CN108122186A - Location estimation method is lived in a kind of duty based on bayonet data - Google Patents
Location estimation method is lived in a kind of duty based on bayonet data Download PDFInfo
- Publication number
- CN108122186A CN108122186A CN201711468924.4A CN201711468924A CN108122186A CN 108122186 A CN108122186 A CN 108122186A CN 201711468924 A CN201711468924 A CN 201711468924A CN 108122186 A CN108122186 A CN 108122186A
- Authority
- CN
- China
- Prior art keywords
- data
- vehicle
- point
- record
- bayonet
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 10
- 239000000284 extract Substances 0.000 claims abstract description 6
- 238000003064 k means clustering Methods 0.000 claims abstract description 6
- 238000013480 data collection Methods 0.000 claims description 6
- 238000007621 cluster analysis Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000005215 recombination Methods 0.000 claims description 3
- 230000006798 recombination Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000037361 pathway Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- 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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Educational Administration (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Evolutionary Computation (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of vehicle duties to live location estimation method, the described method includes:Step (1) obtains bayonet data and bayonet data is pre-processed;Step (2) divides vehicle travel;Step (3) extracts important record point;Step (4) uses improved k means clustering algorithms, and sample point is clustered.The home location of vehicle can be effectively estimated, is laid the foundation for further analysis vehicle driving feature etc..
Description
Technical field
The present invention relates to field of traffic more particularly to a kind of duty based on bayonet data to live location estimation method.
Background technology
It is real to road that tollgate devices are disposed upon being combined using video camera as the software and hardware relied on for transport hub road
The intelligent management and control devices of monitoring are applied, including license plate number, the vehicle time can be spent, traveling with 24 hour record of whole day by information of vehicles
Direction etc. provides reliable data supporting for vehicle driving feature and road network state analysis.In trip characteristics, if it is known that
Position is lived in the duty of vehicle, has larger meaning for the estimate analysis for waiting relevant traffics situation and management and control of the magnitude of traffic flow.Institute
Report position work (duty) and residence (living) where position.
The method on traveler OD (beginning and end) estimations has much at present, wherein Application No.
201110163206.2 patent ---《Dynamic OD Matrix Estimation method based on automatic vehicle identification equipment》Mainly by with
Main means of the microscopic pathways reparation of vehicle as dynamic OD estimation, are operated, above-mentioned skill using Bayesian Estimation algorithm
The data volume intensive that art scheme uses is big, for that can not be applicable under the unappeasable environment of hardware condition, predicts
As a result the space being also improved.
The patent of Application No. 201310213953.1 ---《Mass transit card passenger commuting based on intelligent public transportation system data
OD distribution estimation methods》Three kinds of Resident Trip Characteristics are based primarily upon it is assumed that using frequency statistics method and the cluster analysis side of riding
Method calculates resident OD, and this method is limited to the OD data of bus IC card class.
The content of the invention
This patent is based on the demand of the prior art and proposes, the technical problems to be solved by the patent is to this
Invention propose the vehicle duty based on bayonet data live location estimation method can overcome more than deficiency, with reference to bayonet data in itself
Characteristic realizes that location estimation is lived in vehicle duty.
