CN107862868B - A method of track of vehicle prediction is carried out based on big data - Google Patents
A method of track of vehicle prediction is carried out based on big data Download PDFInfo
- Publication number
- CN107862868B CN107862868B CN201711099729.9A CN201711099729A CN107862868B CN 107862868 B CN107862868 B CN 107862868B CN 201711099729 A CN201711099729 A CN 201711099729A CN 107862868 B CN107862868 B CN 107862868B
- Authority
- CN
- China
- Prior art keywords
- site
- data
- vehicle
- vehicle data
- history
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
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
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/258—Data format conversion from or to a database
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of methods for carrying out track of vehicle prediction based on big data, comprise determining that the driving path of target vehicle, obtain history vehicle data, it is extracted from history vehicle data: traveling basic data and correlation data, by basic data and correlation data, according to acquisition time from closely to being far ranked up, calculating target vehicle passes through first object site F1The first probability P1X, and from current site C to first object site F1First time T used1X.Through the invention, reduce the skill requirement to staff, and can while prediction locus predicted time, while structuring and unstructured data being uniformly processed, improved work efficiency.
Description
Technical field
The present invention relates to tracks of vehicle to predict field, pre- based on big data progress track of vehicle more particularly, to one kind
The method of survey.
Background technique
Tracking is often escaped by the way of escape using driving after offender's crime by modern society, and public security organ is chasing
When crime divides, on the one hand, to send car tracing offender, on the other hand, it is also desirable to which prejudging offender in advance can
The route of running away of energy, arranges police strength on most possible multiple routes, carries out interception in advance.
With the increase for relating to vehicle crime, the angle that play increasingly key in case cracking is investigated and seized for deploying to ensure effective monitoring and control of illegal activities for vehicle
Color.Currently pursue and capture suspected vehicles, mostly uses and manually check that the traditional approach of the data such as video, picture analyzes its history driving trace
The trip rule of target vehicle is found, or the driving in the multiple special time period regional scopes of certain vehicle is inquired by application system
Then track is summarised in specific place of specific time by analysis and deploys to ensure effective monitoring and control of illegal activities and intercepts, heavy workload, spend human and material resources it is more,
Low efficiency, information delay, and also it is higher for the skill requirement of staff, it is difficult to it promotes and applies, and to vehicle route
Prediction is time-consuming more, delays the time for chasing offender, even if in addition, many times predict the driving path of vehicle,
It is difficult the time for predicting that vehicle passes through the driving path, causes to arrange police strength not in time, these defects are all chased to public security organ
Offender causes many difficulties.
On the other hand, the mode for capturing vehicle route data gradually increases, video recording, picture and biography including camera shooting
Manual record data of system etc., increasing for acquisition mode lead to increasing sharply for data volume, majority of case, and people are to be difficult to
Short time quickly grasps so a large amount of data and is analyzed and summarized, thus prejudge the track of vehicle.
Therefore it provides the experience of personnel is wanted in a kind of method that can carry out track of vehicle prediction based on big data, reduction
It asks, is this field urgent problem to be solved.
Summary of the invention
In view of this, being solved existing the present invention provides a kind of method for carrying out track of vehicle prediction based on big data
The problem high to staff's skill requirement in technology.
In order to solve the above-mentioned technical problem, the present invention proposes a kind of method for carrying out track of vehicle prediction based on big data,
Include:
Determine that the driving path of target vehicle, the driving path include: the current site C and first of the target vehicle
History site O1, wherein first history site O1For a upper site adjacent with the current site C；
Any site is chosen as first object site F1；
Obtain history vehicle data, wherein the history vehicle data includes the vehicle of multiple vehicles in preset time period
Data, the vehicle data include driving path of the vehicle in the preset time period, the driving path, comprising:
The acquisition time in the multiple sites and the site；
It is extracted from the history vehicle data:
It successively include first history site O in the driving path1, the current site C and first is referring to site A
The vehicle data, as basic data, in which: described first referring to site A is any site,
It successively include first history site O in the driving path1, the current site C and the first object
Site F1The vehicle data, data as a comparison；
By the basic data, according to the acquisition time from closely to being far ranked up, before obtaining in the basic data
20% vehicle data obtains in the basic data as first foundation group close to after the first foundation group
30% vehicle data obtains last 50% vehicle data in the basic data as the second base set, as
Three base sets；
By the correlation data, according to the acquisition time from closely to being far ranked up, before obtaining in the correlation data
20% vehicle data obtains in the correlation data as the first contrast groups close to after first contrast groups
30% vehicle data obtains last 50% vehicle data in the correlation data as the second contrast groups, as
Three contrast groups；
The target vehicle is calculated using following equation and passes through the first object site F1The first probability P1X, and from working as
Preceding site C to the first object site F1First time T used1X:
P1X=(U11X*5+U12X*3+U13X*2)/(U11*5+U12*3+U13*2)
T1X=(T11X*5+T12X*3+T13X*2)/(T11*5+T12*3+T13*2)
Wherein: U11XIt is the number of vehicle data described in first contrast groups, U12XIt is institute in second contrast groups
State the number of vehicle data, U13XIt is the number of vehicle data described in the third contrast groups, U11It is in the first foundation group
The number of the vehicle data, U12It is the number of vehicle data described in second base set, U13It is the third base set
Described in vehicle data number, T11XBe in first contrast groups each vehicle from the current site C to the first object
Site F1The average used time, T12XBe in second contrast groups each vehicle from the current site C to the first object site
F1The average used time, T13XBe in the third contrast groups each vehicle from the current site C to the first object site F1's
Average used time, T11Be in the first foundation group each vehicle from the current site C to it is described first referring to site A average use
When, T12Be in second base set each vehicle from the current site C to it is described first referring to site A the average used time, T13
Be in the third base set each vehicle from the current site C to it is described first referring to site A the average used time.
