CN110264711A - A kind of traffic accident method of determining probability and device - Google Patents
A kind of traffic accident method of determining probability and device Download PDFInfo
- Publication number
- CN110264711A CN110264711A CN201910458355.8A CN201910458355A CN110264711A CN 110264711 A CN110264711 A CN 110264711A CN 201910458355 A CN201910458355 A CN 201910458355A CN 110264711 A CN110264711 A CN 110264711A
- Authority
- CN
- China
- Prior art keywords
- traffic accident
- information
- section
- attribute
- time period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G08G1/0125—Traffic data processing
-
- 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
Abstract
This application discloses a kind of traffic accident method of determining probability and devices, are related to technical field of transportation, for improving the accuracy rate of traffic accident prediction.It include: the section attribute for obtaining target road section;The section attribute includes one or more of following information: category of roads, number of track-lines grade, curvature, the gradient and highway crosspoint/highway conversion entrance JCT/IC attribute;Obtain information of vehicles of the target road section in first time period;Wherein, the information of vehicles includes the quantity of vehicle and the speed of the vehicle;According to the time attribute of second time period, the information of vehicles, the section attribute and street accidents risks assessment models, traffic accident probability of the target road section in the second time period is determined;Wherein, the street accidents risks assessment models are for determining traffic accident probability, before the first time period is the second time period and the period adjacent with the second time period.
Description
Technical field
The invention relates to technical field of transportation more particularly to a kind of traffic accident method of determining probability and dress
It sets.
Background technique
Traffic safety is the most important thing of road traffic, for traffic-information service provider, how to pass through road conditions
Numerical analysis is the core of traffic safety to avoid traffic accident.
A kind of method that tradition carries out traffic accident prediction is based on backpropagation (back propagation, BP) nerve
The statistical models such as network and autoregressive moving average are predicted that these methods are predicted according to fractional value amount, the form of expression
Number occurs for "ball-park" estimate year or monthly accident;Another method is based on deep neural network (deep neural
Network, DNN) city expressway prediction is carried out, but there are following problems with the method: a DNN relies only on telecommunication flow information,
Have ignored the influence of section attribute and time attribute to traffic accident prediction;Two DNN do not consider traffic accident Time-space serial
Connection.Therefore the accuracy rate of traffic accident prediction is lower in practical engineering applications.
Summary of the invention
Embodiments herein provides a kind of traffic accident method of determining probability and device, pre- for improving traffic accident
The accuracy rate of survey.
In order to achieve the above objectives, embodiments herein adopts the following technical scheme that
In a first aspect, a kind of traffic accident method of determining probability is provided, this method comprises:
Obtain the section attribute of target road section;Section attribute includes one or more of following information: category of roads, vehicle
Road number grade, curvature, the gradient and highway crosspoint (junction, JCT)/highway conversion entrance (inter
Change, IC) attribute;
Obtain information of vehicles of the target road section in first time period;Wherein, information of vehicles includes the quantity and vehicle of vehicle
Speed;
According to the time attribute of second time period, information of vehicles, section attribute and street accidents risks assessment models, really
Set the goal traffic accident probability of the section in second time period;Wherein, street accidents risks assessment models are for determining traffic
Accident probability, first time period be second time period before and the period adjacent with second time period.
Traffic accident method of determining probability provided by the embodiments of the present application, according to the section attribute of target road section, first
The time attribute and street accidents risks of the second time period after information of vehicles, first time period in period assess mould
Type determines the traffic accident probability of target road section.It, can be with by multi-analysis such as time attribute, information of vehicles, section attributes
Improve the accuracy rate of the traffic accident probabilistic forecasting of target road section.
Second aspect, provides a kind of determining device of traffic accident probability, which can be server, or
Applied to the chip of server, the apparatus may include:
Communication unit, for obtaining the section attribute of target road section;Section attribute includes one or more in following information
It is a: category of roads, number of track-lines grade, curvature, the gradient and JCT/IC attribute;
Communication unit is also used to obtain information of vehicles of the target road section in first time period;Wherein, information of vehicles includes
The quantity of vehicle and the speed of vehicle;
Determination unit, for according to the time attribute of second time period, information of vehicles, section attribute and traffic accident wind
Dangerous assessment models determine traffic accident probability of the target road section in second time period;Wherein, street accidents risks assessment models
For determining traffic accident probability, before first time period is second time period and the period adjacent with second time period.
The third aspect provides a kind of readable storage medium storing program for executing, instruction is stored in readable storage medium storing program for executing, when instruction is performed
When, realize the method such as first aspect.
Fourth aspect provides a kind of computer program product, and computer program product includes at least one instruction, when extremely
When a few instruction is run on computers, so that computer executes the method such as first aspect.
5th aspect, provides a kind of chip, and chip includes at least one processor and communication interface, communication interface and at least
One processor coupling, at least one processor is for running computer program or instruction, the method to realize first aspect.
The device or computer storage medium or computer program product or chip of above-mentioned offer are used to execute institute above
The corresponding method provided, therefore, the attainable beneficial effect of institute can refer to counterparty in corresponding method provided above
The beneficial effect of case, details are not described herein again.
Detailed description of the invention
Fig. 1 is a kind of JCT structural schematic diagram that embodiments herein provides;
Fig. 2 is a kind of IC structural schematic diagram that embodiments herein provides;
Fig. 3 is a kind of traffic accident method of determining probability flow diagram one that embodiments herein provides;
Fig. 4 is a kind of traffic accident method of determining probability flow diagram two that embodiments herein provides;
Fig. 5 be embodiments herein provide a kind of shot and long term memory network (long short-term memory,
LSTM structural schematic diagram);
Fig. 6 is the knot for a kind of cycling element (gated recurrent unit, GRU) that embodiments herein provides
Structure schematic diagram;
Fig. 7 is a kind of structural schematic diagram one of the determining device for traffic accident probability that embodiments herein provides;
Fig. 8 is a kind of structural schematic diagram two of the determining device for traffic accident probability that embodiments herein provides;
Fig. 9 is a kind of structural schematic diagram for chip that embodiments herein provides.
Specific embodiment
Below first to the invention relates to related terms simply introduced.
JCT:JCT refers to linking-up road (such as highway and highway, highway between two highways
With expressway around city road), Fig. 1 shows a kind of JCT structural schematic diagram of the embodiment of the present application.
IC:IC refer between connection highway and other different brackets roads road (uniquely upper highway or
Locate the road of highway), Fig. 2 shows a kind of IC structural schematic diagrams of the embodiment of the present application.
Time attribute: it includes whether as festivals or holidays, whether be working day and period etc..
Wherein, the period refers to the time interval in one day.It, then can be by one day for dividing one day by the hour
24 time intervals are divided into, each time interval is a period, such as: 0:00am-1:00am is period, 1:
00am-2:00am is period etc..
Illustratively, by taking the 8:00am-8:05am on May 1st, 2019 as an example, then the 8 of 1 day May in 2019:
The time attribute of 00am-8:05am is festivals or holidays or nonworkdays, corresponding period are 8:00am-9:00am.With 2019 5
For the 8:00am-8:05am on the moon 22, then the time attribute of the 8:00am-8:05am on May 22nd, 2019 is non-festivals or holidays
Or working day, corresponding period are 8:00am-9:00am.