In order to solve the above-mentioned technical problem, the technical solution that this patent provides includes:Location estimation side is lived in a kind of vehicle duty
Method, the described method includes:Step (1) obtains bayonet data and bayonet data is pre-processed;The pretreatment includes:
(1.1) Data Dimensionality Reduction goes out bayonet number K, license plate number M from original bayonet extracting data, crosses vehicle time S, form direction F, deletes
Remaining column data;(1.2) data recombination is grouped first, in accordance with license plate number;All data are grouped according to license plate number M, i.e., by car plate
Number identical data are put into M in a seti={ Mi1, Mi2..., Min, wherein MinExpression license plate number is MiVehicle n-th
Data;Secondly, vehicle time-sequencing was pressed;Then, the data of identical license plate number are ranked up according to vehicle time S excessively, obtained
The vehicle data set M being sequentially arrangedis={ Mis1, Mis2..., Misn, wherein MisnExpression license plate number is MiVehicle
Nth bar data;(1.3) unidentified data are deleted in data cleansing;It deletes and repeats to record;Step (2) division vehicle travel (2.1)
Based on data record time interval, each vehicle is recorded daily and is a little divided, be divided into independent stroke;First, in accordance with day
Phase is to data set MisIn packet, obtain the daily track data set M of each carid={ Mid1, Mid2..., Midn,
Middle MidnExpression license plate number is MiNth bar data of the vehicle at the d days;(2.2) threshold value B, division record point are set;According to vehicle
Daily track data collection Mid, calculate the adjacent two time difference Δ t for recording pointd={ Δ td1, Δ td2..., Δ tdn-1, and be based on
Bayonet data and actual road network, given threshold B, if Δ tdi<B, then i+1 record, which is put, belongs to next stroke, otherwise, belongs to
Thus current track collection obtains the daily all travel path collection of each car, Mid={ Mid1, Mid2..., Midn};Step (3) carries
Important record is taken to click pick up the car daily first and the last one stroke, and extracts two classes record point therein:The first kind
For the first record point of first stroke and the last one record point of the last one stroke, A classes are denoted as;Second class is first
The last one record point of stroke and first record point of the last one stroke, are denoted as B classes;And according to bayonet number and bayonet
The latitude and longitude coordinates of the record point of extraction are put into new sample set P, P by physical location latitude and longitude coordinatesi={ PAi, PBi,
Wherein PAiRepresent all A classes record point data collection of i-th vehicle;Step (4) uses improved k-means clustering algorithms, right
Sample point is clustered the measure that (4.1) improve k-means clustering algorithms, is replaced using spherical distance in K-means
Euclidean distance measures the similitude between sample;The calculation formula of spherical distance is:D(Pi, Pj)=R × arccos (a+
b);Wherein, D (Pi, Pj) represent PiPoint and point PjBetween most short spherical distance;R is earth radius;A=sin (PiLon×π/
180)×sin(PjLon× π/180), PiLonFor point PiLongitude, PjLonFor point PjLongitude;B=cos (PiLat×π/180)×
cos(PjLat×π/180)×cos(PjLon-PiLon), PiLatFor point PiLatitude, PjLatFor point PjLatitude;(4.2) set poly-
Class center K carries out cluster analysis, the number k for the data subset to be generated is set, by the data set PA of inputiK classes are divided into,
Obtain data set C={ c1..., ck, wherein ckExpression is divided into the set of the data of kth class;The wherein initial value of k is from sample
It is randomly selected in this, obtains k classes, therefrom chosen comprising the most class max { C } of sample size, using its cluster centre as estimation
Home location, equally from data set PBiThe middle operating position for clustering obtained cluster centre as estimation.
Preferably, the calculating of threshold value B is specific as follows:(1) the length information L={ l in all sections on m road are obtained1...,
lmAnd every road maximum permission speed V={ v1..., vm};(2) the hourage T=in each section is calculated
{t1..., tm, wherein
(3) B=max { T }, from road network in the journey time in all sections maximizing as threshold value B.
The present invention is around exploratory algorithm, and based on bayonet data, proposition is a kind of to record time interval based on track of vehicle
Trip stroke identification method;The present invention takes into full account actual road network information, and combines card in the calculating of time interval threshold value B
Mouth data calculate the road travel time, ensure the reliability of threshold value B;The present invention combines vehicle commuting feature, from vehicle one-stroke
Tracing point sets out, and a series of " important " record points during traveling is chosen, as location estimation sample set, and according to bayonet data
Self character improves the Euclidean distance in K-means algorithms as spherical distance, important record point is clustered, the side of reducing
The complexity of method;The home location of vehicle can be effectively estimated, is laid the foundation for further analysis vehicle driving feature etc..
Description of the drawings
Fig. 1 is a kind of structure diagram of road network in the specific embodiment of the invention;
Fig. 2 is that the vehicle travel based on exploratory algorithm divides schematic diagram;
Fig. 3 is vehicle driving travel path dot-dash split flow figure;
Fig. 4 is the important record point schematic diagram of track of vehicle.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples so that those skilled in the art's reference
Specification word can be implemented according to this.Present embodiment proposes a kind of home location estimation side based on bayonet data
Method is as follows:
(1) bayonet data are pre-processed;
Data Dimensionality Reduction, data are included from group, data cleansing to the pretreatment of the buckle data.Pass through the pretreatment of data
It is capable of the data content of value -capture maximum, facilitates the implementation of this method, improve the accuracy and efficiency of this method.
