CN105355047B - The Data Fusion method of many Vehicle Detection source dynamic time granularities - Google Patents
The Data Fusion method of many Vehicle Detection source dynamic time granularities Download PDFInfo
- Publication number
- CN105355047B CN105355047B CN201510736033.7A CN201510736033A CN105355047B CN 105355047 B CN105355047 B CN 105355047B CN 201510736033 A CN201510736033 A CN 201510736033A CN 105355047 B CN105355047 B CN 105355047B
- Authority
- CN
- China
- Prior art keywords
- data
- detection
- time
- source
- traffic
- 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.)
- Expired - Fee Related
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 156
- 235000019580 granularity Nutrition 0.000 title claims abstract description 47
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 15
- 230000004927 fusion Effects 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 25
- 230000008859 change Effects 0.000 claims abstract description 17
- 238000000605 extraction Methods 0.000 claims abstract description 11
- 230000007613 environmental effect Effects 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000011068 loading method Methods 0.000 claims abstract description 6
- 238000010790 dilution Methods 0.000 claims description 6
- 239000012895 dilution Substances 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 4
- 230000008439 repair process Effects 0.000 claims description 4
- 241001269238 Data Species 0.000 claims description 3
- 241000545442 Radix Species 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 108700041286 delta Proteins 0.000 claims description 3
- 238000007499 fusion processing Methods 0.000 claims description 3
- 230000008676 import Effects 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000003908 quality control method Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 abstract description 2
- 238000007405 data analysis Methods 0.000 abstract description 2
- NJPPVKZQTLUDBO-UHFFFAOYSA-N novaluron Chemical compound C1=C(Cl)C(OC(F)(F)C(OC(F)(F)F)F)=CC=C1NC(=O)NC(=O)C1=C(F)C=CC=C1F NJPPVKZQTLUDBO-UHFFFAOYSA-N 0.000 abstract description 2
- 230000035945 sensitivity Effects 0.000 abstract description 2
- 238000013480 data collection Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000006698 induction Effects 0.000 description 3
- 230000010365 information processing Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000005267 amalgamation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000005389 magnetism Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/012—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from other sources than vehicle or roadside beacons, e.g. mobile networks
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of Data Fusion method of many Vehicle Detection source dynamic time granularities, belong to urban transportation data analysis technique field, it includes being spaced extraction time radix by detection data acquisition time；To detection data loading spatial information and the data structure of generalized treatment traffic data；Data mutuality degree, device data precision and the environmental sensitivity index for evaluating detection source obtain detection source degree of belief coefficient；By position in detection zone mobile terminal register data dynamic change update obtain time granularity application demand；Traffic parameter fusion is carried out to many detection sources according to application demand and detection source degree of belief.The present invention is mutually agreed with the trend that urban transportation data processing develops with big data using the traffic data combination mobile terminal locations data of registering that pedestal sensor is obtained, changing dynamic fusion data according to time granularity can effectively improve speed, have the advantages that acquisition computing is quick, reduce storage redundancy.
Description
Technical field
The invention belongs to urban transportation data analysis technique field, more particularly to a kind of many Vehicle Detection sources dynamic time grain
The Data Fusion method of degree.
Background technology
With the rise of smart city, the various roads sensor such as earth magnetism wagon detector, induction coil, vehicle flowrate radar
The pattern of common gather data in real time is gradually popularized.On the other hand with data collection and data storage ability it is continuous
Enhancing, the particularly mobile terminal client terminal user such as the popularization of individual mobile terminal and microblogging, wechat gradually increases, shifting therein
Dynamic terminal positioning technology provides the data source of magnanimity for traffic data collection.Under the propulsion of these two aspects technology, new
Traffic data collection pattern and traditional data drainage pattern the data space shortage that interweaves lower jointly associate, and cause actual friendship
Logical application aspect is made slow progress the bottleneck as Modern City Traffic data processing.
Traffic data fusion refers to that the traffic data of multisensor is comprehensive under certain criterion, is determined with the traffic needed for completing
Plan and the transport information processing procedure assessed and carry out.The transport information that multiple sensors are gathered is carried out into effective integration, not only
Cost performance, the degree of accuracy and the reliability of acquisition of information can be effectively improved, and single information source can be avoided to fail and cause
Judgement and decision error.According to the position of each Loop detector layout in current domestic communication data fusion method, to including upstream
The roadway element in intersection and section carries out traffic data fusion by certain hour granularity, during such transport information handling and
Storage redundancy is excessive, and cannot ensure practical application effect.