Optionally, further includes:
Any site is chosen as the second target site F2；
It is extracted from the history vehicle data:
It successively include first history site O in the driving path1, the current site C, described first referring to position
Point A and second referring to site B the vehicle data, as master data, in which: described second referring to site B is any bit
Point,
It successively include first history site O in the driving path1, the current site C, the first object position
Point F1With the second target site F2Vehicle data, as contrasting data；
By the master data, according to the acquisition time from closely to being far ranked up, before obtaining in the master data
20% vehicle data obtains after organizing substantially in the master data close to described first as first basic group
30% vehicle data obtains last 50% vehicle data in the master data as second basic group, as
Three basic groups；
By the contrasting data, according to the acquisition time from closely to being far ranked up, before obtaining in the contrasting data
20% vehicle data obtains in the contrasting data as the first control group close to after first control group
30% vehicle data obtains last 50% vehicle data in the contrasting data as the second control group, as
Three control groups；
The target vehicle is calculated using following equation and passes through the first object site F1With the second target site F2
The second probability P2X, and from the current site C pass through the first object site F1With the second target site F2Institute
Second time T2X:
P2X=(U21X*5+U22X*3+U23X*2)/(U21*5+U22*3+U23*2)
T2X=(T21X*5+T22X*3+T23X*2)/(T21*5+T22*3+T23*2)
Wherein: U21XIt is the number of vehicle data described in first control group, U22XIt is institute in second control group
State the number of vehicle data, U23XIt is the number of vehicle data described in the third control group, U21It is in described first basic group
The number of the vehicle data, U22It is the number of vehicle data described in described second basic group, U23It is third group substantially
Described in vehicle data number, T21XBe in first control group each vehicle from the current site C by first mesh
Mark point F1With the second target site F2The average used time, T12XBe in second control group each vehicle from the present bit
Point C passes through the first object site F1With the second target site F2The average used time, T23XIt is in the third control group
Each vehicle passes through the first object site F from the current site C1With the second target site F2The average used time, T21
Be in described first basic group each vehicle from the current site C by described first referring to site A and described second referring to position
The average used time of point B, T12Be in described second basic group each vehicle from the current site C by described first referring to site A
With the average used time of the second reference site B, T13It is that each vehicle from the current site C passes through institute in third group substantially
Stated for the first average used time referring to site A and described second referring to site B.
Optionally, any site is being chosen as first object site F1Before, further includes: from the institute of the target vehicle
It states in driving path and obtains, with first history site O1An adjacent upper site, as the second history site O2；
It all successively include: described second to go through at this point, in the driving path of the basic data and the correlation data
History site O2, first history site O1With the current site C.
Optionally, any site is being chosen as first object site F1Before, further includes: from the institute of the target vehicle
It states in driving path and obtains, with first history site O1An adjacent upper site, as the second history site O2；
At this point, the traveling road of the basic data, the master data, the correlation data and the contrasting data
It all successively include: second history site O in diameter2, first history site O1With the current site C.
Optionally, the acquisition history vehicle data, further are as follows:
Obtain unstructured vehicle data；
Structured vehicle data are converted by the unstructured vehicle data；
All structured vehicle data are obtained as the history vehicle data.
Optionally, the structured vehicle data are stored in the distributed data base system towards column, the database
System external provides unified interface；
It is described to convert structured vehicle data for the unstructured vehicle data, further are as follows:
The information is carried out typing according to the format of the structure by the information for extracting the unstructured vehicle data.
Optionally, the unstructured vehicle data, comprising: video, picture and voice；
The information for extracting the unstructured vehicle data, comprising:
Judge the classification of the unstructured vehicle data；
When the unstructured vehicle data are the videos, the frame picture in the video is extracted, is known using picture
Other method obtains the information in the frame picture；
When the unstructured vehicle data are the pictures, the frame picture is obtained using the image identification method
In information；
When unstructured vehicle data are voices, text is converted for the voice using speech recognition system, obtains institute
State the information in text.
Optionally, the vehicle data, further includes: license plate number.
Compared with prior art, a kind of method carrying out track of vehicle prediction based on big data provided by the invention, is realized
It is following the utility model has the advantages that
A kind of method that track of vehicle prediction is carried out based on big data is provided, the personnel's rawness for using this method is wanted
Ask, solve the problems, such as it is high to staff's skill requirement in the prior art, and using this method can prediction vehicle row
While sailing path, the time of prediction vehicle driving to the path solves the problems, such as that it is difficult to predict the times in the prior art, originally
Unified interface is used in method, uniformly converts structural data for unstructured data, makes each separate sources, format
Vehicle data conveniently accesses analysis system, reduces the complexity of data access, improves the accessible range of data.