(1.1) Data Dimensionality Reduction
Original bayonet data message amount is big, and not all information is useful all for present embodiment.For
Analysis efficiency is improved, key message is extracted from bayonet data in present embodiment, reduces data dimension.Specifically,
In this embodiment, go out bayonet number K, license plate number M from original bayonet extracting data, cross vehicle time S, traveling side
To F, remaining column data is deleted.
(1.2) data recombination
1. it is grouped according to license plate number;All data according to license plate number M are grouped, i.e., the identical data of license plate number are put into one
M in a seti={ Mi1, Mi2..., Min, wherein MinExpression license plate number is MiVehicle nth bar data.
2. pressed vehicle time-sequencing;Then the data of identical license plate number are ranked up according to vehicle time S excessively, obtained on time
Between tactic vehicle data set Mis={ Mis1, Mis2..., Misn, wherein MisnExpression license plate number is MiVehicle n-th
Data.
(1.3) data cleansing
Delete unidentified data;Fail the record of identification for license plate number in bayonet data, carries out delete operation;It deletes
Except repetition records;The repetition continuously occurred for vehicle by repeated detection records, and carries out delete operation.
It is commonly considered as data and license plate number identification mistake of the license plate number labeled as " unidentified " printed words, such as mess code,
Data are that license plate number fails the record of identification, these records need to delete.Repeated data includes bayonet number K, license plate number
M crosses vehicle time S, the identical data of travel direction F, and the data overlapped for partial content can not be considered weight
Complex data only retains a data in repeated data for repeated data, deletes remaining data.
(2) vehicle travel is divided
(2.1) based on data record time interval, each vehicle is recorded daily and is a little divided, is divided into independent row
Journey,
First, in accordance with the date to data set MisIn packet, obtain the daily track data set M of each carid=
{Mid1, Mid2..., Midn, wherein MidnExpression license plate number is MiNth bar data of the vehicle at the d days.
(2.2) threshold value B, division record point are set
According to the daily track data collection M of vehicleid, calculate the adjacent two time difference Δ t for recording pointd={ Δ td1, Δ
td2..., Δ tdn-1, and based on bayonet data and actual road network, given threshold B, if Δ tdi<B, then i+1 record, which is put, belongs to
Otherwise next stroke, belongs to current track collection, thus obtain the daily all travel path collection of each car, Mid={ Mid1,
Mid2..., Midn}。
Such as in road network shown in Fig. 1, the length L={ l of all roads in road network are obtained1..., lmAnd it is maximum allowable
Velocity information V={ v1..., vm, and calculate the journey time T={ t in all sections1..., tm, choose the range time
Max { T } is the value of threshold value B.
Based on the track data of each car according to time sequence daily is obtained in step (1), the time of adjacent recorded dots is calculated
Interval, and one by one compared with B, if less than B, the corresponding two adjacent tracing points in this time interval belong to a track collection;
Otherwise if more than B, then the two tracing points belong to two different track collection, and the tracing point in time series rearward is divided into
Next track collection, as shown in Fig. 2, t1, t2, t3, t4All greater than threshold value B, flow chart is as shown in Figure 3;
(3) important record point is extracted
According to vehicle commuting rule, vehicle daily first and the last one stroke are chosen, and extracts two class therein
Record point:
The first kind is the first record point of first stroke and the last one record point of the last one stroke, is denoted as A classes;
Second class is the last one record point of first stroke and first record point of the last one stroke, is denoted as B
Class;
And according to bayonet number and bayonet physical location latitude and longitude coordinates, the latitude and longitude coordinates of the record point of extraction are put into
In new sample set P, Pi={ PAi, PBi, wherein PAiRepresent all A classes record point data collection of i-th vehicle.