Traditional road traffic fusion method is remained in many deficiencies under new traffic data collection pattern, and its is main
Performance number exists：Time granularity is set as definite value in current data fusion method, expends time and memory space；Information processing is only
Public presupposed information demand aspect is rested on, and cannot be formed with the change of regional traffic customer volume and then dynamic change time grain
Degree carries out traffic data fusion；For the number that the magnanimity location information that individual mobile terminal is produced does not effectively become traffic characteristic
According to foundation.
The content of the invention
In order to time granularity fixation causes consuming time and memory space in overcoming the shortcomings of data fusion, the present invention is provided
A kind of Data Fusion method of many Vehicle Detection source dynamic time granularities.The method is with urban highway traffic detection data
Research object, the time-space attribute according to detection data extracts Uniform data format, is signed by detection zone mobile terminal APP positions
Determine traffic application demand to data dynamic change, and then information source degree of belief coefficient is pressed to the data for detecting source in detection zone more
Traffic parameter fusion is carried out, the flexible analyzing and processing of urban transportation data is realized.
Dynamic time granularity Fusion Module of the present invention comprising Vehicle Detection data generalization processing module with traffic data, should
Method is comprised the following steps that：
(1), Vehicle Detection data generalization treatment
The treatment of Vehicle Detection data generalization is main to realize multi-source ground fundamental mode traffic detector normalization of data structure；Its treatment
Process mainly includes the time interval feature extraction of multi-source detection data and two steps of data structure generalized of traffic data：
I) time interval of gathered data, the extraction time radix on the premise of data precision is ensured are extracted；Examined to obtaining
Temporally radix carries out data quality control to survey device data；
Ii space position parameter) is extracted for each Vehicle Detection source, spatial information loading is carried out to detection data, and it is right
The generalized data structure of traffic data；
According to above-mentioned basic implementing procedure, the Vehicle Detection data generalization module specific implementation is comprised the following steps：
Step 1, traffic parameter is extracted to data record, and by Import data records database；
Step 2, record adjacent to same detection source historical data extract the acquisition time interval of data；
Step 3, the average of the desired value for extracting multi-source detection data acquisition time interval are time radix, i.e., in the time period
[ts, te] in, data record m kind time interval extraction time radixes tbFor Wherein pikIt is i-th
K-th acquisition time interval of delta t in detection sourcekThe probability of appearance, riFor the gathered data amount in i-th detection source accounts for all detections
The ratio of data, n is detection source number in region.
Step 4, with time radix be standard adjustment record time interval, and by after adjustment time interval repair missing data
Record
The spatial information of step 5, over the ground fundamental mode detector, obtain traffic flow parameter unified field DLatitude,
DLongitude, LaneNo, TimeStamp, Flow, Speed, Occ, wherein DLatitude are detector location latitude,
DLongitude is detector location longitude, and LaneNo is numbered for road track, and TimeStamp is detection time label, Flow
It is the magnitude of traffic flow that detection is obtained, Speed is the vehicle location speed that detection is obtained, and Occ is the time occupancy that detection is obtained.
According to Vehicle Detection data generalization of the present invention process particular technique embodiment, the traffic detector data when
Between spaced features be for analysis collection Vehicle Detection data reliability and accuracy, correspondence obtain data time interval category
Property, specifically include：1) acquisition time interval：It refer to the time interval in traffic detector detection data acquisition time；2) time
Radix：It refer to the desired value of Vehicle Detection data collection interval；3) time granularity：Refer to multi-source traffic data fusion when
Between interval setting, the time granularity of this method depends on the data volume change of registering of mobile terminal APP positions in corresponding region and determines
Transport need.
(2), the dynamic time granularity fusion of traffic data
The dynamic time granularity fusion part of traffic data mainly includes that detection source Trust Values Asses module, position are registered number
According to three parts of traffic parameter Fusion Module for updating statistical module and dynamic time granularity.Wherein detection source degree of belief is commented
Estimate module and position registers that to update statistical module be the forerunner of the traffic parameter Fusion Module of dynamic time granularity, Liang Zhewei to data
Implementation relation in parallel：
I) source Trust Values Asses module is detected：To detect the historical data degree of correlation factor in source, device data dilution of precision,
The aspect qualitative assessment such as Environmental Factors, comprehensively obtains detection source Trust Values Asses coefficient；
Ii) position register data update statistical module：Mobile terminal APP positions are registered data filtering, reject traffic ginseng
Record and repeat to record with person's positional information abnormity, and travel through mobile terminal APP in region and register position data, statistics dynamic becomes
Change and update；
Iii) the traffic parameter Fusion Module of dynamic time granularity：By position register data variation obtain time granularity should
With demand, traffic parameter fusion is carried out to many detection sources according to application demand and detection source degree of belief.