For the ease of clearly describing the technical solution of the embodiment of the present application, in embodiments herein, use " the
One ", the printed words such as " second " distinguish function and the essentially identical identical entry of effect or similar item.For example, first time period
With second time period just for the sake of differentiation different period, its sequencing is not defined.Those skilled in the art
Member is understood that the printed words such as " first ", " second " are not defined quantity and execution order, and " first ", " second " etc.
Printed words do not limit certain different yet.
It should be noted that in the application, " illustrative " or " such as " etc. words make example, illustration for indicating or say
It is bright.Described herein as " illustrative " or " such as " any embodiment or design scheme be not necessarily to be construed as than it
His embodiment or design scheme more preferably or more advantage.Specifically, use " illustrative " or " such as " etc. words be intended to
Related notion is presented in specific ways.
In the application, "at least one" refers to one or more, and " multiple " refer to two or more."and/or",
The incidence relation of affiliated partner is described, indicates may exist three kinds of relationships, for example, A and/or B, can indicate: individualism A,
Exist simultaneously A and B, the case where individualism B, wherein A, B can be odd number or plural number.Character "/" typicallys represent front and back and closes
Connection object is a kind of relationship of "or".At least one of " following (a) " or its similar expression, refer to these in any group
It closes, any combination including individual event (a) or complex item (a).For example, at least one (a) in a, b or c, can indicate:
A, b, c, a-b, a-c, b-c or a-b-c, wherein a, b, c can be individually, be also possible to multiple.
Below in conjunction with Fig. 3 to Fig. 6 to a kind of traffic accident method of determining probability progress provided by the embodiments of the present application
It is specifically described.
It should be pointed out that mutually can use for reference or refer between each embodiment of the application, for example, the same or similar step
Suddenly, between embodiment of the method and Installation practice, it can mutually refer to, not limit.
As shown in figure 3, Fig. 3 shows a kind of traffic accident method of determining probability provided by the embodiments of the present application, the party
The executing subject of method can be server.Following embodiments are described so that executing subject is server as an example.
It should be noted that server can be the property server of traffic-information service provider, it can also be traffic
The virtual server of information service provider, such as Cloud Server.
As shown in figure 3, this method may include:
S201, the section attribute for obtaining target road section.
Wherein, section attribute is for indicating section feature.Section attribute may include one or more of following information:
Category of roads, number of track-lines grade, curvature, the gradient and JCT/IC attribute.
In the embodiment of the present application, category of roads may include national highway, provincial highway, county road etc..Number of track-lines grade may include list
Lane, two-way traffic, multilane etc..
A kind of possible implementation can be configured with the information in multiple sections in server, as each in multiple sections
The corresponding identification information in section (or title), the section attribute in each section etc..Wherein, target road section is appointing in multiple sections
One section.
Alternatively possible implementation, server can pass through digital navigation map, basic map or high-precision map
Deng the section attribute for obtaining target road section.
S202, information of vehicles of the target road section in first time period is obtained.
Wherein, information of vehicles includes the quantity of vehicle and the speed of vehicle.
It should be noted that server can also be according to vehicle of the target road section in first time period in the embodiment of the present application
Quantity and vehicle speed, determine the speed mean value namely mode of target road section vehicle in first time period
Pattern numerical value.
S203, mould is assessed according to the time attribute of second time period, information of vehicles, section attribute and street accidents risks
Type determines traffic accident probability of the target road section in second time period.
Wherein, street accidents risks assessment models are for determining traffic accident probability.
In the embodiment of the present application, server can time attribute according to target road section in second time period, information of vehicles
And section attribute and street accidents risks assessment models, determine traffic accident probability of the target road section in second time period.
In a kind of mode in the cards, first time period can be before second time period and adjacent with second time period
Period.
Illustratively, by taking first time period is the 8:00am-8:05am of Monday as an example, then second time period can the Monday
7:40am-8:00am.
In another mode in the cards, first time period can for before second time period and with second time period
The time attribute identical period.
Illustratively, by taking first time period is the 8:00am-8:05am of a Monday in May as an example, then second time period
It can be the 7:40am-8:00am of one or more Mondays of times such as 3 months or April.
It should be noted that the time attribute of one or more Mondays of times such as a Monday and March or April in May
It is identical.For example, being all non-festivals or holidays or being all working day.
It should be noted that if the time attribute of a Monday in May is festivals or holidays, then server can choose other moons
The time attribute of part is the Monday of festivals or holidays.
In a kind of possible embodiment, server can also obtain target road section in the Weather property of first time period and
The Weather property of two periods, such as whether rainfall or whether snowfall, can be further improved by the Weather property of target road section
The accuracy rate of traffic accident probabilistic forecasting.
Traffic accident method of determining probability provided by the embodiments of the present application, according to the section attribute of target road section, first
The time attribute and street accidents risks of the second time period after information of vehicles, first time period in period assess mould
Type determines the traffic accident probability of target road section.It is multi-party by the time attribute of target road section, information of vehicles, section attribute etc.
The accuracy rate of the traffic accident prediction of target road section can be improved in surface analysis.
As a kind of possible implementation, as shown in figure 4, server can pass through S301- in the embodiment of the present application
S303 determines street accidents risks assessment models:
S301, traffic accident information and traffic normal information are obtained.
In the embodiment of the present application, traffic accident information include the time attribute of the traffic accident in each section in M section,
The information of vehicles of multiple periods of corresponding road section before the identification information and section attribute in M section and traffic accident occur,
Traffic normal information includes: the time attribute of the traffic normal information in each section in N number of section, the mark of N number of corresponding road section
Information and section attribute;Wherein, wherein M, N are the integer greater than 0.
It should be noted that M section and N number of section can be identical section, or different sections, it can also
Think the identical section in part.
In a kind of possible implementation, server can pass through frequency modulation(PFM) (frequency modulation, FM)
Speech recognition obtains traffic accident information and traffic normal information.
Alternatively possible is in implementation, and server can be captured by day eye and obtain traffic accident information and traffic just
Normal information.
It is captured it should be noted that the quantity of the traffic accident information obtained by FM speech recognition will be lower than by day eye
The quantity of the traffic accident information of acquisition;The quantity of the traffic normal information obtained by FM speech recognition will be lower than by day eye
Capture the quantity of the traffic normal information obtained.
It should be noted that server can get multiple traffic accidents by above two mode or any mode
Information and multiple traffic normal informations.
Illustratively, the area A includes 5000 sections, and available 5000 sections to the area A of server were at 2 months 2018
To 98854 traffic accident informations and 14348831 traffic normal informations between May.
It should be noted that in the embodiment of the present application, in order to reduce the error for the traffic accident information that aforesaid way obtains,
Server can be using the section of traffic accident generation and section upstream and downstream pre-determined distance as the corresponding section of traffic accident, and incites somebody to action
Preset duration is as traffic injury time before traffic accident time of origin and generation.
Illustratively, pre-determined distance can be 200 meters, and preset duration can be 20 minutes.
S302, processing traffic accident information and traffic normal information, obtain multiple numerical value.
Wherein, multiple numerical value include that corresponding multiple first numerical value of traffic accident information and traffic normal information are corresponding more
A second value.
It should be noted that in order to improve the accuracy rate of traffic accident probabilistic forecasting, server can be sent out traffic accident
Period before death is spaced the information of vehicles for being divided into multiple sub- periods, and obtaining each sub- period at preset timed intervals.
Illustratively, for by the section A, on Monday traffic accident occurs for 8:00am, then the available section A of server is sent out
7:30am-8:00am before raw traffic accident is divided into 7:30am-7:40am, 7:40am-7:50am, 7:50am-8:00am.