:All travel paths of the vehicle that is obtained based on step (2) record point, extract the A classes record point and B of each car
Class records point, as shown in figure 4, being respectively put into sample set A and B, the two class sample sets for obtaining i-th vehicle are respectively PAi, PBi;
(4) k-means clustering algorithms are improved, sample point is clustered
(4.1) measure is improved
In K-means, the similitude between sample is measured usually using Euclidean distance.Tracing point is in bayonet data
Latitude and longitude coordinates, therefore spherical distance is used to replace Euclidean distance.The calculation formula of spherical distance is:
D(Pi, Pj)=R × arccos (a+b)
Wherein, D (Pi, Pj) represent PiPoint and point PjBetween most short spherical distance;R is earth radius;
A=sin (PiLon×π/180)×sin(PjLon× π/180), PiLonFor point PiLongitude, PjLonFor point PjWarp
Degree;
B=cos (PiLat×π/180)×cos(PjLat×π/180)×cos(PjLon-PiLon), PiLatFor point PiLatitude,
PjLatFor point PjLatitude.
(4.2) cluster centre K is set, carries out cluster analysis
The number k for the data subset to be generated is set, by the data set PA of inputiK classes are divided into, obtain data set C=
{c1..., ck, wherein ckExpression is divided into the set of the data of kth class.The wherein initial value of k is randomly selected from sample.
K classes are obtained, are therefrom chosen comprising the most class max { C } of sample size, using its cluster centre as the home location estimated, together
Sample is from data set PBiThe middle operating position for clustering obtained cluster centre as estimation.
Claims (2)
1. location estimation method is lived in a kind of vehicle duty, which is characterized in that the described method includes:
Step (1) obtains bayonet data and bayonet data is pre-processed;
The pretreatment includes:
(1.1) Data Dimensionality Reduction
Go out bayonet number K, license plate number M from original bayonet extracting data, cross vehicle time S, form direction F deletes remaining columns
According to;
(1.2) data recombination
It is grouped first, in accordance with license plate number;All data according to license plate number M are grouped, i.e., the identical data of license plate number are put into one
M in seti={ Mi1, Mi2..., Min, wherein MinExpression license plate number is MiVehicle nth bar data;Secondly, when pressing vehicle
Between sort;Then, the data of identical license plate number are ranked up according to vehicle time S excessively, the vehicle being sequentially arranged
Data acquisition system Mis={ Mis1, Mis2..., Misn, wherein MisnExpression license plate number is MiVehicle nth bar data;
(1.3) data cleansing
Delete unidentified data;It deletes and repeats to record;
Step (2) divides vehicle travel
(2.1) based on data record time interval, each vehicle is recorded daily and is a little divided, is divided into independent stroke;
First, in accordance with the date to data set MisIn packet, obtain the daily track data set M of each carid={ Mid1,
Mid2..., Midn, wherein MidnExpression license plate number is MiNth bar data of the vehicle at the d days;
(2.2) threshold value B, division record point are set;According to the daily track data collection M of vehicleid, calculated for adjacent twice for recording point
Poor Δ td={ Δ td1, Δ td2..., Δ tdn-1, and based on bayonet data and actual road network, given threshold B, if Δ tdi<B, then
I+1 record point belongs to next stroke, otherwise, belongs to current track collection, thus obtains the daily all strokes of each car
Track collection, Mid={ Mid1, Mid2..., Midn};
Step (3) extracts important record point
Vehicle daily first and the last one stroke are chosen, and extracts two classes record point therein:The first kind is first
The first record point of stroke and the last one record point of the last one stroke, are denoted as A classes;Second class for first stroke most
First record point of the latter record point and the last one stroke, is denoted as B classes;And according to bayonet number and bayonet physical location
The latitude and longitude coordinates of the record point of extraction are put into new sample set P, P by latitude and longitude coordinatesi={ PAi, PBi, wherein PAiTable
Show all A classes record point data collection of i-th vehicle;
Step (4) uses improved k-means clustering algorithms, and sample point is clustered
(4.1) measure of k-means clustering algorithms is improved
Similitude between sample is measured instead of the Euclidean distance in K-means using spherical distance;Spherical distance
Calculation formula is:D(Pi, Pj)=R × arccos (a+b);Wherein, D (Pi, Pj) represent PiPoint and point PjBetween most short ball identity distance
From;R is earth radius;A=sin (PiLon×π/180)×sin(PjLon× π/180), PiLonFor point PiLongitude, PjLonFor point
PjLongitude;B=cos (PiLat×π/180)×cos(PjLat×π/180)×cos(PjLon-PiLon), PiLatFor point PiLatitude
Degree, PjLatFor point PjLatitude;
(4.2) cluster centre K is set, carries out cluster analysis
The number k for the data subset to be generated is set, by the data set PA of inputiK classes are divided into, obtain data set C=
{c1..., ck, wherein ckExpression is divided into the set of the data of kth class;The wherein initial value of k is randomly selected from sample,
K classes are obtained, are therefrom chosen comprising the most class max { C } of sample size, using its cluster centre as the home location estimated, together
Sample is from data set PBiThe middle operating position for clustering obtained cluster centre as estimation.