Basic implementing procedure of the invention, the detection source Trust Values Asses module specific implementation includes following step
Suddenly：
Step 1, section n Data Detection source to be fused is determined, to i-th detection source data xiCalculate and j-th inspection
Survey source data xjThe discrete data correlation coefficient r in one dayijFor
Wherein xik
It is k-th data of time radix, x in i-th detection source data one dayjkIt is k-th time in j-th detection source data one day
The data of radix.
Step 2, the Data mutuality degree factor C for detecting source for i-th according to calculating in detection source discrete data coefficient correlationiForWherein rijIt is i-th detection source data xiCalculate and j-th detection source data xjCoefficient correlation.
Step 3, to i-th detection source computing device data precision factors Ai, andWherein aiIt is such detection source
Innate detection precision.
Step 4, according to i-th detection source characteristic assignment Environmental Factors Si, and
Step 5, calculate i-th independent detection source Trust Values Asses coefficient ωi, andWherein CiIt is i-th
The detection source data degree of correlation factor, AiIt is i-th device data dilution of precision in detection source, SiIt is i-th environment shadow in detection source
Ring the factor.
According to Trust Values Asses module particular technique embodiment in detection source of the present invention, detection source Trust Values Asses are
To including detecting the transport information that source obtains same section, from Data mutuality degree, three sides of equipment accuracy and environmental sensitivity index more
The quantitative overall merit in face detects the degree of belief in source, is embodied as including：1) the multi-source data degree of correlation factor：Refer to that certain section is treated
The related proportionality coefficient of other class detector datas of the data that the traffic detector of evaluation is measured and same section；2) device data precision
The factor：It refer to the proportionality coefficient of certain Vehicle Detection equipment innate detection precision and all kinds of detector precision ratios；3) ambient influnence
The factor：It refer to the influence proportionality coefficient of certain traffic detector testing result under conditions of environmental change.
Basic implementing procedure of the invention, the position register data update statistical module specific implementation include it is following
Step：
Step 1, information data of being registered to mobile terminal APP positions filtering, conversion and loading, reject traffic participant position
Information abnormity records and repeats to record.To obtain unified field Venue id, Venue of traffic participant positional information
Name, Latitude, Longitude, Checked UserID, wherein Venue id are to register data number, Venue position
Name is register position longitude, Checked to register location name, Latitude to register position latitude, Longitude
UserID for position register data source mobile terminal numbering；
Step 2, all adjacent detection intersections are linked to be triangle, make the perpendicular bisector on each side of triangle, then
Some perpendicular bisectors around each intersection surround Thiessen polygon.Unique intersection is encompassed within unique Tyson
In polygon.Initialization δ is 1, and wherein δ is the time radix multiple in dynamic time cycle.
Step 3, [the t-2 δ * t before current time tb,t-δ*tb] interval and [t- δ * tb, t] and it is interval interior, judge that position is marked
In polygonal region belonging to signing, and count position in each region and register the number m of data0And m1。
Step 4, with current statistic cycle [t- δ * tb, t] and measurement period [t-2 δ * tb,t-δ*tb] in position register number
According to variation delta m=| m1-m0| with upper limit threshold Mmax, lower threshold MminCompare；If Δ m<Mmin, then δ=δ+1 simultaneously repeats to walk
Rapid 3；If Δ m>Mmax, then δ=δ -1 and repeat step 3；If Δ m ∈ [Mmin,Mmax], it is determined that dynamic time granularity is δ times
Time radix.
According to position of the present invention register data update statistical technique embodiment, the filtering refer to traffic participant position
Confidence breath rejecting abnormalities record and repetition are recorded.
Basic implementing procedure of the invention, the traffic parameter Fusion Module specific implementation bag of the dynamic time granularity
Include following steps：
Section is in dynamic time cycle [t, t+ δ * t in step 1, same detection zone Thiessen polygonb] in n detector
Weight matrix W={ the ω of degree of belief1,ω2...ωn}T, wherein ωiRepresent i-th detector data measured in fusion process
Weight ratio.
Step 2, in dynamic time cycle [t, t+ δ * tb] interior to i-th common n in detection sourceiIndividual data calculate i-th it is only
The information fusion in vertical detection sourceWherein xaIt is the traffic data of i-th detector acquisition to be fused.
Step 3, in dynamic time cycle [t, t+ δ * tb] in the information fusion knot that obtains of various traffic information collecting methods
It is really：WhereinI-th information fusion result in independent detection source.