By referring to the drawings to the detailed description of exemplary embodiment of the present invention, other feature of the invention and its
Advantage will become apparent.
Detailed description of the invention
It is combined in the description and the attached drawing for constituting part of specification shows the embodiment of the present invention, and even
With its explanation together principle for explaining the present invention.
Fig. 1 is a kind of flow chart for the method that track of vehicle prediction is carried out based on big data in the embodiment of the present invention 1；
Fig. 2 is a kind of flow chart for the method that track of vehicle prediction is carried out based on big data in the embodiment of the present invention 2.
Specific embodiment
Carry out the various exemplary embodiments of detailed description of the present invention now with reference to attached drawing.It should also be noted that unless in addition having
Body explanation, the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
The range of invention.
Be to the description only actually of at least one exemplary embodiment below it is illustrative, never as to the present invention
And its application or any restrictions used.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as part of specification.
It is shown here and discuss all examples in, any occurrence should be construed as merely illustratively, without
It is as limitation.Therefore, other examples of exemplary embodiment can have different values.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need that it is further discussed.
Embodiment 1
Fig. 1 is a kind of flow chart for the method that track of vehicle prediction is carried out based on big data in the embodiment of the present invention 1, is such as schemed
Shown in 1, a kind of method that track of vehicle prediction is carried out based on big data provided by the invention, comprising:
S101: the driving path of target vehicle is determined.
Specifically, it needs to be determined that carry out the target vehicle of trajectory predictions, by camera, take pictures, the side such as manual record
Formula obtains the driving path that target vehicle has been passed by, and driving path includes: the current site C and the first history bit of target vehicle
Point O1, wherein the first history site O1For a upper site adjacent with current site C, that is to say, that target vehicle is from first
History site O1Current site C is driven to, the driving path for being now to pass by based on target vehicle predicts that it may be walked
Next site, driving path may include multiple history sites, only choose the first history site in the present embodiment, can be with
Speed when improving operation improves the accuracy of prediction but it is also possible to choose multiple history sites.
S102: any site is chosen as first object site F1。
Specifically, first object site F1It is the next site that speculates the target vehicle and may drive to, that is, guesses mesh
It marks vehicle and drives to first object site F from current site C1, method provided by the invention is it is expected that drive to first object
Position F1Probability and the time, first object site F1It can be any site, that is to say, that can be the first history site O1,
It can be current site C, i.e., prediction target vehicle returns to the first history site O from current site C1Or rest on current site
C is not moved.First object site F1, with current site C or the first history site O1It can also be different.
S103: obtaining history vehicle data, extracts basic data and correlation data.
Specifically, history vehicle data includes the vehicle data of multiple vehicles in preset time period, vehicle data includes one
The driving path of a vehicle within a preset period of time, driving path include the site of multiple acquisition times, from history vehicle data
Middle extraction: successively including the first history site O in driving path1, current site C and first referring to the vehicle data of site A, make
For basic data, in which: first referring to site A is any site, then is extracted from history vehicle data: in driving path successively
Including the first history site O1, current site C and first object site F1Vehicle data, data as a comparison.History vehicle number
According to data volume it is larger, in order to improve calculating speed, can set preset time period is 1 year, statistics this section till now the year before
In time, in all driving paths, comprising: the first history site O1With the vehicle data of current site C, then therefrom acquisition traveling
It simultaneously include first object site F in path1Vehicle data, data as a comparison, and including the first history site O1, when
Preceding site C and first is used as basic data referring to the vehicle data of site A.It should be noted that a vehicle, when preset
In, repeatedly run over the first history site O1, current site C and first object site F1, or repeatedly run over the first history
Site O1, current site C and first referring to site A, should be considered as multiple vehicle datas at this time, and cannot be because of being a vehicle
In the data of different time, a vehicle data is treated as that is, by site specified several times and is just denoted as several vehicle datas.
S104: correlation data and basic data are grouped.
By basic data, according to acquisition time from closely to being far ranked up, obtaining in basic data preceding 20% vehicle number
According to, as first foundation group, obtain in basic data close to rear 30% vehicle data of first foundation group as the second base set,
Last 50% vehicle data in basic data is obtained, as third base set.
By correlation data, according to acquisition time from closely to being far ranked up, obtaining in correlation data preceding 20% vehicle number
According to, as the first contrast groups, obtain in correlation data close to rear 30% vehicle data of the first contrast groups as the second contrast groups,
Last 50% vehicle data in correlation data is obtained, as third contrast groups.
Specifically, the data acquired in the recent period are placed on front, most according to acquisition time from closely to being far ranked up, just referring to
The data early acquired are put behind, such as acquisition time section is the January to December of same year, then before the data of acquisition in December are placed on
Face, January, the data of acquisition were placed on finally.
S105: the first probability is calculated.
Target vehicle is calculated using following equation and passes through first object site F1The first probability P1X:
P1X=(U11X*5+U12X*3+U13X*2)/(U11*5+U12*3+U13*2)
Wherein: U11XIt is the number of vehicle data in the first contrast groups, U12XIt is the number of vehicle data in the second contrast groups,
U13XIt is the number of vehicle data in third contrast groups, U11It is the number of vehicle data in first foundation group, U12It is the second basis
The number of vehicle data, U in group13It is the number of vehicle data in third base set.