2. location estimation method is lived in a kind of vehicle duty according to claim 1, which is characterized in that the calculating of threshold value B is specific
It is as follows:
(1) the length information L={ l in all sections on m road are obtained1..., lmAnd every road maximum permission speed V
={ v1..., vm};
(2) the hourage T={ t in each section is calculated1..., tm, wherein1<i<m;
(3) B=max { T }, from road network in the journey time in all sections maximizing as threshold value B.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711468924.4A CN108122186B (en) | 2017-12-29 | 2017-12-29 | Job and live position estimation method based on checkpoint data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711468924.4A CN108122186B (en) | 2017-12-29 | 2017-12-29 | Job and live position estimation method based on checkpoint data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108122186A true CN108122186A (en) | 2018-06-05 |
CN108122186B CN108122186B (en) | 2021-11-05 |
Family
ID=62232398
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711468924.4A Active CN108122186B (en) | 2017-12-29 | 2017-12-29 | Job and live position estimation method based on checkpoint data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108122186B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108717790A (en) * | 2018-07-06 | 2018-10-30 | 广州市交通运输研究所 | A kind of vehicle driving analysis method based on bayonet license plate identification data |
CN109308804A (en) * | 2018-08-08 | 2019-02-05 | 北京航空航天大学 | Hourage estimation method based on tensor resolution |
CN109471907A (en) * | 2018-11-15 | 2019-03-15 | 刘长山 | A kind of driving law-analysing system and method based on bayonet data |
CN111080351A (en) * | 2019-12-05 | 2020-04-28 | 任子行网络技术股份有限公司 | Clustering method and system for multi-dimensional data set |
CN111523562A (en) * | 2020-03-20 | 2020-08-11 | 浙江大学 | Commuting mode vehicle identification method based on license plate identification data |
CN111640303A (en) * | 2020-05-29 | 2020-09-08 | 青岛大学 | City commuting path identification method and equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103646187A (en) * | 2013-12-27 | 2014-03-19 | 中国科学院自动化研究所 | Method for obtaining vehicle travel path and OD (Origin-Destination) matrix in statistic period |
CN105608528A (en) * | 2015-12-18 | 2016-05-25 | 东南大学 | Taxi driver income-pressure assessment method and system based on big data analysis |
WO2016085549A1 (en) * | 2014-11-24 | 2016-06-02 | Raytheon Comapny | Method and system for processing a sequence of images to identify, track, and/or target an object on a body of water |
CN105869405A (en) * | 2016-05-25 | 2016-08-17 | 银江股份有限公司 | Urban road traffic congestion index calculating method based on checkpoint data |
-
2017
- 2017-12-29 CN CN201711468924.4A patent/CN108122186B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103646187A (en) * | 2013-12-27 | 2014-03-19 | 中国科学院自动化研究所 | Method for obtaining vehicle travel path and OD (Origin-Destination) matrix in statistic period |
WO2016085549A1 (en) * | 2014-11-24 | 2016-06-02 | Raytheon Comapny | Method and system for processing a sequence of images to identify, track, and/or target an object on a body of water |
CN105608528A (en) * | 2015-12-18 | 2016-05-25 | 东南大学 | Taxi driver income-pressure assessment method and system based on big data analysis |
CN105869405A (en) * | 2016-05-25 | 2016-08-17 | 银江股份有限公司 | Urban road traffic congestion index calculating method based on checkpoint data |
Non-Patent Citations (1)
Title |
---|
巧克力工厂的查理: "用户地理位置的聚类算法实现—基于DBSCAN和Kmeans的混合算法", 《HTTPS://BLOG.CSDN.NET/JACKADUMA/ARTICLE/DETAILS/52734731》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108717790A (en) * | 2018-07-06 | 2018-10-30 | 广州市交通运输研究所 | A kind of vehicle driving analysis method based on bayonet license plate identification data |