Beneficial effects of the present invention：
1) traffic data combination mobile terminal APP positions are obtained using pedestal sensor and registers data to urban transportation data
The trend developed with big data is processed mutually to agree with.
2) class statistic that mobile terminal APP registers position, is existed by position number change of registering to time granularity demand
Line is divided.
3) changing dynamic data combining according to time granularity can effectively improve speed, with obtaining that computing is quick, reduce and deposit
Store up the advantage of redundancy.
Brief description of the drawings
Fig. 1 is flow chart of the invention.
Fig. 2 is many source detection apparatus acquisition interval characteristic patterns of the invention.
Fig. 3 is intersection Thiessen polygon organigram of the invention.
Fig. 4 is flow chart of the judgement location tags of the invention in Thiessen polygon region.
Specific embodiment
Based on the fundamental mode traffic detector data by multi-source ground of the invention, the mobile terminal that additional transport participant participates in
APP dynamic positions are registered data, it is proposed that a kind of to be dynamically determined the section multi-source traffic number of time granularity with position data of registering
According to method for amalgamation processing.Below in conjunction with accompanying drawing and embodiment explanation technical scheme.
As shown in figure 1, the flow of embodiment specifically includes following steps：
Step 1：Extract example section geomagnetism detecting device, induction coil detector and video monitor acquisition time interval simultaneously
Extraction time radix.From example section geomagnetism detecting device (DE537919498), video monitor (DC00004838) and the line of induction
Circle detector (COIL1518017) is in time period [ts, te] in various detectors historical data time interval feature, such as Fig. 2 institutes
Show.
The flow of the step extraction time radix specifically includes following steps：
1. traffic parameter is extracted to multi-source data record and imports database, and adjust data field DetectorID,
LaneNo, TimeStamp, Flow, Speed, Occ, wherein DetectorID are numbered for detector, and LaneNo is compiled for road track
Number, TimeStamp is detection time label, and Flow is the magnitude of traffic flow that detection is obtained, and Speed is the vehicle location that detection is obtained
Speed, Occ is the time occupancy that detection is obtained.
2. to historgraphic data recording extraction time sequence, then traffic data time series the finite aggregate { (t for gathering1,o1),
(t2,o2),…,(tn,on) meet ts≤ti≤teAnd ti<ti+1(i=1,2 ..., n-1), oiIt is correspondence time tiTraffic parameter
Collection includes magnitude of traffic flow Flow, vehicle location speed Speed, time occupancy Occ, calculates the time period [ts, te] gathered data
Time interval { Δ tk|Δtk=ti+1-tiAnd i=1,2 ..., n-1 }；
3. multi-source detection data acquisition time interlude radix is calculated, i.e., in time period [ts, te] in, data record
M kind time interval extraction time radixes tbForWherein pikIt is k-th collection in i-th detection source
Time interval Δ tkThe probability of appearance, riThe ratio of all detection datas is accounted for for the gathered data amount in i-th detection source, n is region
Interior detection source number.
Step 2, by various regions fundamental mode Vehicle Detection source extract space position parameter DLatitude, DLongitude, to detection
Data carry out spatial information loading obtain unified field DetectorID, DLatitude of traffic data, DLongitude,
LaneNo, TimeStamp, Flow, Speed, Occ, wherein DLatitude are detector location latitude, and DLongitude is inspection
Survey device position longitude.
Singly detect that data quality control flow of the source data based on time radix specifically includes following steps in the step：
If 1. uniqueThen data (ti, oi) be adapted to (oi)；
If 2. multipleThen k evidence is set to data (ti_1, oi_1), (ti_2,
oi_2) ..., (ti_k, oi_k), dataObtain generalized data
If 3. there is no (ti, oi)∈(nTb,(n+1)tb], then traffic data has continuity in time-domain, thus utilizes
Front and rear two time cycle carries out the translation of time cycle data and repairs.If front and rear cycle data is set to data (ti-2, oi-2),
(ti-1, oi-1) and (ti+1, oi+1), (ti+2, oi+2), dataObtain generalized data
Step 3, to being related to the historical data degree of correlation factor, device data dilution of precision, the Environmental Factors in detection source
Etc. aspect qualitative assessment, comprehensively obtain detection source Trust Values Asses coefficient；
Step detection source Trust Values Asses flow specifically includes following steps：
1. section n Data Detection source to be fused is determined, to i-th detection source data xiCalculate and j-th detection source
Data xjThe discrete data correlation coefficient r in one dayijFor
Wherein xik
It is k-th data of time radix, x in i-th detection source data one dayjkIt is k-th time in j-th detection source data one day
The data of radix.