Specific: the driving path of people's driving vehicle and time are closely related, because of the originals such as urban construction, road change
Cause, even destination is identical, path selected by interior people is also different in different times, so most reference price
The data of value are the data acquired in the recent period, so the maximum of the weight setting of the first contrast groups, the data of acquisition time more rearward
Weighted value is smaller, so that estimating target vehicle passes through first object site F1Probability, by repeating the above steps, it will be able to
The driving path that target vehicle is most possibly chosen is calculated, optionally, chooses first object site F1When, it can be referring initially to
The driving path of one base set chooses wherein most possible site as first object site F1。
S106: first time T is calculated1X。
Target vehicle is calculated from current site C to first object site F using following equation1First time T used1X:
T1X=(T11X*5+T12X*3+T13X*2)/(T11*5+T12*3+T13*2)
Wherein: T11XBe in the first contrast groups each vehicle from current site C to first object site F1The average used time, T12X
Be in the second contrast groups each vehicle from current site C to first object site F1The average used time, T13XIt is each in third contrast groups
Vehicle is from current site C to first object site F1The average used time, T11Be in first foundation group each vehicle from current site C to
The average used time of first reference site A, T12It is each vehicle putting down from the current reference of site C to first site A in the second base set
Equal used time, T13It is average used time of each vehicle from current site C to first referring to site A in third base set.
Specifically, first time T1XIt is to reach first object site F from current site C1Time used, when calculating first
Between T1XWhen setting weight the considerations of with calculate the first probability P1XWhen the considerations of it is identical, in a practical situation, since target vehicle is past
Toward being constantly to move, so calculated first probability P1XBe it is effective, in the prior art, majority do not provide target
The vehicle mobile time for working as next site, or predicted only by distance divided by the mode of target vehicle average speed,
And in actual conditions, the road conditions in each place are different, and are existed very with distance divided by the prediction that the mode of speed carries out
Large deviation combines specific vehicle data in the present invention, and assigns different weights according to different acquisition times, give
Most possible first time T1X, improve prediction accuracy.
Embodiment 2
The improvement that embodiment 2 carries out on the basis of embodiment 1, all explanations done in embodiment 1 are also applied for
Embodiment 2 is based on current site C and the first history site O in embodiment 11Prediction will by next site, implement
Example 2 is based on current site C and the first history site O1Prediction will by lower two sites.
Fig. 2 is a kind of flow chart for the method that track of vehicle prediction is carried out based on big data in the embodiment of the present invention 2, is such as schemed
Shown in 2, a kind of method that track of vehicle prediction is carried out based on big data provided by the invention, comprising:
S201: the driving path of target vehicle is determined.
S202: any site is chosen as first object site F1, any site is chosen as the second target site F2。
Specifically, first object site F1With the second target site F2Speculate that the target vehicle may successively drive to
Lower two sites, i.e. conjecture target vehicle drive to first object site F from current site C1, then drive to the second target site
F2, first object site F1With the second target site F2It can be any site, that is to say, that the two can be identical or not
Together.
S203: obtaining history vehicle data, extracts master data and contrasting data.
It is extracted from history vehicle data: successively including the first history site O in driving path1, current site C, first
Referring to site A and second referring to the vehicle data of site B, as master data, in which: second referring to site B is any site.
It is extracted from history vehicle data: successively including the first history site O in driving path1, current site C, first
Target site F1With the second target site F2Vehicle data, as contrasting data.
S204: master data and contrasting data are grouped.
By master data, according to acquisition time from closely to being far ranked up, obtaining in master data preceding 20% vehicle number
According to, as first basic group, rear 30% vehicle data organized substantially in master data close to first is obtained, it is basic as second
Group obtains last 50% vehicle data in master data, as third group substantially.
By contrasting data, according to acquisition time from closely to being far ranked up, obtaining in contrasting data preceding 20% vehicle number
According to, as the first control group, obtain in contrasting data close to rear 30% vehicle data of the first control group as the second control group,
Last 50% vehicle data in contrasting data is obtained, as third control group.
S205: the second probability P is calculated2X。
Target vehicle is calculated using following equation and passes through first object site F1With the second target site F2The second probability
P2X:
P2X=(U21X*5+U22X*3+U23X*2)/(U21*5+U22*3+U23*2)
Wherein: U21XIt is the number of vehicle data in the first control group, U22XIt is the number of vehicle data in the second control group,
U23XIt is the number of vehicle data in third control group, U21It is the number of vehicle data in first basic group, U22It is second basic
The number of vehicle data, U in group23It is the number of vehicle data in third group substantially.