It should be noted that if the quantity for the traffic accident information that server obtains and the quantity of traffic normal information are poor
Away from threshold value is greater than, then service can be with the quantity of the quantity of well-balanced adjustment traffic accident information and traffic normal information.
Illustratively, with 5000 sections in the area A 2 months 2018 between May 98854 traffic accident informations and
For 14348831 traffic normal informations, server sorts after can upsetting 14348831 traffic normal informations, takes out
Preceding 10% is taken to be used as traffic normal information.Or server can be randomly selected in 14348831 traffic normal informations
10% is used as traffic normal information, in conjunction with 98854 traffic accident informations, available traffic information stream bulk sample sheet.Clothes
Device be engaged in by customized loss function adjusting traffic information stream bulk sample sheet, to solve traffic accident information and traffic normal information
The unbalanced problem of quantity.
Wherein, customized loss function can be with are as follows:
posrIt is the ratio of traffic normal information Yu traffic information stream bulk sample sheet, negrIt is that traffic accident information and traffic are believed
The ratio of breath stream bulk sample sheet, and posr+negr=1.
Illustratively, server adjusts this when of traffic information stream bulk sample by the customized loss function, with posr=
0.3,negr=1-posrFor=0.7, then when server predicts traffic accident probability by the above method, accuracy rate 60%,
Recall rate is 30%.In practical application, needing higher recall rate, therefore can be posrIt tunes up, such as pos=0.7,
Neg=0.3, then the recall rate of prediction traffic accident probability can be improved in server.
In a kind of possible implementation, server can do following processing to the information of vehicles in each section:
S3011, the information of vehicles for obtaining each period in each section multiple periods.
Wherein, in multiple periods each period include multiple preset durations the sub- period.
Illustratively, by taking the period is 9:00am-10:00am as an example, 9:00am-10:00am can be pressed 5 points by server
Clock is that time interval is divided into 12 sub- periods, therefore server can determine the vehicle of 12 sub- periods in 9:00-10:00
Information.
If the quantity of S3012, the sub- period for lacking information of vehicles in a period are greater than threshold value, the time is deleted
The information of vehicles of section.
Illustratively, which can be 6.By taking the period is between 9:00am-10:00am as an example, if in 9:00am-
There are information of vehicles is lacked in 5 minutes of 6 or more between 10:00am, then server can delete 9:00am-10:00am it
Between information of vehicles.
If the quantity of S3013, the sub- period for lacking information of vehicles in a period are less than or equal to threshold value, should
The information of vehicles of period, which is done, linearly to be filled up.
Illustratively, which can be 6.By taking the period is between 9:00am-10:00am as an example, if in 9:00am-
There are 6 or 6 between 10:00am lacks information of vehicles in 5 minutes below, then server can by time attribute and/or
The information of vehicles in other identical sections of section attribute adds to 9:00am-10:00am.
If each of S3014, a period sub- period do not lack information of vehicles, according to preset formula and preset flat
The information of vehicles of the sliding coefficient processing period.
Wherein, preset formula are as follows: V2=V1*(1±a);V2For the information of vehicles of the sub- period of the period, V1
For the information of vehicles of adjacent sub- period sub- period, a is default smoothing factor.
Illustratively, default smoothing factor can be 0.3.With vehicle of the sub- period between 9:20am-9:25am
For average speed is 100km/h, if the average speed of the vehicle between 9:25am-9:30am is 140km/h, server
The average speed of vehicle between 9:25am-9:30am can be adjusted to 100* (1+0.3)=130km/h.
Illustratively, default smoothing factor can be 0.3.With vehicle of the sub- period between 9:20am-9:25am
For average speed is 100km/h, if the average speed of the vehicle between 9:25am-9:30am is 60km/h, server
The average speed of vehicle between 9:25am-9:30am can be processed into 100* (1-0.3)=70km/h.
In a kind of possible implementation, the embodiment of the present application provides a kind of processing method of traffic accident information, packet
Include S3015-S3016:
S3015, by traffic accident occur before and traffic accident occur when duration be spaced at preset timed intervals be divided into it is multiple
The sub- period.
Illustratively, the available B road section traffic volume accident of server occurs preceding 20 minutes and 5 minutes when traffic accident generation
This can be divided into 55 minutes sub- periods for 25 minutes by interior information of vehicles, server.
Illustratively, if traffic accident time of origin is 08:12am, then duration when traffic accident occurs can be 8:
10am-8:15am, traffic accident occur before when a length of 7:50am-8:10am.Server can be by 7:50am-8:10am and 8:
10am-8:15am is divided into 7:50am-7:55am, 7:55am-8:00am, 8:00am-8:05am, 8:05am-8:10am, 8:
10am-8:15am5 sub- periods.
S3016, the pre-determined distance that the upstream and downstream in section occurs for traffic accident is divided into multiple sub- sections.
Illustratively, by pre-determined distance be 1 kilometer for, server it is available to traffic accident occur preceding 20 minutes with
The information of vehicles of 1 kilometer of the section B upstream and downstream when traffic accident occurs in 5 minutes.Server can distinguish the section B upstream and downstream
It is divided into 50 sub- sections.
Wherein, in conjunction with S3015, then each sub- section in a section all corresponds to the information of vehicles of multiple sub- periods, because
The information of vehicles of the multiple groups of each traffic accident information in this available section.
Illustratively, by taking traffic accident occurs for the 9:02am of the section B Monday as an example, above and below the available section B of server
Information of vehicles of 1 kilometer of the trip between the information of vehicles and 9:00am-9:05am between 8:40am-9:00am.Then by above-mentioned
The above-mentioned time can be divided into 5 sub- periods: 8:40am-8:45am, 8:45am-8:50am, 8 by division methods, server:
50am-8:55am,8:55am-9:00am,9:00am-9:05am.Server is by 1 kilometer each stroke of 1 kilometer of the section B upstream and downstream
It is divided into 50 sub- sections, obtains 100 sub- sections, wherein every sub- section corresponding 5 sub- period.Therefore, server can be with
Determine corresponding 500 groups of the information of vehicles of a traffic accident in the section B.
In a kind of possible implementation, server can be according to only hot One-hot coding techniques by above-mentioned each section
Road attribute and the information of vehicles of time attribute and the corresponding multiple groups of each traffic accident information be processed into multi dimensional numerical.
Illustratively, as shown in table 1, table 1 is in the embodiment of the present application, and server is according to One-hot coding techniques to the road B
The processing result of one traffic accident information of section.
The processing result of one traffic accident information in 1 section B of table
In a kind of possible implementation, server can be according to One-hot coding techniques by each traffic normal information
The road attribute and time attribute in corresponding section, are processed into multi dimensional numerical.
Illustratively, as shown in table 2, table 2 is in the embodiment of the present application, and server is according to One-hot coding techniques to the road B
The processing result of one traffic normal information of section.
The processing result of one traffic normal information in 2 section B of table
Field attribute | Dimension |
Category of roads | 11 |
Number of track-lines grade | 3 |
JCT/IC attribute | 3 |
Curvature | 1 |
The gradient | 1 |
Festivals or holidays | 2 |
Working day | 2 |
Period | 24 |
It amounts to | 47 |
It should be noted that number of track-lines grade includes 3 grades (corresponding to be identified as 1,2,3) in table 1, table 2, then take
Business device can be handled number of track-lines grade according to One-hot coding techniques as (0,0,1), (0,1,0), (1,0,0).Server
Can handle in a manner described the category of roads in the section B, JCT/IC attribute, festivals or holidays, working day and period are processed into it is more
Dimension value.