2. the Data mutuality degree factor according to i-th detection source of calculating in detection source discrete data coefficient correlation isWherein rijIt is i-th detection source data xiCalculate and j-th detection source data xjCoefficient correlation.
3. to i-th detection source computing device data precision factorWherein aiIt is such detection source innate detection
Precision.
4. according to i-th detection source characteristic assignment Environmental Factors SI,And
5. i-th independent detection source Trust Values Asses coefficient ω is calculatediForWherein CiIt is i-th detection
The source data degree of correlation factor, AiIt is i-th device data dilution of precision in detection source, SiFor i-th detection source ambient influnence because
Son.
Step 4, information data of being registered to mobile terminal APP positions conversion and filtering, and to mobile terminal in detection zone
APP registers position data dynamic change update；
The step specifically includes following steps：
1. check position register data category information it is whether complete, the position data of registering for lacking attribute are deleted
Remove, and to data according to criteria field Venue id, Venue name, Latitude, Longitude, Checked UserID
Form is modified, and it is register location name, Latitude that wherein Venue id register data number, Venue name for position
For position latitude of registering, Longitude are to register position longitude, Checked UserID for data source movement is registered eventually in position
End numbering；
2. data of being registered to the repeatable position of identical Checked UserID carry out data merging treatment, reduce the superfluous of data
Remaining.
3. all adjacent detection intersections are linked to be triangle, make the perpendicular bisector on each side of these triangles, then
Some perpendicular bisectors around each intersection just surround Thiessen polygon, as shown in Figure 3.Unique intersection is forgiven
In unique Thiessen polygon, and initialization time radix coefficient δ is 1.
4. a large amount of discrete location points of registering are collected in polygon, the δ times of time with time radix as measurement period,
[t-2 δ * t before current time tb,t-δ*tb] interval and [t- δ * tb, t] and it is interval interior, judged belonging to location tags with scanning Beam Method
Polygonal region, and count position and register the number m of data0And m1, flow is as shown in Figure 4.
5. with current statistic cycle [t- δ * tb, t] and measurement period [t-2 δ * tb,t-δ*tb] in position register data change
Change amount Δ m=| m1-m0| with upper limit threshold Mmax, lower threshold MminCompare；If Δ m<Mmin, then 4. δ=δ+1 simultaneously repeats；If
Δm>Mmax, then 4. δ=δ -1 simultaneously repeats；If Δ m ∈ [Mmin,Mmax], it is determined that dynamic time granularity is δ times of time base
Number.
Step 5, change to attributes of being registered by position change are obtained to time granularity demand, to obtain the unification of traffic data
Field DLatitude, DLongitude, LaneNo, TimeStamp, Flow, Speed, Occ, according to time granularity application demand
Many detection sources are carried out with the fusion of traffic parameter field data.
Time granularity application demand specifically includes following to the fusion flow that many detection sources carry out traffic data in the step
Step：
1. in same detection zone Thiessen polygon section in dynamic time cycle [t, t+ δ * tb] in n detector measure
Weight matrix W={ the ω of data1,ω2...ωn}T, wherein ωiRepresent that i-th detector measures traffic parameter field data xi
The shared weight in fusion process.
2. in dynamic time cycle [t, t+ δ * tb] interior to i-th common n in detection sourceiIndividual data calculate i-th independent detection
The information fusion in sourceWherein xaIt is the traffic data of i-th detector acquisition to be fused.
4. in dynamic time cycle [t, t+ δ * tb] in the information fusion result that obtains of various traffic information collecting methods be：WhereinI-th information fusion result in independent detection source.
General principle of the invention be by the use of position register data as road traffic demand dynamic update foundation, enter
And meet the requirement of novel traffic data acquisition, wherein to the number in many Vehicle Detection sources on the basis of detection source degree of belief is evaluated
According to data processing is carried out under dynamic change time granularity, so as to realize the data dynamic fusion to many Vehicle Detection sources.