CN109308804A (en) * | 2018-08-08 | 2019-02-05 | 北京航空航天大学 | Hourage estimation method based on tensor resolution |
CN109308804B (en) * | 2018-08-08 | 2020-05-05 | 北京航空航天大学 | Travel time estimation method based on tensor decomposition |
CN109471907A (en) * | 2018-11-15 | 2019-03-15 | 刘长山 | A kind of driving law-analysing system and method based on bayonet data |
CN109471907B (en) * | 2018-11-15 | 2022-04-29 | 刘长山 | Traffic law analysis system and method based on checkpoint data |
CN111080351A (en) * | 2019-12-05 | 2020-04-28 | 任子行网络技术股份有限公司 | Clustering method and system for multi-dimensional data set |
CN111523562A (en) * | 2020-03-20 | 2020-08-11 | 浙江大学 | Commuting mode vehicle identification method based on license plate identification data |
CN111523562B (en) * | 2020-03-20 | 2021-06-08 | 浙江大学 | Commuting mode vehicle identification method based on license plate identification data |
CN111640303A (en) * | 2020-05-29 | 2020-09-08 | 青岛大学 | City commuting path identification method and equipment |
Also Published As
Publication number | Publication date |
---|---|
CN108122186B (en) | 2021-11-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108122186A (en) | Location estimation method is lived in a kind of duty based on bayonet data | |
WO2020238631A1 (en) | Population type recognition method based on mobile phone signaling data | |
CN110728842B (en) | Abnormal driving early warning method based on reasonable driving range of vehicles at intersection | |
CN108959466B (en) | Taxi passenger carrying hot spot visualization method and system based on BCS-DBSCAN | |
CN105070053B (en) | A kind of intelligent traffic monitoring video camera for recognizing rule-breaking vehicle motor pattern | |
CN105608505B (en) | Resident rail transit trip mode identification method based on mobile phone signaling data | |
CN109359880B (en) | Urban road traffic duty police force deployment method | |
US7697720B2 (en) | Visual sensing for large-scale tracking | |
CN107656987B (en) | Subway station function mining method based on L DA model | |
CN108230724B (en) | Urban public transport system vehicle-mounted station reporting missing data repairing method based on maximum probability estimation | |
CN114416710B (en) | Method and system for extracting OD position of express way vehicle | |
Li et al. | A general framework for unmet demand prediction in on-demand transport services | |
CN101329734A (en) | License plate character recognition method based on K-L transform and LS-SVM | |
He et al. | Inferring high-resolution traffic accident risk maps based on satellite imagery and gps trajectories | |
CN102044146A (en) | Method for reckoning getting-on stops on basis of data of one-ticket public-transport integrated circuit (IC) card | |
US20060062431A1 (en) | Method of producing a transit graph | |
CN106227859A (en) | The method identifying the vehicles from gps data | |
CN106956680A (en) | A kind of electric automobile driving behavior identifying and analyzing method | |
Dey et al. | Turning video into traffic data–an application to urban intersection analysis using transfer learning | |
CN107329977A (en) | A kind of false-trademark car postsearch screening method based on probability distribution | |
CN114372503A (en) | Cluster vehicle motion trail prediction method | |
Hamedmoghadam et al. | Automated extraction of origin-destination demand for public transportation from smartcard data with pattern recognition | |
CN114091581A (en) | Vehicle operation behavior type identification method based on sparse track | |
Zhou et al. | Detecting unmetered taxi rides from trajectory data | |
Shi et al. | A sequential pattern mining based approach to adaptively detect anomalous paths in floating vehicle trajectories |
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 |