It should be noted that in the embodiment of the present application, server can pass through digital navigation map, basic map or high-precision
Spend the gradient and curvature that map etc. obtains multiple sections.
It should be noted that some sections correspond to multiple gradients, if in multiple gradients the gradient of maximum absolute value be greater than or
Equal to first threshold, the gradient in the section can be processed into 1 by server.The gradient in some sections is less than first threshold, service
The processing of the gradient in section can be 0 by device.Some sections correspond to multiple curvature, if the curvature of maximum absolute value is big in multiple curvature
In or equal to second threshold, the processing of the curvature in the section can be 1 by server.The curvature in some sections is less than second threshold,
It is 0 that server can be handled the curvature in section respectively.
In the embodiment of the present application, server can be corresponding more according to the available traffic accident information of above-mentioned processing method
A first numerical value and the corresponding multiple second values of traffic normal information.
Illustratively, in conjunction with table 1, table 2 and above-mentioned processing method, server can determine that traffic accident information is corresponding more
A first numerical value is X={ x1,x2,...,x547, the corresponding multiple second values of traffic normal information are Y={ y1,y2,...,
y47}。
S303, street accidents risks assessment models are determined according to multiple numerical value and dynamic circulation neural network.
In a kind of possible implementation, server can be by corresponding multiple first numbers of the traffic accident information in S301
Value and the corresponding multiple second values of traffic normal information input dynamic circulation neural network, determine that street accidents risks assess mould
Type.
It should be noted that dynamic refer to feedback neural network, Recognition with Recurrent Neural Network by feedback can will before
The numerical value at one moment retains, it is made to be added to the calculating of subsequent time numerical value, make Recognition with Recurrent Neural Network not only and have dynamic and
And the system information retained is also more complete.
It should be noted that dynamic circulation neural network includes input layer, hidden layer and output layer.Wherein, input layer can
Corresponding multiple first numerical value of traffic accident information and traffic normal information corresponding multiple second to receive each moment
Numerical value.Multiple numerical value is propagated forward in the neuron of hidden layer by dynamic circulation network.Hidden layer gradually propagated forward
To output layer, which can finally be presented the output result of dynamic circulation neural network.
Wherein, output result includes traffic accident probability, and accident whether can occur (0 indicates to occur accident, and 1 indicates
Accident does not occur).
In the embodiment of the present application, the structure of dynamic circulation neural network are as follows: LSTM connection GRU.Wherein, GRU is first layer
Structure, LSTM are second layer structures.
Illustratively, Fig. 5 shows one of the embodiment of the present application LSTM structural schematic diagram.
It should be noted that LSTM includes multiple cell A (only showing 3 in Fig. 5), the structure of each cell A is identical.Its
In, each cell A of LSTM includes 12 hiding nodes.
It should be noted that as shown in figure 5, each cell A includes forgeing door, input gate, cell state to update and export
Door:
Wherein, forget door:
ft=σ (Wf·[ht-1,xt]+bf)
Input gate:
Cell state updates:
Out gate:
Wherein, sigma function is sigmoid function, and sigmoid function is most direct excitation function, be can be applied to various
Door, sigmoid function can be by numerical value processing between 0~1.Tanh function is used for the output of dynamic circulation neural network,
Tanh function can be by numerical value processing between -1~1.Wf、bf、Wi、bi、Wc、bc、Wo、boThe parameter of LSTM, t be greater than or
Integer equal to 1.
It should be noted that the input of each cell of LSTM are as follows: xnAnd the output of a upper cell A, wherein n is big
In 0 integer.
Illustratively, the input in Fig. 5, in second cell of LSTM are as follows: xt、ht-1.Wherein, ht-1It is in Fig. 5 first
The output of a cell A.
It should be noted that inputting h in LSTM0Corresponding initial numerical value can be preset value.
Wherein, the input x of each cell of LSTMnA corresponding numerical value.
Illustratively, the input x of LSTM3X in multiple numerical value of corresponding traffic accident information3。
Illustratively, Fig. 6 shows one of the embodiment of the present application GRU structural schematic diagram.
It should be noted that GRU includes multiple cell A ' (only showing 3 in Fig. 6), the structure of each cell A ' is identical.
Wherein, each cell of GRU includes 6 hiding nodes.
Here 6 concealed nodes are provided that.
It should be noted that as shown in fig. 6, each cell A ' includes resetting door, updates door in GRU.
It should be noted that in Fig. 6, rt=σ (Wr·[ht-1·xt]), zt=σ (Wz·[ht-1·xt]),
Wherein, Wr、Wz、It is the parameter of GRU.Consistent in sigma function and tanh function and LSTM, details are not described herein again.
It should be noted that the input of each cell A ' of GRU are as follows: upper one of the output of cell A and GRU is thin in LSTM
The output of born of the same parents A '.
Illustratively, the input x in Fig. 6tOutput h in corresponding diagram 5t。
It should be noted that dynamic circulation neural network exports result are as follows: yt=σ (Wq·ht).Wherein, WqIt is parameter.
It should be noted that LSTM is from GRU the difference is that the function of cell interior is different.The phase of LSTM and GRU
It is the reuse repeatedly of cell with place, such as cell A is reused repeatedly in LSTM layers and cell A ' is weighed repeatedly in GRU layers
With.LSTM and both structures of GRU can remember the pervious input of sequence.Therefore, when be related to it is continuous, with context phase
When the task of pass, dynamic circulation neural network is with greater advantage than other neural networks.
In the embodiment of the present application, the framework of LSTM connection GRU can be effectively by each of dynamic circulation neural network
Cells switch improves gradient disappearance problem while improving the accuracy rate of traffic accident prediction at a simulation computer.
In the embodiment of the present application, server can be by the corresponding multiple first numerical value X={ x of traffic accident information1,
x2,...,x547And the corresponding multiple second value Y={ y of traffic normal information1,y2,...,y47, input dynamic circulation mind
Through network, parameter W is obtainedf、bf、Wi、bi、Wc、bc、Wo、bo、Wr、Wz、And corresponding output result yt=σ (Wq·ht)。
Further, server can determine that street accidents risks are assessed according to parameter obtained above and dynamic circulation neural network
Model.
In a kind of possible implementation, server can be by 5 folding cross validations, to street accidents risks assessment models
Power of test be averaging, obtain the test effect of street accidents risks assessment models.
Illustratively, server is by corresponding multiple first numerical value of above-mentioned traffic accident information and traffic normal information
After corresponding multiple second values mixing, 5 groups of numerical value are divided into, such as first group of numerical value, second group of numerical value, third group numerical value, the
Four groups of numerical value, the 5th group of numerical value.Using first group of numerical value as sample to be tested, other four groups of numerical value are as training sample.It takes first
Device be engaged in for other four groups of numerical value input dynamic circulation neural networks, obtains one group of parameter Wf、bf、Wi、bi、Wt、bt、Wo、bo, in turn
Server can determine a street accidents risks assessment models according to dynamic circulation neural network and parameter.Then server will
First group of numerical value inputs the street accidents risks assessment models, general for verifying the street accidents risks assessment models traffic accident
The accuracy rate of rate prediction.
Likewise, in this mode, the available 5 group model parameter of server and 5 street accidents risks assess mould
Type.