Claims (7)
1. a kind of Data Fusion method of many Vehicle Detection source dynamic time granularities, is processed comprising Vehicle Detection data generalization
Two parts are merged with the dynamic time granularity of traffic data, the method is comprised the following steps：
(1), Vehicle Detection data generalization treatment
Vehicle Detection data generalization processing procedure mainly includes time interval feature extraction and the traffic data of multi-source detection data
Two steps of structure generalized；The time interval of gathered data, the extraction time base on the premise of data precision is ensured are extracted first
Number；To obtaining detector data, temporally radix carries out data quality control；Space bit is extracted to each Vehicle Detection source afterwards
Parameter is put, spatial information loading is carried out to detection data and is obtained the generalized data structure of traffic data, and then by data after generalized
For the dynamic time granularity fusion of highway traffic data；
(2), the dynamic time granularity fusion of traffic data
The dynamic time granularity fusion part of traffic data mainly includes that detection source Trust Values Asses module, position register data more
New statistical module and dynamic time granularity traffic parameter three parts of Fusion Module；
I) Trust Values Asses module：To detect the historical data degree of correlation factor in source, device data dilution of precision, ambient influnence because
Sub- qualitative assessment, comprehensively obtains detection source Trust Values Asses coefficient；
Ii) position register data update statistical module：Mobile terminal APP positions are registered data filtering, reject traffic participant
Positional information abnormity records and repeats to record, and travels through mobile terminal APP in region and register position data, and statistics dynamic change is more
Newly；
Iii) dynamic time granularity traffic parameter Fusion Module：By position register data variation obtain time granularity application need
Ask, traffic parameter fusion is carried out to many detection sources according to application demand and detector degree of belief.
2. the Data Fusion method of many Vehicle Detection source dynamic time granularities according to claim 1, its feature exists
In：The time interval feature extracting method of multi-source detection data is comprised the following steps in the Vehicle Detection data generalization treatment：
Step 1, traffic parameter is extracted to data record, and by Import data records database；
Step 2, record adjacent to same detection source historical data extract the acquisition time interval of data；
Step 3, the average of the desired value for extracting multi-source detection data acquisition time interval are time radix, i.e., in time period [ts,
te] in, data record m kind time interval extraction time radixes tbFor Wherein pikIt is i-th detection source
In k-th acquisition time interval of delta tkThe probability of appearance, riFor the gathered data amount in i-th detection source accounts for all detection datas
Ratio, n is detection source number in region；
Step 4, with time radix be standard adjustment record time interval, and by after adjustment time interval repair missing data remember
Record；
The spatial information of step 5, over the ground fundamental mode detector, obtains the unified field of traffic flow parameter.
3. the Data Fusion method of many Vehicle Detection source dynamic time granularities according to claim 1, its feature exists
In：Data structure generalized in the Vehicle Detection data generalization process part is to adjust data record by standard of time radix
Time simultaneously repairs missing data record, and the spatial information of detector is loaded into detection data the unified word for obtaining traffic flow parameter
Section.
4. the Data Fusion method of many Vehicle Detection source dynamic time granularities according to claim 1, its feature exists
In：Detection source Trust Values Asses module and position are registered data renewal in the dynamic time granularity Fusion Module of the traffic data
Statistical module parallel processing, for the traffic parameter Fusion Module of dynamic time granularity provides foundation.
5. the Data Fusion method of many Vehicle Detection source dynamic time granularities according to claim 1, its feature exists
In：Detect that the method for source Trust Values Asses module includes following step in the dynamic time granularity fusion part of the traffic data
Suddenly：
Step 1, section n Data Detection source to be fused is determined, to i-th detection source data xiCalculate and j-th detection source number
According to xjThe discrete data correlation coefficient r in one dayijFor
Wherein xikIt is i-th
K-th data of time radix, x in individual detection source data one dayjkIt is k-th time radix in j-th detection source data one day
Data；
Step 2, the Data mutuality degree factor C for detecting source for i-th according to calculating in detection source discrete data coefficient correlationiFor
Wherein rijIt is i-th detection source data xiCalculate and j-th detection source data xjCoefficient correlation；
Step 3, to i-th detection source computing device data precision factors Ai, andWherein aiFor i-th detection source is consolidated
There is accuracy of detection；
Step 4, according to i-th detection source characteristic assignment Environmental Factors Si, and
Step 5, calculate i-th independent detection source Trust Values Asses coefficient ωi, andWherein CiIt is i-th detection
The source data degree of correlation factor, AiIt is i-th device data dilution of precision in detection source, SiFor i-th detection source ambient influnence because
Son.