S206: the second time T is calculated2X。
Target vehicle is calculated using following equation and passes through first object site F from current site C1With the second target site F2
Second time T used2X:
T2X=(T21X*5+T22X*3+T23X*2)/(T21*5+T22*3+T23*2)
Wherein: T21XBe in the first control group each vehicle from current site C by first object site F1With the second target position
Point F2The average used time, T12XBe in the second control group each vehicle from current site C by first object site F1With the second target
Site F2The average used time, T23XBe in third control group each vehicle from current site C by first object site F1With the second mesh
Mark point F2The average used time, T21It is that each vehicle is joined by first referring to site A and second from current site C in first basic group
According to the average used time of site B, T12It is that each vehicle is joined by first referring to site A and second from current site C in second basic group
According to the average used time of site B, T13It is that each vehicle is joined by first referring to site A and second from current site C in third group substantially
According to the average used time of site B.
Specifically, in the present embodiment, predicting and passing sequentially through first object site F1With the second target site F2Second
Probability P2XAnd the second time T2X, the second time T2XIt is from current site C by first object site F1Reach the second target position
Point F2Time-consuming optionally can also first predict to reach first object site from current site C using the method in embodiment 1
F1Probability and time-consuming, then predict from first object site F1To the second target site F2Probability and time-consuming, thus obtain second
Probability P2XAnd the second time T2X, compared to the method predicted in two steps, the method provided in embodiment 2 is calculated by a step,
It directly obtains and passes sequentially through first object site F1With the second target site F2The second probability P2XAnd the second time T2XIt is general
Rate is crossed to reduce time-consuming and improves calculating speed, by choosing the first different site F1With the second site F2And respectively
Calculate corresponding 2nd P2XIt is known that the driving path that target vehicle is most possible.
Further, in some alternative embodiments, any site is being chosen as first object site F1Before, it also wraps
It includes: being obtained from the driving path of target vehicle, with the first history site O1An adjacent upper site, as the second history site
O2；
At this point, all successively including: the second history site O in the driving path of basic data and correlation data2, first go through
History site O1With current site C.
Further, in some alternative embodiments, any site is being chosen as first object site F1Before, also
It include: to be obtained from the driving path of target vehicle, with the first history site O1An adjacent upper site, as the second history bit
Point O2；
At this point, basic data, master data, correlation data and contrasting data driving path in, all successively include: second
History site O2, the first history site O1With current site C.
Specifically, being all by current site C and the first history site O in embodiment 1 and embodiment 21To predict down
One or two site can be based on the second history site O to improve precision of prediction2, the first history site O1And present bit
Point C is predicted, optionally, can be in the case where having plenty of time to improve the reliability and accuracy of calculated result
More history sites are chosen, the reliability of calculated result is further increased.
Further, in some optional embodiments, history vehicle data is obtained, further are as follows: obtain unstructured vehicle
Data, convert structured vehicle data for unstructured vehicle data, obtain all structured vehicle data as history
Vehicle data.
Specifically, vehicle data may be from various aspects, there are data, the computers such as the electric police for being easier to access, bayonet can
Analysis is directly read, also there is video information, other relevant informations such as confession of witness from society's monitoring.Single platform
Do not have the ability for summarizing and integrating different channel informations.The information such as video, picture about vehicle are unstructured vehicle number
According to, be not easy to computer directly analyze with use, structuring vehicle is converted by architectural system by unstructured vehicle data
Data can solve unstructured vehicle data and be not readily used for the problem of computer is directly analyzed with using, using this method
System, the unstructured vehicle data that will acquire transfer to unstructured storage system uniformly to store and provide access service, knot
After unstructured vehicle data are converted structured vehicle data by structure system, can further it locate with resource description information
Storage is into the distributed data base system towards column after reason.
Further, in some optional embodiments, structured vehicle data are stored in the distributed data base towards column
In system, Database Systems externally provide unified interface；Structured vehicle data are converted by unstructured vehicle data, into
One step are as follows: information is carried out typing according to the format of structure by the information for extracting unstructured vehicle data.
Specifically, being that the distributed data base towards column externally provides a unified data-interface, information of vehicles is solved
The problem of format is inconsistent, and source is inconsistent, is not easy to obtain, can not analyze comprehensively, and the distributed data of using face nematic
The vehicle data of inventory's storage structure can solve conventional relationship type database and be unable to vehicle data exponential increase and a large amount of
The problem of data store and system are not easy dilatation and the bad problem of fault-tolerant redundancy ability.
Optionally, unstructured vehicle data, comprising: video, picture and voice；Extract the letter of unstructured vehicle data
Breath, comprising: judge the classification of unstructured vehicle data；When unstructured vehicle data are videos, the frame in video is extracted
Picture, using the information in image identification method getting frame picture；When unstructured vehicle data are pictures, known using picture
Information in other method getting frame picture；When unstructured vehicle data are voices, voice is converted using speech recognition system
For text, the information in text is obtained.
Specific: general bayonet, electricity police etc. are disposed by planning big companies, government department according to certain Specification Design, can
Vehicle data is directly read by data address or timing inquires its database and obtains vehicle data, then the online money that will be read
The temp directory that source storage is specified to system, and resource associated description information is obtained automatically.The view of the acquisitions such as social monitoring device
Frequently, the resources such as picture do not access network generally, can not directly read, and can only be obtained by offline mode, can directly will be related
Resource is copied directly to system and specifies temp directory, and fills in the relevant informations such as time, the position of the data, system meeting in system
It is handled respectively for the type of resource, using image recognition technology or speech recognition technology, obtains information therein.