In the embodiment of the present application, server is repeatedly tested according to the above method, available multiple groups as shown in table 3
Test result.
3 test result of table
Numerical value | Accuracy rate | Recall rate | F value |
1 | 65.78% | 74.11% | 70% |
2 | 64.52% | 75.22% | 69.46% |
3 | 67.33% | 66.63% | 66.98% |
4 | 67.36% | 67.53% | 67.45% |
5 | 67.95% | 65.11% | 66.5% |
By table 3 it is found that server not only may be used according to traffic accident method of determining probability provided by the embodiments of the present application
To improve the accuracy rate of traffic accident prediction, higher recall rate and comprehensive evaluation index F value can also be obtained.
It should be noted that server determines the accuracy rate, recall rate, F value of traffic accident prediction in the embodiment of the present application
Reference standard it is as shown in table 4.
4 reference standard of table
Wherein, in table 4, TP indicates that prediction result is traffic accident to occur, and traffic accident also occurs for reality.FP indicates pre-
Surveying result is traffic accident to occur, but traffic accident does not occur for reality.FP indicates prediction result as traffic accident does not occur, but real
Traffic accident has occurred in border.TN indicates that prediction result is traffic accident not to occur, and traffic accident does not also occur for reality.
Accuracy rate calculation formula is as follows:
Recall rate calculation formula is as follows:
Comprehensive evaluation index F value calculates as follows:
As a kind of possible implementation, as shown in figure 4, S203 can be realized by S304 in the embodiment of the present application.
S304, by the corresponding numerical value of the time attribute of second time period, the corresponding numerical value of information of vehicles and section attribute
Corresponding numerical value inputs street accidents risks assessment models, determines traffic accident probability of the target road section in second time period.
It should be noted that in the embodiment of the present application, server can be according to One-hot coding mode by second time period
Information of vehicles in second time period of time attribute, target road section, target road section section attribute be processed into target road section pair
The multiple numerical value answered.
In the embodiment of the present application, server comments the street accidents risks of the corresponding multiple numerical value input S303 of target road section
Estimate model, can determine traffic accident probability of the target road section in second time period.
In an alternative embodiment, as shown in figure 4, method provided by the embodiments of the present application can also include:
S305, traffic accident probability are greater than or equal to preset threshold, determine and traffic accident occurs.
S306, traffic accident probability are less than preset threshold, and traffic accident does not occur for determination.
Illustratively, in the embodiment of the present application, preset threshold can be 0.7.
As a kind of possible implementation, as shown in figure 4, method provided by the embodiments of the present application can also include:
S307, determine occur traffic accident in the case where, send indicating risk information.
The application asks in embodiment, when determine target road section second time period occur traffic accident in the case where, service
Device can send indicating risk information to terminal.The indicating risk information is for prompting target road section that may send out in second time period
Raw traffic accident.
It should be noted that terminal can be the equipment for needing to receive traffic accident probability.For example, the terminal can be use
Family handheld communication devices, for example, mobile phone, tablet computer, mobile unit, wearable device etc..The embodiment of the present application does not make this
Any restriction.For example, handheld device can be smart phone.Mobile unit can be onboard navigation system.Wearable device can
To be Intelligent bracelet.Computer can be personal digital assistant (personal digital assistant, PDA) computer, put down
Template computer and laptop computer (laptop computer).
The embodiment of the present application can carry out functional module according to determining device of the above method example to traffic accident probability
Perhaps the division of functional unit be for example, each functional module of each function division or functional unit can be corresponded to, can also will
Two or more functions are integrated in a processing module.Above-mentioned integrated module both can take the form of hardware reality
It is existing, it can also be realized in the form of software function module or functional unit.Wherein, in the embodiment of the present application to module or
The division of unit is schematically that only a kind of logical function partition, there may be another division manner in actual implementation.
The embodiment of the present application provides a kind of determining device of traffic accident probability, which can be server, can also
The chip applied to server is thought, as shown in fig. 7, the device 100 may include communication unit 101 and determination unit 102.
Communication unit 101, for obtaining the section attribute of target road section;Section attribute include one in following information or
It is multiple: category of roads, number of track-lines grade, curvature, the gradient and JCT/IC attribute;
Communication unit 101 is also used to obtain information of vehicles of the target road section in first time period;Wherein, information of vehicles
The speed of quantity and vehicle including vehicle;
Determination unit 102, for according to the time attribute of second time period, information of vehicles, section attribute and traffic accident
Risk evaluation model determines traffic accident probability of the target road section in second time period;Wherein, street accidents risks assess mould
For type for determining traffic accident probability, first time period is after second time period and the period adjacent with second time period.
Optionally, communication unit 101 are also used to obtain traffic accident information and traffic normal information.
Determination unit 102, is also used to: processing traffic accident information and traffic normal information obtain multiple numerical value;Wherein,
Multiple numerical value include corresponding multiple first numerical value of traffic accident information and the corresponding multiple second values of traffic normal information;Root
Street accidents risks assessment models are determined according to multiple numerical value and dynamic circulation nerual network technique.
Optionally, determination unit 102 are specifically used for: by the corresponding numerical value of the time attribute of second time period, information of vehicles
Corresponding numerical value and the corresponding numerical value of section attribute input street accidents risks assessment models, determine target road section second
Traffic accident probability in period.
Optionally, traffic accident information includes the time attribute of the traffic accident in each section in M section, M section
Identification information and section attribute and traffic accident occur before corresponding road section multiple periods information of vehicles, traffic is just
Normal information includes the identification information and section attribute of the normal time attribute of traffic in each section in N number of section, N number of section;
Wherein, M, N are the integer greater than 0.
Optionally, traffic accident probability is greater than or equal to preset threshold, and determination unit 102 is also used to determine generation traffic
Accident.
Optionally, in the case where determining generation traffic accident, communication unit 101 is also used to send indicating risk information.
Optionally, traffic accident probability is less than preset threshold, and determination unit 102 is also used to determination and traffic accident does not occur.
Another possible structure that Fig. 8 shows the determining device of involved traffic accident probability in above-described embodiment is shown
It is intended to.When the device is server, which includes: one or more processors 161 and communication interface 162.Processor 161
Control management is carried out for the movement to device, for example, executing the step of above-mentioned determination unit 102 executes, and/or for executing
Other processes of techniques described herein.
In the concrete realization, as one embodiment, processor 161 may include one or more CPU, such as in Fig. 8
CPU0 and CPU1.
In the concrete realization, as one embodiment, communication equipment may include multiple processors, such as the place in Fig. 8
Manage device 161.Each of these processors can be monokaryon (single-CPU) processor, be also possible to a multicore
(multi-CPU) processor.Here processor can refer to one or more equipment, circuit, and/or for handling data (example
Such as computer program instructions) processing core.
Optionally, which can also include memory 163 and communication line 164, and memory 163 is for storage device
Program code and data.
Fig. 9 is the structural schematic diagram of chip 170 provided by the embodiments of the present application.Chip 170 includes one or more
(including two) processor 1710 and communication interface 1730.
Optionally, which further includes memory 1740, and memory 1740 may include read-only memory and deposit at random
Access to memory, and operational order and data are provided to processor 1710.The a part of of memory 1740 can also include non-volatile
Property random access memory (non-volatile random access memory, NVRAM).
In some embodiments, memory 1740 stores following element, execution module perhaps data structure or
Their subset or their superset.