6. the Data Fusion method of many Vehicle Detection source dynamic time granularities according to claim 1, its feature exists
In：The position register data update statistical module method comprise the following steps：
Step 1, information data of being registered to mobile terminal APP positions filtering, conversion and loading, reject traffic participant positional information
Exception record and repetition are recorded, to obtain the unified field of traffic participant positional information；
Step 2, all adjacent detection intersections are linked to be triangle, make the perpendicular bisector on each side of these triangles, then
Some perpendicular bisectors around each intersection just surround Thiessen polygon；Unique intersection is encompassed within unique Thailand
Gloomy polygon, and it is 1 to initialize δ, wherein δ is the time radix coefficient for determining the time cycle；
Step 3, [the t-2 δ * t before current time tb,t-δ*tb] interval and [t- δ * tb, t] and interval interior, wherein tbIt is what is extracted
Time radix, judges the polygonal region belonging to location tags, and counts position in each region and register the number m of data0With
m1；
Step 4, with current statistic cycle [t- δ * tb, t] and measurement period [t-2 δ * tb,t-δ*tb] in position register data change
Change amount Δ m=| m1-m0| with upper limit threshold Mmax, lower threshold MminCompare；If Δ m<Mmin, then δ=δ+1 and repeat step 3；
If Δ m>Mmax, then δ=δ -1 and repeat step 3；If Δ m ∈ [Mmin,Mmax], it is determined that dynamic time granularity for δ times when
Between radix.
7. the Data Fusion method of many Vehicle Detection source dynamic time granularities according to claim 1, its feature exists
In：The traffic parameter Fusion Module of the dynamic time granularity determines according to the evaluation weight matrix of each detector in detection zone
Shared weight in data fusion process, by position register data update statistical module determine the time cycle, it is desirable to by same inspection
The traffic data of many discrete time points collection in polytype detection source in region time period is surveyed by evaluating weights than weighting
Mode carries out data fusion, and then obtains more accurate data result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510736033.7A CN105355047B (en) | 2015-11-03 | 2015-11-03 | The Data Fusion method of many Vehicle Detection source dynamic time granularities |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510736033.7A CN105355047B (en) | 2015-11-03 | 2015-11-03 | The Data Fusion method of many Vehicle Detection source dynamic time granularities |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105355047A CN105355047A (en) | 2016-02-24 |
| CN105355047B true CN105355047B (en) | 2017-06-06 |
Family
ID=55331011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510736033.7A Expired - Fee Related CN105355047B (en) | 2015-11-03 | 2015-11-03 | The Data Fusion method of many Vehicle Detection source dynamic time granularities |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105355047B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105792110B (en) * | 2016-03-30 | 2019-04-30 | 上海申腾信息技术有限公司 | A kind of data fusion of multiple data origin, the processing method of intelligent search |
| CN106710238B (en) * | 2017-01-26 | 2023-09-05 | 深圳市迅朗科技有限公司 | Method for improving accuracy of geomagnetic vehicle detector |
| CN110111563A (en) * | 2019-04-08 | 2019-08-09 | 东南大学 | A kind of real-time traffic states estimation method of city expressway |
| CN110322688A (en) * | 2019-05-20 | 2019-10-11 | 华为技术有限公司 | A kind of method of data processing, the method for data query and relevant device |
| CN110133685B (en) * | 2019-05-22 | 2023-04-14 | 吉林大学 | OCC-based street lamp auxiliary mobile phone detailed positioning communication system |
| CN114333301B (en) * | 2021-12-06 | 2023-09-26 | 北京东土正创科技有限公司 | Traffic signal control optimization method, system and traffic signal optimization equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493991A (en) * | 2009-02-20 | 2009-07-29 | 北京交通大学 | Method and device for obtaining traffic status based on sensor network |
| CN101710361A (en) * | 2009-11-13 | 2010-05-19 | 北京航空航天大学 | Distributed traffic simulating platform based on Service-Oriented Architecture (SOA) and simulating method thereof |
| CN101930668A (en) * | 2009-04-29 | 2010-12-29 | 上海电器科学研究所（集团）有限公司 | Road traffic OD (Optical Density) information collection system for license plate recognition and processing method thereof |
| CN102074126A (en) * | 2011-01-19 | 2011-05-25 | 清华大学 | Historical radar data based air traffic flow analysis system and method |
| CN102254001A (en) * | 2011-07-14 | 2011-11-23 | 青岛海信网络科技股份有限公司 | Efficient data management method and system |
| CN103220340A (en) * | 2013-03-28 | 2013-07-24 | 北京交通大学 | Semantic coding method for road transport sensor network nodes |
| CN103456163A (en) * | 2012-06-04 | 2013-12-18 | 北京市市政工程设计研究总院 | City expressway interchange traffic capacity and running status discrimination method and system |
