehicle occurs; the set of pre-collision speeds S is obtained according to the collision point m,k =(s 1,1 ,s 1,2 …s m,k ), where s m,k , is the speed of the m-th collision vehicle before the k-th collision; in; By calculating the temperature t corresponding to the k-th collision point of the m-th vehicle m,k Find the mean Thus, the temperature before the collision is obtained; the mean Let x be the temperature, and divide it into different levels according to the rules. The corresponding formula is: T1, T2…T N Respectively represent the threshold value of division; Based on regularized matching, the climate and wind are combined and divided into levels, where x2 represents the climate and wind characteristics corresponding to this collision point. The corresponding formula is: c rei Indicates the relevant climate conditions at level i, where i = 1, 2…M2, and the priority of the rules is from M2 to 1; w rei Indicates the relevant situation of wind force at level i, where i = 1, 2…M2, and the priority of the rules is also from M2 to 1; S33, based on the GEOHASH algorithm, the collision point set L m,k Perform a-bit GEOHAS encoding to obtain the corresponding GEOHASH encoding dataset G m,k [a]=(g 1,1 [a],g 1,2 [a]…g m,k [a]), where G m,k The same value exists in [a]; S34, obtaining a set of environmental data C when the collision occurs according to the collision point o,e , where the environmental data calculation formula is: {(c 1,1 ,w 1,1 ,t 1,1 ,di 1,1 ,date1), C o,e =(c1,2,w1,2,t1,2,di1,2,date2), … (c o,e ,w o,e ,t o,e ,di o,e ,date o )} Where date is the date corresponding to the environmental data, o is the date of the oth day, and each day contains e cities; di represents the province, city, and region address, i.e. di o,e is the address of the city area in the province e on the oth day; c represents the climate, c r,j is the specific climate of the jth urban area on the rth day, where r = 1, 2…o; j = 1, 2…e; w represents the wind force, w r,j represents the specific wind speed in the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; t represents the temperature of the day, t r,j Represents the specific temperature of the jth city area on the rth day, where r = 1, 2…o; j = 1, 2…e; the environmental data set C o,e By comparing the variables such as address and time with the collision point set L obtained in step S32 m,k The time and address (D, time) are spliced together, and the environment data corresponding to the collision is GEOHASH coded by the GEOHASH algorithm to obtain a collection with environment and GEOHASH coding l f,h ={(lng 1,1 ,lat 1,1 ,g 1,1 [a],time 1,1 ,d 1,1 ,s 1,1 ,T 1,1 ,C 1,1 ), (lng 1,2 ,lat 1,2 ,g 1,2 [a],time 1,2 ,d 1,2 ,s 1,2 ,T 1,2 ,C 1,2 ) … (lng f,h ,lat f,h ,g f,h [a],time f,h ,d f,h ,s f,h ,T f,h ,C f,h )} Obtain road network data based on OSM to obtain the address D of each province and city area of the collision point m,k =(d 1,1 ,d 1,2 …d m,k ), matching the obtained b-digit GEOHASH code of the longitude and latitude of the collision point with the b-digit GEOHASH code of the longitude and latitude before the collision point, if they are the same, the collision point is considered to be the address D on the road network data; S35. By setting a threshold value Q for the number of accidents, it is determined whether it is a new accident-prone area when the number of accidents in different environmental combinations is q; when the GEOHASH code of the new accident-prone area is the same as the GEOHASH code of the historical accident-prone area and the environmental classification is consistent, it is classified as an accident-prone area and the collision information is deleted, and only collision points that have not reached the accident-prone area are retained; the retained collision points are set as new accident-prone areas, thereby updating the accident-prone areas.

5. The accident-prone area identification method according to claim 4, characterized in that: The accident occurrence area monitoring step includes: S41, by obtaining the travel data B transmitted by the vehicle terminal M , where B M =(lng1,lat1,time1,s1) M , where M is the number of vehicles with all driving behaviors, lng1, lat1 are the longitude and latitude information collected by the vehicle in real time, time1 is the time when the data is generated, and s1 represents the driving speed of the vehicle; GEOHASH code and corresponding address information are obtained through real-time driving behavior data; S42. Obtain the road network data based on OSM to obtain the environmental information of the day. By associating the environmental information with the driving information, obtain the real-time temperature and climate level, and thus obtain the trip data with the environment B M ; That is B M =(lng1,lat1,time1,s1,g1[a],T1,C1) M ; S43. When a vehicle approaches or enters an accident-prone area under a corresponding environmental level, a real-time reminder of the accident-prone area is issued to the vehicle, and the vehicle is reminded to control its driving speed.

6. The accident-prone area identification method according to claim 5, characterized in that: The step S41 further includes: S410, obtain a GEOHASH code through the travel data, and associate it with the road network data obtained based on OSM in step S3 to obtain the address D of each province and city area of the collision point m,k =(d 1,1 ,d 1,2 …d m,k ), based on the b-bit GEOHASH code matching, the corresponding provincial city area address information is obtained; that is, based on the acquired road network data and the corresponding b-bit GEOHASH code, the b-bit GEOHASH before the collision point is matched. If they are the same, the collision point is considered to be an address on the road network.

7. A device for identifying accident-prone areas, characterized in that: The device comprises: The data acquisition and processing module is used to obtain the travel data, historical vehicle accident alarm data, real-time accident alarm data transmitted by the vehicle terminal, and the daily environmental data of cities in each province collected by the weather website API, and pre-process the travel data, historical vehicle accident alarm data, real-time accident alarm data, and daily environmental data of cities in each province collected by the weather website API; The historical accident-prone areas generation module is used to perform GEOHASH matching on historical vehicle accident alarm data and generate historical accident-prone areas based on whether the number of accidents reaches a threshold; The accident-prone area update module combines real-time accident alarm data with accident locations that do not reach the accident-prone area threshold to form a new accident location. It then performs GEOHASH matching on new alarm collision point data obtained from real-time alarm data, retaining collision points that do not reach the accident-prone area. The retained collision points are then set as new accident-prone areas, thereby updating the accident-prone areas. The accident area monitoring module is used to monitor whether the vehicle enters the accident-prone area while driving, issue a warning to the vehicle entering the accident-prone area, and remind the vehicle to control the driving speed; According to the collision point, the environmental data set at the time of collision is obtained, and the environmental data set is spliced with the two associated keys of time and address of the obtained collision point set through the address and time variables. The environmental data corresponding to the collision is GEOHASH encoded by the GEOHASH algorithm to obtain a collection with environment and GEOHASH codes.

8. A computer device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the accident-prone area identification method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the accident-prone area identification method according to any one of claims 1 to 6 is implemented.

Images