In the embodiment of the present application, by calling the operational order of the storage of memory 1740, (operational order is storable in
In operating system), execute corresponding operation.
Wherein, above-mentioned processor 1710 may be implemented or execute various exemplary in conjunction with described in present disclosure
Logic block, unit and circuit.The processor can be central processing unit, general processor, and digital signal processor is dedicated
Integrated circuit, field programmable gate array or other programmable logic device, transistor logic, hardware component or its
Any combination.It, which may be implemented or executes, combines various illustrative logic blocks described in present disclosure, unit
And circuit.The processor is also possible to realize the combination of computing function, such as combines comprising one or more microprocessors,
DSP and the combination of microprocessor etc..
Memory 1740 may include volatile memory, such as random access memory;The memory also may include
Nonvolatile memory, such as read-only memory, flash memory, hard disk or solid state hard disk;The memory can also include upper
State the combination of the memory of type.
Bus 1720 can be expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..Bus 1720 can be divided into address bus, data/address bus, control bus etc..For convenient for
It indicates, is only indicated with a line in Fig. 9, it is not intended that an only bus or a type of bus.
Through the above description of the embodiments, it is apparent to those skilled in the art that, for description
It is convenienct and succinct, only with the division progress of above-mentioned each functional unit for example, in practical application, can according to need and will be upper
It states function distribution to be completed by different functional units, i.e., the internal structure of device is divided into different functional units, to complete
All or part of function described above.The specific work process of the system, apparatus, and unit of foregoing description, before can referring to
The corresponding process in embodiment of the method is stated, details are not described herein.
The embodiment of the present application also provides a kind of computer readable storage medium, and finger is stored in computer readable storage medium
It enables, when computer executes the instruction, which executes each step in method flow shown in above method embodiment.
Wherein, computer readable storage medium, such as electricity, magnetic, optical, electromagnetic, infrared ray can be but not limited to or partly led
System, device or the device of body, or any above combination.The more specific example of computer readable storage medium is (non-poor
The list of act) it include: the electrical connection with one or more conducting wires, portable computer diskette, hard disk.Random access memory
(Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM), erasable type may be programmed read-only
It is memory (Erasable Programmable Read Only Memory, EPROM), register, hard disk, optical fiber, portable
Compact disc read-only memory (Compact Disc Read-Only Memory, CD-ROM), light storage device, magnetic memory
Part or above-mentioned people are in appropriate combination or the computer readable storage medium of any other form of this field numerical value.
A kind of illustrative storage medium is coupled to processor, to enable a processor to from the read information, and can be to
Information is written in the storage medium.Certainly, storage medium is also possible to the component part of processor.Pocessor and storage media can be with
In application-specific IC (Application Specific Integrated Circuit, ASIC).In the application
In embodiment, computer readable storage medium can be any tangible medium for including or store program, which can be referred to
Enable execution system, device or device use or in connection.
The embodiment of the present invention provides a kind of computer program product, and the computer program product refers to comprising at least one
It enables, when at least one instruction is run on computers, so that computer executes the traffic accident probability as described in Fig. 3, Fig. 4
Determination method.
By the determining device of traffic accident probability in an embodiment of the present invention, computer readable storage medium, calculating
Machine program product can be applied to the above method, therefore, can be obtained technical effect see also above method embodiment,
Details are not described herein for the embodiment of the present invention.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method, it can be with
It realizes by another way.For example, apparatus embodiments described above are merely indicative, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of equipment or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application should be based on the protection scope of the described claims.
Claims (16)
1. a kind of traffic accident method of determining probability characterized by comprising
Obtain the section attribute of target road section;The section attribute includes one or more of following information: category of roads, vehicle
Road number grade, curvature, the gradient and highway crosspoint/highway conversion entrance JCT/IC attribute;
Obtain information of vehicles of the target road section in first time period;Wherein, the information of vehicles includes the quantity of vehicle
And the speed of the vehicle;
Mould is assessed according to the time attribute of second time period, the information of vehicles, the section attribute and street accidents risks
Type determines traffic accident probability of the target road section in the second time period;Wherein, the street accidents risks assessment
Model for determining traffic accident probability, before the first time period is the second time period and with the second time period
The adjacent period.
2. the method according to claim 1, wherein the method also includes:
Obtain traffic accident information and traffic normal information;
The traffic accident information and the traffic normal information are handled, multiple numerical value are obtained;Wherein, the multiple numerical value includes
Corresponding multiple first numerical value of traffic accident information and the corresponding multiple second values of the traffic normal information;
The street accidents risks assessment models are determined according to the multiple numerical value and dynamic circulation neural network.
3. according to the method described in claim 2, it is characterized in that, the time attribute according to second time period, the vehicle
Information, the section attribute and street accidents risks assessment models, determine the target road section in the second time period
Traffic accident probability, comprising:
By the corresponding numerical value of the time attribute of the second time period, the corresponding numerical value of the information of vehicles and the section
The corresponding numerical value of attribute inputs the street accidents risks assessment models, determines the target road section in the second time period
The traffic accident probability.
4. according to the described in any item methods of claim 2-3, which is characterized in that the traffic accident information includes M section
In the time attribute of traffic accident in each section, the identification information in the M section, section attribute and traffic accident hair
The information of vehicles of multiple periods of corresponding road section before death, the traffic normal information include the friendship in each section in N number of section
Lead to the time attribute of normal information, the identification information and section attribute in N number of section;Wherein, M, N are the integer greater than 0.
5. method according to claim 1-3, which is characterized in that the method also includes:
The traffic accident probability is greater than or equal to preset threshold, determines and traffic accident occurs.
6. according to the method described in claim 5, it is characterized in that, the method also includes:
In the case where determining generation traffic accident, indicating risk information is sent.
7. method according to claim 1-3, which is characterized in that the method also includes:
The traffic accident probability is less than preset threshold, and traffic accident does not occur for determination.
8. a kind of determining device of traffic accident probability characterized by comprising
Communication unit, for obtaining the section attribute of target road section;The section attribute includes one or more in following information
It is a: category of roads, number of track-lines grade, curvature, the gradient and JCT/IC attribute;
The communication unit is also used to obtain information of vehicles of the target road section in first time period;Wherein, the vehicle
Information includes the quantity of vehicle and the speed of the vehicle;
Determination unit, for according to the time attribute of second time period, the information of vehicles, the section attribute and traffic thing
Therefore risk evaluation model, determine traffic accident probability of the target road section in the second time period;Wherein, the traffic
Accident risk assessment models for determining traffic accident probability, the first time period be the second time period before and with institute
State the second time period adjacent period.
9. device according to claim 8, which is characterized in that
The communication unit, is also used to: obtaining traffic accident information and traffic normal information;
The determination unit, is also used to: handling the traffic accident information and the traffic normal information, obtains multiple numerical value;
Wherein, the multiple numerical value includes that corresponding multiple first numerical value of the traffic accident information and the traffic normal information are corresponding
Multiple second values;
The determination unit, is also used to: determining the traffic accident wind according to the multiple numerical value and dynamic circulation neural network
Dangerous assessment models.
10. device according to claim 9, which is characterized in that the determination unit is specifically used for:
By the corresponding numerical value of the time attribute of the second time period, the corresponding numerical value of the information of vehicles and the section
The corresponding numerical value of attribute inputs the street accidents risks assessment models, determines the target road section in the second time period
The traffic accident probability.