| CN104765974A (en) * | 2015-04-24 | 2015-07-08 | 北京城建设计发展集团股份有限公司 | Urban rail transit passenger flow density index calculating method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7046166B2 (en) * | 2003-04-29 | 2006-05-16 | Rockwell Scientific Licensing, Llc | Modular wireless integrated network sensor (WINS) node with a dual bus architecture |
| US20060095518A1 (en) * | 2004-10-20 | 2006-05-04 | Davis Jesse H Z | Software application for modular sensor network node |
-
2015
- 2015-11-03 CN CN201510736033.7A patent/CN105355047B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493991A (en) * | 2009-02-20 | 2009-07-29 | 北京交通大学 | Method and device for obtaining traffic status based on sensor network |
| CN101930668A (en) * | 2009-04-29 | 2010-12-29 | 上海电器科学研究所（集团）有限公司 | Road traffic OD (Optical Density) information collection system for license plate recognition and processing method thereof |
| CN101710361A (en) * | 2009-11-13 | 2010-05-19 | 北京航空航天大学 | Distributed traffic simulating platform based on Service-Oriented Architecture (SOA) and simulating method thereof |
| CN102074126A (en) * | 2011-01-19 | 2011-05-25 | 清华大学 | Historical radar data based air traffic flow analysis system and method |
| CN102254001A (en) * | 2011-07-14 | 2011-11-23 | 青岛海信网络科技股份有限公司 | Efficient data management method and system |
| CN103456163A (en) * | 2012-06-04 | 2013-12-18 | 北京市市政工程设计研究总院 | City expressway interchange traffic capacity and running status discrimination method and system |
| CN103220340A (en) * | 2013-03-28 | 2013-07-24 | 北京交通大学 | Semantic coding method for road transport sensor network nodes |
| CN104765974A (en) * | 2015-04-24 | 2015-07-08 | 北京城建设计发展集团股份有限公司 | Urban rail transit passenger flow density index calculating method |
Non-Patent Citations (3)
| Title |
|---|
| "城市交通控制与诱导***协同的信息分析与组织研究";陈昕 等;《公路交通科技》;20070228;正文全文 * |
| "基于视频技术的微观交通流信息采集***研究";张汝华 等;《交通与计算机》;20040630;正文全文 * |
| "空间区域中对象流动模式构建方法研究";刘俊岭 等;《空计算机工程与科学》;20151031;正文全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105355047A (en) | 2016-02-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105355047B (en) | The Data Fusion method of many Vehicle Detection source dynamic time granularities | |
| Zhang et al. | A methodology for obtaining spatiotemporal information of the vehicles on bridges based on computer vision | |
| Biljecki et al. | Transportation mode-based segmentation and classification of movement trajectories | |
| CN104330089B (en) | A kind of method that map match is carried out using history gps data | |
| CN104134349B (en) | A kind of public transport road conditions disposal system based on traffic multisource data fusion and method | |
| CN109544932A (en) | A kind of city road network flow estimation method based on GPS data from taxi Yu bayonet data fusion | |
| CN108629978A (en) | A kind of traffic trajectory predictions method based on higher-dimension road network and Recognition with Recurrent Neural Network | |
| CN110505583A (en) | A kind of path matching algorithm based on bayonet data and signaling data | |
| CN107958031B (en) | Resident travel OD distribution extraction method based on fusion data | |
| Sharman et al. | Analysis of freight global positioning system data: Clustering approach for identifying trip destinations | |
| CN106446208A (en) | Method for identifying traveling mode of smart phone by considering integrating degree of road network | |
| CN105206057B (en) | Detection method and system based on Floating Car resident trip hot spot region | |
| CN107730115A (en) | A kind of method for evaluating quality of the multi-source location track data based on AHP | |
| CN107767669A (en) | Public bus network passenger flow OD methods of estimation based on WiFi and bluetooth recognition | |
| CN105738587A (en) | Water quality monitoring system | |
| CN110852243B (en) | Road intersection detection method and device based on improved YOLOv3 | |
| CN106845559A (en) | Take the ground mulching verification method and system of POI data special heterogeneity into account | |
| CN112000755B (en) | Regional travel corridor identification method based on mobile phone signaling data | |
| CN108154387B (en) | Method and device for evaluating bus body advertisement delivery route scheme | |
| CN105844038A (en) | Combination optimized layout method for multiple types of highway traffic detectors | |
| CN105844031B (en) | A kind of urban transportation gallery recognition methods based on mobile phone location data | |
| CN107368480A (en) | A kind of interest point data type of error positioning, repeat recognition methods and device | |
| CN110413855A (en) | A kind of region entrance Dynamic Extraction method based on taxi drop-off point | |
| CN107301649A (en) | A kind of region merging technique SAR image coastline Detection Method algorithm based on super-pixel | |
| CN107230350A (en) | A kind of urban transportation amount acquisition methods based on bayonet socket Yu mobile phone flow call bill data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170606 Termination date: 20191103 |