Further, in some alternative embodiments, vehicle data, further includes: license plate number.
The present invention provides it is a kind of based on big data carry out track of vehicle prediction method, to use the personnel of this method without
Skill requirement, solves the problems, such as high to staff's skill requirement in the prior art, and can predicted using this method
While vehicle running path, the time of prediction vehicle driving to the path, solve in the prior art that it is difficult to predict the times
Problem uses unified interface in this method, uniformly converts structural data for unstructured data, make it is each it is different come
Source, format vehicle data conveniently access analysis system, reduce the complexity of data access, improve what data can access
Range.
Although some specific embodiments of the invention are described in detail by example, the skill of this field
Art personnel it should be understood that example above merely to being illustrated, the range being not intended to be limiting of the invention.The skill of this field
Art personnel are it should be understood that can without departing from the scope and spirit of the present invention modify to above embodiments.This hair
Bright range is defined by the following claims.
Claims (8)
1. a kind of method for carrying out track of vehicle prediction based on big data characterized by comprising
Determine that the driving path of target vehicle, the driving path include: the current site C and the first history of the target vehicle
Site O1, wherein first history site O1For a upper site adjacent with the current site C；
Any site is chosen as first object site F1；
Obtaining history vehicle data, wherein the history vehicle data includes the vehicle data of multiple vehicles in preset time period,
The vehicle data includes driving path of the vehicle in the preset time period, the driving path, comprising: Duo Gesuo
The acquisition time of rheme point and the site；
It is extracted from the history vehicle data:
It successively include first history site O in the driving path1, the current site C and first referring to site A institute
Vehicle data is stated, as basic data, in which: described first referring to site A is any site,
It successively include first history site O in the driving path1, the current site C and the first object site F1
The vehicle data, data as a comparison；
By the basic data, according to the acquisition time from closely to being far ranked up, obtaining preceding 20% in the basic data
The vehicle data obtain rear 30% institute in the basic data close to the first foundation group as first foundation group
Vehicle data is stated as the second base set, obtains last 50% vehicle data in the basic data, as third basis
Group；
By the correlation data, according to the acquisition time from closely to being far ranked up, obtaining preceding 20% in the correlation data
The vehicle data obtain rear 30% institute in the correlation data close to first contrast groups as the first contrast groups
Vehicle data is stated as the second contrast groups, obtains last 50% vehicle data in the correlation data, is compared as third
Group；
The target vehicle is calculated using following equation and passes through the first object site F1The first probability P1X, and from present bit
Point C to the first object site F1First time T used1X:
P1X=(U11X*5+U12X*3+U13X*2)/(U11*5+U12*3+U13*2)
T1X=(T11X*5+T12X*3+T13X*2)/(T11*5+T12*3+T13*2)
Wherein: U11XIt is the number of vehicle data described in first contrast groups, U12XIt is vehicle described in second contrast groups
The number of data, U13XIt is the number of vehicle data described in the third contrast groups, U11It is described in the first foundation group
The number of vehicle data, U12It is the number of vehicle data described in second base set, U13It is institute in the third base set
State the number of vehicle data, T11XBe in first contrast groups each vehicle from the current site C to the first object site
F1The average used time, T12XBe in second contrast groups each vehicle from the current site C to the first object site F1's
Average used time, T13XBe in the third contrast groups each vehicle from the current site C to the first object site F1Be averaged
Used time, T11Be in the first foundation group each vehicle from the current site C to it is described first referring to site A the average used time,
T12Be in second base set each vehicle from the current site C to it is described first referring to site A the average used time, T13It is
Each vehicle is from the current site C to described first referring to the average used time of site A in the third base set.
2. a kind of method for carrying out track of vehicle prediction based on big data according to claim 1, which is characterized in that also wrap
It includes:
Any site is chosen as the second target site F2；
It is extracted from the history vehicle data:
It successively include first history site O in the driving path1, the current site C, described first referring to site A and
Second referring to site B the vehicle data, as master data, in which: described second referring to site B is any site,
It successively include first history site O in the driving path1, the current site C, the first object site F1
With the second target site F2Vehicle data, as contrasting data；
By the master data, according to the acquisition time from closely to being far ranked up, obtaining preceding 20% in the master data
The vehicle data obtain rear 30% institute organized substantially in the master data close to described first as first basic group
Vehicle data is stated, as second basic group, obtains last 50% vehicle data in the master data, it is basic as third
Group；
By the contrasting data, according to the acquisition time from closely to being far ranked up, obtaining preceding 20% in the contrasting data
The vehicle data obtain rear 30% institute in the contrasting data close to first control group as the first control group
Vehicle data is stated as the second control group, obtains last 50% vehicle data in the contrasting data, is compareed as third
Group；
The target vehicle is calculated using following equation and passes through the first object site F1With the second target site F2?