11. according to the described in any item devices of claim 9-10, which is characterized in that the traffic accident information includes M road
The time attribute of traffic accident, the identification information in the M section and the section attribute in each section and traffic accident in section
The information of vehicles of multiple periods of corresponding road section before occurring, the traffic normal information includes each section in N number of section
The time attribute of traffic normal information, N number of section identification information and section attribute;Wherein, M, N are the integer greater than 0.
12. according to the described in any item devices of claim 8-10, which is characterized in that
The traffic accident probability is greater than or equal to preset threshold, and the determination unit is also used to: determining and traffic accident occurs.
13. according to the described in any item devices of claim 8-10, which is characterized in that
In the case where determining generation traffic accident, the communication unit is also used to: sending indicating risk information.
14. according to the described in any item devices of claim 8-10, which is characterized in that
The traffic accident probability is less than preset threshold, and the determination unit is also used to determination and traffic accident does not occur.
15. a kind of readable storage medium storing program for executing, which is characterized in that instruction is stored in the readable storage medium storing program for executing, when described instruction quilt
When execution, the method as described in any one of claims 1 to 7 is realized.
16. a kind of chip, which is characterized in that the chip includes at least one processor and communication interface, the communication interface
It is coupled at least one described processor, at least one described processor is for running computer program or instruction, to realize power
Benefit require any one of 1 to 7 described in method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910458355.8A CN110264711A (en) | 2019-05-29 | 2019-05-29 | A kind of traffic accident method of determining probability and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910458355.8A CN110264711A (en) | 2019-05-29 | 2019-05-29 | A kind of traffic accident method of determining probability and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110264711A true CN110264711A (en) | 2019-09-20 |
Family
ID=67915844
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910458355.8A Pending CN110264711A (en) | 2019-05-29 | 2019-05-29 | A kind of traffic accident method of determining probability and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110264711A (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110852563A (en) * | 2019-10-09 | 2020-02-28 | 天津大学 | Deep learning attention mechanism-based social risk multi-objective optimization generation method |
CN111009122A (en) * | 2019-11-13 | 2020-04-14 | 青岛国信城市信息科技有限公司 | Tunnel traffic operation risk assessment method |
CN111062240A (en) * | 2019-10-16 | 2020-04-24 | 中国平安财产保险股份有限公司 | Method and device for monitoring automobile driving safety, computer equipment and storage medium |
CN111859291A (en) * | 2020-06-23 | 2020-10-30 | 北京百度网讯科技有限公司 | Traffic accident recognition method, device, equipment and computer storage medium |
CN112101670A (en) * | 2020-09-21 | 2020-12-18 | 北京嘀嘀无限科技发展有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN112530163A (en) * | 2020-11-23 | 2021-03-19 | 腾讯科技(深圳)有限公司 | Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium |
CN112562337A (en) * | 2020-12-10 | 2021-03-26 | 之江实验室 | Expressway real-time traffic accident risk assessment method based on deep learning |
CN112634614A (en) * | 2020-12-16 | 2021-04-09 | 安徽百诚慧通科技有限公司 | Long downhill traffic incident real-time detection method, device and storage medium |
CN113379187A (en) * | 2021-04-29 | 2021-09-10 | 武汉理工大学 | Traffic meteorological disaster assessment method and device and computer readable storage medium |
CN113781777A (en) * | 2021-08-31 | 2021-12-10 | 同济大学 | Traffic accident number prediction method and system based on multivariate time series model |
CN114627643A (en) * | 2022-02-28 | 2022-06-14 | 青岛海信网络科技股份有限公司 | Expressway accident risk prediction method, device, equipment and medium |
CN114863680A (en) * | 2022-04-27 | 2022-08-05 | 腾讯科技(深圳)有限公司 | Prediction processing method, prediction processing device, computer equipment and storage medium |
CN115100867A (en) * | 2022-07-27 | 2022-09-23 | 武汉微晶石科技股份有限公司 | Urban intelligent traffic simulation method based on digital twins |
CN115171372A (en) * | 2022-06-20 | 2022-10-11 | 青岛海信网络科技股份有限公司 | Traffic anomaly detection method, equipment and device |
CN115331449A (en) * | 2022-10-17 | 2022-11-11 | 四川省公路规划勘察设计研究院有限公司 | Method and device for identifying accident prone area of long and large continuous longitudinal slope section and electronic equipment |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104751642A (en) * | 2015-03-11 | 2015-07-01 | 同济大学 | Real-time estimating method for high-grade road traffic flow running risks |
CN105608902A (en) * | 2016-03-28 | 2016-05-25 | 辽宁省交通科学研究院 | Expressway black spot identification system and method |
CN106355883A (en) * | 2016-10-20 | 2017-01-25 | 同济大学 | Risk evaluation model-based traffic accident happening probability acquiring method and system |
CN106504534A (en) * | 2016-11-28 | 2017-03-15 | 北京世纪高通科技有限公司 | A kind of method of predicted link road conditions, device and user equipment |
CN106507315A (en) * | 2016-11-24 | 2017-03-15 | 西安交通大学 | A kind of urban traffic accident Forecasting Methodology and system based on network social intercourse media data |
CN106781458A (en) * | 2016-11-30 | 2017-05-31 | 成都通甲优博科技有限责任公司 | A kind of traffic accident monitoring method and system |
CN107845262A (en) * | 2017-11-09 | 2018-03-27 | 山东浪潮云服务信息科技有限公司 | A kind of Predictive Methods of Road Accidents and device |
JP2018067116A (en) * | 2016-10-19 | 2018-04-26 | 株式会社東芝 | Accident prediction system and accident prediction method |
CN108198415A (en) * | 2017-12-28 | 2018-06-22 | 同济大学 | A kind of city expressway accident forecast method based on deep learning |
CN108417033A (en) * | 2018-03-23 | 2018-08-17 | 四川高路交通信息工程有限公司 | Expressway traffic accident analysis prediction technique based on multi-dimensional factors |
CN108877213A (en) * | 2018-05-11 | 2018-11-23 | 上海赢科信息技术有限公司 | The recognition methods and system of traffic accident |
CN109147325A (en) * | 2018-09-04 | 2019-01-04 | 广州视源电子科技股份有限公司 | road condition predicting method and device, storage medium, processor |
CN109697852A (en) * | 2019-01-23 | 2019-04-30 | 吉林大学 | Urban road congestion degree prediction technique based on timing traffic events |
-
2019
- 2019-05-29 CN CN201910458355.8A patent/CN110264711A/en active Pending
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104751642A (en) * | 2015-03-11 | 2015-07-01 | 同济大学 | Real-time estimating method for high-grade road traffic flow running risks |
CN105608902A (en) * | 2016-03-28 | 2016-05-25 | 辽宁省交通科学研究院 | Expressway black spot identification system and method |
JP2018067116A (en) * | 2016-10-19 | 2018-04-26 | 株式会社東芝 | Accident prediction system and accident prediction method |
CN106355883A (en) * | 2016-10-20 | 2017-01-25 | 同济大学 | Risk evaluation model-based traffic accident happening probability acquiring method and system |