Two probability Ps2X, and from the current site C pass through the first object site F1With the second target site F2Used
Second time T2X:
P2X=(U21X*5+U22X*3+U23X*2)/(U21*5+U22*3+U23*2)
T2X=(T21X*5+T22X*3+T23X*2)/(T21*5+T22*3+T23*2)
Wherein: U21XIt is the number of vehicle data described in first control group, U22XIt is vehicle described in second control group
The number of data, U23XIt is the number of vehicle data described in the third control group, U21It is described in described first basic group
The number of vehicle data, U22It is the number of vehicle data described in described second basic group, U23It is institute in third group substantially
State the number of vehicle data, T21XBe in first control group each vehicle from the current site C by the first object position
Point F1With the second target site F2The average used time, T12XBe in second control group each vehicle from the current site C
Pass through the first object site F1With the second target site F2The average used time, T23XIt is each vehicle in the third control group
From the current site C pass through the first object site F1With the second target site F2The average used time, T21It is institute
State each vehicle in first basic group from the current site C by described first referring to site A and described second referring to site B's
Average used time, T12Be in described second basic group each vehicle from the current site C by described first referring to site A and described
The average used time of second reference site B, T13Be in third group substantially each vehicle from the current site C by described first
Referring to site A and described second referring to the average used time of site B.
3. a kind of method for carrying out track of vehicle prediction based on big data according to claim 1, which is characterized in that selecting
Take any site as first object site F1Before, further includes: it is obtained from the driving path of the target vehicle, with
First history site O1An adjacent upper site, as the second history site O2；
At this point, all successively including: second history bit in the driving path of the basic data and the correlation data
Point O2, first history site O1With the current site C.
4. a kind of method for carrying out track of vehicle prediction based on big data according to claim 2, which is characterized in that selecting
Take any site as first object site F1Before, further includes: it is obtained from the driving path of the target vehicle, with
First history site O1An adjacent upper site, as the second history site O2；
At this point, the driving path of the basic data, the master data, the correlation data and the contrasting data
In, it all successively include: second history site O2, first history site O1With the current site C.
5. a kind of method for carrying out track of vehicle prediction based on big data according to claim 1 to 4, feature exist
In, the acquisition history vehicle data, further are as follows:
Obtain unstructured vehicle data；
Structured vehicle data are converted by the unstructured vehicle data；
All structured vehicle data are obtained as the history vehicle data.
6. a kind of method for carrying out track of vehicle prediction based on big data according to claim 5, which is characterized in that described
Structured vehicle data are stored in the distributed data base system towards column, and the Database Systems externally provide unified connect
Mouthful；
It is described to convert structured vehicle data for the unstructured vehicle data, further are as follows:
The information is carried out typing according to the format of the structure by the information for extracting the unstructured vehicle data.
7. a kind of method for carrying out track of vehicle prediction based on big data according to claim 6, which is characterized in that described
Unstructured vehicle data, comprising: video, picture and voice；
The information for extracting the unstructured vehicle data, comprising:
Judge the classification of the unstructured vehicle data；
When the unstructured vehicle data are the videos, the frame picture in the video is extracted, using picture recognition side
Method obtains the information in the frame picture；
When the unstructured vehicle data are the pictures, obtained in the frame picture using the image identification method
Information；
When unstructured vehicle data are voices, text is converted for the voice using speech recognition system, obtains the text
Information in word.
8. according to a kind of method for carrying out track of vehicle prediction based on big data described in claim 1, which is characterized in that the vehicle
Data, further includes: license plate number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711099729.9A CN107862868B (en) | 2017-11-09 | 2017-11-09 | A method of track of vehicle prediction is carried out based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711099729.9A CN107862868B (en) | 2017-11-09 | 2017-11-09 | A method of track of vehicle prediction is carried out based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107862868A CN107862868A (en) | 2018-03-30 |
| CN107862868B true CN107862868B (en) | 2019-08-20 |
Family
ID=61701495
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711099729.9A Active CN107862868B (en) | 2017-11-09 | 2017-11-09 | A method of track of vehicle prediction is carried out based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107862868B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108389385B (en) * | 2018-04-03 | 2020-09-18 | 北京万集科技股份有限公司 | Method and device for intercepting overrun vehicle, electronic equipment and storage medium |
| CN109448363B (en) * | 2018-09-30 | 2021-06-08 | 佳都科技集团股份有限公司 | Intelligent suspected vehicle sealing and controlling method and system based on track prediction and processing terminal |
| CN109493266A (en) * | 2018-11-15 | 2019-03-19 | 重庆市城投金卡信息产业(集团)股份有限公司 | Vehicle supervision method and its system based on RFID |