CN106507315A (en) * | 2016-11-24 | 2017-03-15 | 西安交通大学 | A kind of urban traffic accident Forecasting Methodology and system based on network social intercourse media data |
CN106504534A (en) * | 2016-11-28 | 2017-03-15 | 北京世纪高通科技有限公司 | A kind of method of predicted link road conditions, device and user equipment |
CN106781458A (en) * | 2016-11-30 | 2017-05-31 | 成都通甲优博科技有限责任公司 | A kind of traffic accident monitoring method and system |
CN107845262A (en) * | 2017-11-09 | 2018-03-27 | 山东浪潮云服务信息科技有限公司 | A kind of Predictive Methods of Road Accidents and device |
CN108198415A (en) * | 2017-12-28 | 2018-06-22 | 同济大学 | A kind of city expressway accident forecast method based on deep learning |
CN108417033A (en) * | 2018-03-23 | 2018-08-17 | 四川高路交通信息工程有限公司 | Expressway traffic accident analysis prediction technique based on multi-dimensional factors |
CN108877213A (en) * | 2018-05-11 | 2018-11-23 | 上海赢科信息技术有限公司 | The recognition methods and system of traffic accident |
CN109147325A (en) * | 2018-09-04 | 2019-01-04 | 广州视源电子科技股份有限公司 | road condition predicting method and device, storage medium, processor |
CN109697852A (en) * | 2019-01-23 | 2019-04-30 | 吉林大学 | Urban road congestion degree prediction technique based on timing traffic events |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110852563A (en) * | 2019-10-09 | 2020-02-28 | 天津大学 | Deep learning attention mechanism-based social risk multi-objective optimization generation method |
CN111062240A (en) * | 2019-10-16 | 2020-04-24 | 中国平安财产保险股份有限公司 | Method and device for monitoring automobile driving safety, computer equipment and storage medium |
CN111009122A (en) * | 2019-11-13 | 2020-04-14 | 青岛国信城市信息科技有限公司 | Tunnel traffic operation risk assessment method |
CN111859291B (en) * | 2020-06-23 | 2022-02-25 | 北京百度网讯科技有限公司 | Traffic accident recognition method, device, equipment and computer storage medium |
CN111859291A (en) * | 2020-06-23 | 2020-10-30 | 北京百度网讯科技有限公司 | Traffic accident recognition method, device, equipment and computer storage medium |
US11328600B2 (en) | 2020-06-23 | 2022-05-10 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for identifying traffic accident, device and computer storage medium |
CN112101670A (en) * | 2020-09-21 | 2020-12-18 | 北京嘀嘀无限科技发展有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN112530163A (en) * | 2020-11-23 | 2021-03-19 | 腾讯科技(深圳)有限公司 | Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium |
CN112530163B (en) * | 2020-11-23 | 2022-03-01 | 腾讯科技(深圳)有限公司 | Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium |
CN112562337A (en) * | 2020-12-10 | 2021-03-26 | 之江实验室 | Expressway real-time traffic accident risk assessment method based on deep learning |
CN112634614B (en) * | 2020-12-16 | 2022-05-06 | 安徽百诚慧通科技股份有限公司 | Long downhill traffic incident real-time detection method, device and storage medium |
CN112634614A (en) * | 2020-12-16 | 2021-04-09 | 安徽百诚慧通科技有限公司 | Long downhill traffic incident real-time detection method, device and storage medium |
CN113379187A (en) * | 2021-04-29 | 2021-09-10 | 武汉理工大学 | Traffic meteorological disaster assessment method and device and computer readable storage medium |
CN113781777A (en) * | 2021-08-31 | 2021-12-10 | 同济大学 | Traffic accident number prediction method and system based on multivariate time series model |
CN114627643A (en) * | 2022-02-28 | 2022-06-14 | 青岛海信网络科技股份有限公司 | Expressway accident risk prediction method, device, equipment and medium |
CN114627643B (en) * | 2022-02-28 | 2023-07-14 | 青岛海信网络科技股份有限公司 | Highway accident risk prediction method, device, equipment and medium |
CN114863680B (en) * | 2022-04-27 | 2023-04-18 | 腾讯科技(深圳)有限公司 | Prediction processing method, prediction processing device, computer equipment and storage medium |
CN114863680A (en) * | 2022-04-27 | 2022-08-05 | 腾讯科技(深圳)有限公司 | Prediction processing method, prediction processing device, computer equipment and storage medium |
CN115171372A (en) * | 2022-06-20 | 2022-10-11 | 青岛海信网络科技股份有限公司 | Traffic anomaly detection method, equipment and device |
CN115171372B (en) * | 2022-06-20 | 2023-10-24 | 青岛海信网络科技股份有限公司 | Traffic abnormality detection method, equipment and device |
CN115100867B (en) * | 2022-07-27 | 2022-11-29 | 武汉微晶石科技股份有限公司 | Urban intelligent traffic simulation method based on digital twins |
CN115100867A (en) * | 2022-07-27 | 2022-09-23 | 武汉微晶石科技股份有限公司 | Urban intelligent traffic simulation method based on digital twins |
CN115331449B (en) * | 2022-10-17 | 2023-02-07 | 四川省公路规划勘察设计研究院有限公司 | Method and device for identifying accident-prone area of long and large continuous longitudinal slope section and electronic equipment |
CN115331449A (en) * | 2022-10-17 | 2022-11-11 | 四川省公路规划勘察设计研究院有限公司 | Method and device for identifying accident prone area of long and large continuous longitudinal slope section and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110264711A (en) | A kind of traffic accident method of determining probability and device | |
Bao et al. | A spatiotemporal deep learning approach for citywide short-term crash risk prediction with multi-source data | |
CN112700072B (en) | Traffic condition prediction method, electronic device, and storage medium | |
Chen et al. | Predicting station level demand in a bike‐sharing system using recurrent neural networks | |
CN102436751B (en) | Short-time forecasting method for traffic flow based on urban macroscopic road network model | |
CN106504534B (en) | A kind of method, apparatus and user equipment for predicting road conditions | |
CN109840660A (en) | A kind of vehicular characteristics data processing method and vehicle risk prediction model training method | |
Fukuda et al. | Short‐term prediction of traffic flow under incident conditions using graph convolutional recurrent neural network and traffic simulation | |
Mo et al. | Speed profile estimation using license plate recognition data | |
CN110390349A (en) | Bus passenger flow volume based on XGBoost model predicts modeling method | |
CN105513337A (en) | Passenger flow volume prediction method and device | |
CN110232820A (en) | A kind of method for building up and device of road condition predicting model | |
Du et al. | Predicting freeway work zone delays and costs with a hybrid machine-learning model | |
CN111160622A (en) | Scenic spot passenger flow prediction method and device based on hybrid neural network model | |
Lu et al. | Short-term demand forecasting for online car-hailing using ConvLSTM networks | |
Du et al. | Traffic events oriented dynamic traffic assignment model for expressway network: a network flow approach | |
CN104599002A (en) | Order value predicting method and equipment | |
Liu et al. | Modelling the energy consumption of electric vehicles under uncertain and small data conditions | |
CN108985518A (en) | A kind of traffic congestion charging policy optimization method based on VISSIM emulation | |
Girijan et al. | Dynamic thresholds identification for green extension and red truncation strategies for bus priority | |
Jin et al. | Determining the breakpoints of fundamental diagrams | |
Zhang et al. | Road section traffic flow prediction method based on the traffic factor state network | |
Arasan et al. | Microsimulation study of vehicular interactions in heterogeneous traffic flow on intercity roads | |
Zheng et al. | A deep learning–based approach for moving vehicle counting and short-term traffic prediction from video images | |
Wu et al. | A novel dynamically adjusted regressor chain for taxi demand prediction |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190920 |