| CN111402580B (en) * | 2020-03-04 | 2022-03-01 | 杭州海康威视系统技术有限公司 | Vehicle running track prediction method and device and electronic equipment |
| CN113282782B (en) * | 2021-05-21 | 2022-09-09 | 三亚海兰寰宇海洋信息科技有限公司 | Track acquisition method and device based on multi-point phase camera array |
| CN114973642B (en) * | 2022-01-17 | 2023-09-08 | 深圳市聚业美家科技有限公司 | Computer path decision system and decision method based on big data track analysis |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101395647A (en) * | 2006-02-28 | 2009-03-25 | 丰田自动车株式会社 | Object course prediction method, device, program, and automatic driving system |
| CN103325245A (en) * | 2013-06-25 | 2013-09-25 | 中国科学院信息工程研究所 | Method for predicting space-time traveling track of blacklisted vehicle |
| CN104239556A (en) * | 2014-09-25 | 2014-12-24 | 西安理工大学 | Density clustering-based self-adaptive trajectory prediction method |
| CN104464344A (en) * | 2014-11-07 | 2015-03-25 | 湖北大学 | Vehicle driving path prediction method and system |
| CN104462190A (en) * | 2014-10-24 | 2015-03-25 | 中国电子科技集团公司第二十八研究所 | On-line position prediction method based on mass of space trajectory excavation |
| CN104834669A (en) * | 2015-03-18 | 2015-08-12 | 广西师范学院 | Multi-order position prediction method of suspect on the basis of spatiotemporal semantics transfer matrix |
| JP5762474B2 (en) * | 2013-01-22 | 2015-08-12 | エヌイーシー（チャイナ）カンパニー， リミテッドＮＥＣ（Ｃｈｉｎａ）Ｃｏ．，Ｌｔｄ． | Hierarchical destination prediction model construction device, destination prediction device, hierarchical destination prediction model construction method, and destination prediction method |
| US9240124B2 (en) * | 2012-08-08 | 2016-01-19 | Hitachi, Ltd. | Traffic-volume prediction device and method |
-
2017
- 2017-11-09 CN CN201711099729.9A patent/CN107862868B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101395647A (en) * | 2006-02-28 | 2009-03-25 | 丰田自动车株式会社 | Object course prediction method, device, program, and automatic driving system |
| US9240124B2 (en) * | 2012-08-08 | 2016-01-19 | Hitachi, Ltd. | Traffic-volume prediction device and method |
| JP5762474B2 (en) * | 2013-01-22 | 2015-08-12 | エヌイーシー（チャイナ）カンパニー， リミテッドＮＥＣ（Ｃｈｉｎａ）Ｃｏ．，Ｌｔｄ． | Hierarchical destination prediction model construction device, destination prediction device, hierarchical destination prediction model construction method, and destination prediction method |
| CN103325245A (en) * | 2013-06-25 | 2013-09-25 | 中国科学院信息工程研究所 | Method for predicting space-time traveling track of blacklisted vehicle |
| CN104239556A (en) * | 2014-09-25 | 2014-12-24 | 西安理工大学 | Density clustering-based self-adaptive trajectory prediction method |
| CN104462190A (en) * | 2014-10-24 | 2015-03-25 | 中国电子科技集团公司第二十八研究所 | On-line position prediction method based on mass of space trajectory excavation |
| CN104464344A (en) * | 2014-11-07 | 2015-03-25 | 湖北大学 | Vehicle driving path prediction method and system |
| CN104834669A (en) * | 2015-03-18 | 2015-08-12 | 广西师范学院 | Multi-order position prediction method of suspect on the basis of spatiotemporal semantics transfer matrix |
Non-Patent Citations (1)
| Title |
|---|
| 基于路网的不确定性轨迹预测;郭黎敏 等;《计算机研究与发展》;20101231;第104-112页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107862868A (en) | 2018-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107862868B (en) | A method of track of vehicle prediction is carried out based on big data | |
| CN109598066B (en) | Effect evaluation method, apparatus, device and storage medium for prediction module | |
| US10346888B2 (en) | Systems and methods to obtain passenger feedback in response to autonomous vehicle driving events | |
| EP2980767B1 (en) | Video search and playback interface for vehicle monitor | |
| CN102682610B (en) | Automatic vehicle speed measurement and enforcement approach and system | |
| CN112052826A (en) | Intelligent enforcement multi-scale target detection method, device and system based on YOLOv4 algorithm and storage medium | |
| CN108765239A (en) | Land resource information management system | |
| CN115828112B (en) | Fault event response method and device, electronic equipment and storage medium | |
| US10984254B2 (en) | Investigation assist system and investigation assist method | |
| CN104679864A (en) | Intelligent tracking method and intelligent tracking device for suspected target based on GIS (Geographic Information System) | |
| CN111125290B (en) | Intelligent river patrol method and device based on river growth system and storage medium | |
| CN105989610A (en) | Target tracking method and device | |
| EP4170605A1 (en) | Multi-level transferable region-based domain adaptive object detection apparatus and method | |
| EP2735984A1 (en) | Video query method, device and system | |
| US20120093395A1 (en) | Method and system for hierarchically matching images of buildings, and computer-readable recording medium | |
| JP2019192209A (en) | Learning target image packaging device and method for artificial intelligence of video movie | |
| CN113807622A (en) | Event decision generation method and device, electronic equipment and storage medium | |
| Li et al. | DBUS: Human driving behavior understanding system | |
| US9953466B2 (en) | Monitoring objects in motion along a static route using sensory detection devices | |
| CN108156406A (en) | Information processing method and device of automobile data recorder | |
| KR102546193B1 (en) | Method for learning data classification using color information, and computer program recorded on record-medium for executing method thereof | |
| CN112182302B (en) | Subway taking route tracing method and device based on transfer management | |
| Ogawa et al. | Identifying Parking Lot Occupancy with YOLOv5 | |
| CN113421067A (en) | Shared bicycle brand identification and quantity management method and system | |
| Fujita et al. | Fine-tuned Surface Object Detection Applying Pre-trained Mask R-CNN Models |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |