CN108734008A - Anonymous method is gone to the mobile trajectory data of vehicle anonymity based on parking record - Google Patents
Anonymous method is gone to the mobile trajectory data of vehicle anonymity based on parking record Download PDFInfo
- Publication number
- CN108734008A CN108734008A CN201810385918.0A CN201810385918A CN108734008A CN 108734008 A CN108734008 A CN 108734008A CN 201810385918 A CN201810385918 A CN 201810385918A CN 108734008 A CN108734008 A CN 108734008A
- Authority
- CN
- China
- Prior art keywords
- stop
- track
- parking
- anonymous
- vehicle
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
- G06F21/55—Detecting local intrusion or implementing counter-measures
Abstract
Anonymous method removes the mobile trajectory data of vehicle anonymity based on parking record the present invention relates to a kind of, includes the following steps：All stops in track are extracted, are stopped caused by removing on-board and off-board, parking is divided into manned parking stops with unloaded；The abnormal point of manned parking is removed, and manned parking is modeled；Waiting traffic lights in unloaded stop and parking of blocking up are removed with gained model, retains interest stop；Feature vector based on TF-IDF is constructed to interest stop, by the feature vector of known trajectory with the feature vector of anonymous track compared with pair, find out in anonymous track set with the matched anonymity track of known trajectory.The present invention, which realizes to concentrate from the motion track of anonymity by stop record, match not anonymous initial trace, verify the stability and uniqueness of single driver's interest parking point feature, and disclose anonymous track data security risk.
Description
Technical field
The present invention relates to the personal secrets technical fields of anonymous motion track, more particularly to one kind based on parking record pair
The mobile trajectory data of vehicle anonymity goes anonymous method.
Background technology
With the development of mobile terminal and location technology, the mobile trajectory data of the vehicles can be easily obtained, and
These motion tracks generally comprise abundant space time information, and reasonably being excavated and being analyzed using these data can be had
The information of value.Simultaneously because can include personal privacy information in motion track, malicious attacker can pass through these movements
Track deduces its all kinds of interested privacy event, to cause personal secrets problem.
In order to protect the privacy of track data, generally associated privacy protection technique can be used to rail before the data publication of track
Mark data prediction.Relatively common at present has two major classes, one is modification initial trace, the essence in reducing track in sky
Degree, for example reduce the resolution ratio of recording track or be inserted into noise in track, to protect privacy.But data distortion can be caused tight
Weight, availability are low；For another kind using the method handled track anonymization, it is by (with uniqueness random with assumed name
Indications) substitute participant true identity, make the true identity of participant can not be associated with assumed name in any manner.This
Kind method is easy to implement, and privacy is protected in the case where keeping low overhead, not changing former data, keeps data maximum available
Property, it therefore is widely adopted.
Attacker can be identified from anonymous track by the track of attacker by bypass message.Assuming that attacker can be with
Some anonymous track set is accessed, including the track of its target of attack.The attack method needs to obtain anonymous track generation
In period, several space-times " snapshot " (i.e. the location information of vehicle at a certain moment) of target of attack vehicle, thus from anonymous rail
Mark concentration is identified with obtained space-time " snapshot " to the track met.Such attack is stronger to the space-time limitation of " snapshot ", to
Limit the implementation of anonymous attack.However, with the presence or absence of more generally attack pattern, need not obtain and anonymous track data
Survive time equitant space time information, you can successful implementation goes anonymous attack to there is no conclusion.
Invention content
Technical problem to be solved by the invention is to provide a kind of based on parking record to the motion track number of vehicle anonymity
It verifies single driver's interest according to anonymous method, analysis interest parking point feature caused by driver personal preference is gone and stops
The stability of vehicle point feature and uniqueness, and disclose anonymous track data security risk.
The technical solution adopted by the present invention to solve the technical problems is：One kind is provided based on parking record to vehicle anonymity
Mobile trajectory data go anonymous method, several motion track segments of a certain taxi any time period are obtained, using holding
There is track to identify the track of the taxi compared with the track that anonymous track is concentrated, includes the following steps：
(1) all stops in extraction track, stop caused by removing on-board and off-board, and parking is divided into manned parking and sky
It carries and stops, wherein manned parking includes parking caused by parking caused by equal traffic lights and traffic congestion；
(2) abnormal point of manned parking is removed, and manned parking is modeled；
(3) equal traffic lights and traffic congestion in the unloaded stop of gained model removal are used to stop, reservation interest stop；
(4) feature vector based on TF-IDF is constructed to interest stop, by the feature vector of known trajectory and anonymous rail
The feature vector of mark compared to pair, find out in the set of anonymous track with the matched anonymous track of known trajectory.
The step (1) is specially：The point that screening out position does not change over time is concentrated from track, regards it as stop,
Whether it is on-board and off-board parking according to the manned condition adjudgement before and after the parking for hiring out wheel paths concentration.
Indicate that down time, dist indicate stop to the distance at crossing, inter expressions with dura in the step (2)
The interval time of two stops, construction feature vector Stp=(dura, dist, inter) removes manned stop using mahalanobis distance
The abnormal point of vehicle.
The abnormal point that manned parking is removed using mahalanobis distance is specially：Set the threshold value of mahalanobis distance asWherein, p is degree of freedom, and the stop more than threshold value is considered as abnormal point；The calculation formula of mahalanobis distance is：Wherein StpiIndicate that the feature vector of i-th of stop, μ indicate one group of stop
The feature vector mean value of set；V then indicates the covariance matrix of stop set feature vector.
One-classSVM graders are trained with the manned stop after removal exception parking point in the step (3), are come
Distinguish whether stop is traffic congestion and waits the stop of traffic lights；It is identified and is removed in unloaded stop with grader and blocked up and wait
The stop of traffic lights, reservation is interest stop.
The step (4) is specially：Regard the track that taxi travels as a text, section number regards each word as
Language, it is exactly number that word occurs in the text that the parking in section, which is counted out,；Weighted value tf-idfI, jCalculation formula be：tf-
idfI, j=tfI, j×idfI, j, whereinNiFor vehicle v in training datai
The section sum passed through；mjInclude section r for training datajTrack number；CiFor vehicle v in training dataiTrack belonging to
One kind；mI, jFor CiIn include section rjTrack number；tI, jFor rjIn CiThe number of middle appearance；tfI, jFor rjIn viTrack in
The frequency of appearance；The stop extracted in training data constitutes track characteristic, and the stop in test data constitutes test rail
Mark feature is matched with complementation string similarity value, by measure two inner product of vectors spaces angle cosine value come degree
Measure the similitude between them；It is identical to determine whether two vectors are pointed generally in from the cosine value of the angle between two vectors
Direction.
Advantageous effect
Due to the adoption of the above technical solution, compared with prior art, the present invention having the following advantages that and actively imitating
Fruit：The present invention, which realizes to concentrate from anonymous motion track by stop record, matches not anonymous initial trace,
Analysis interest parking point feature caused by driver personal preference verifies the stability of single driver's interest parking point feature
With uniqueness, and anonymous track data security risk is disclosed.
Description of the drawings
Fig. 1 is the flow chart for removing anonymous methods to hiring out wheel paths based on stop；
Fig. 2 is the exhaustive division figure of taxi stop；
Fig. 3 is the stop feature vector distribution map of Shanghai taxi；
Fig. 4 is the stop feature vector distribution map of Shenzhen taxi；
Fig. 5 is the result figure that Shanghai taxi empty stop passes through svm classifier；
Fig. 6 is the result figure that Shenzhen taxi empty stop passes through svm classifier；
Fig. 7 is the histogram of taxi path matching accuracy rate；
Fig. 8 is that test trails length changes the line chart influenced on matching accuracy rate；
Fig. 9 is that test trails length changes the result figure influenced on matching accuracy rate.
Specific implementation mode
Present invention will be further explained below with reference to specific examples.It should be understood that these embodiments are merely to illustrate the present invention
Rather than it limits the scope of the invention.In addition, it should also be understood that, after reading the content taught by the present invention, people in the art
Member can make various changes or modifications the present invention, and such equivalent forms equally fall within the application the appended claims and limited
Range.
Embodiments of the present invention are related to a kind of anonymous motion track progress stopped and recorded to taxi according to taxi
Anonymous method is gone, is stopped with unloaded as shown in Figure 1, stopping including removal on-board and off-board and dividing manned parking；Remove manned stop
The abnormal point of vehicle models manned parking；The stop of unloaded parking medium traffic lamp and traffic congestion is removed according to model and is retained
Interest stop；It constructs the feature vector of interest stop and carries out path matching, wherein：
It stops except on-board and off-board and divides manned parking and stop with unloaded, the type for distinguishing stop, it is to going out
The preparation that stop of hiring a car is classified.
The abnormal point for removing manned parking, the stop for obtaining equal traffic lights and traffic congestion in manned stop, is used
It trains One-classSVM graders, to distinguish whether stop is traffic congestion and waits the stop of traffic lights.
According to trained SVM classifier, removes the stop of unloaded parking medium traffic lamp and traffic congestion and retain interest
Stop, to obtain the interest stop of reflection taxi driver intention.
Construction track characteristic vector simultaneously carries out path matching, goes out the motion track of corresponding vehicle for identification, and verify
Matched accuracy.
Present embodiment uses the taxi GPS data reportings of Shanghai (906) and Shenzhen (1945), as original
Beginning mobile trajectory data collection.The specifying information that every track data includes has：The number ID of vehicle, current location, current location
Longitude and latitude, timestamp, drive speed, the traveling angle and motion state of vehicle.Motion state can show whether taxi carries
People.And there are unique number in every section and each intersection in respective road network in the road network in Shanghai and Shenzhen.?
Before handling track data, first track is pre-processed.It, first will be by electronically since there are errors for the positioning of GPS
Road network information in figure carries out aligning, and track is all restored on road, and determines position of the vehicle relative to map.It is logical
It crosses and the GPS point of track is mapped on corresponding section using ST-Matching map-matching algorithms.Each GPS in data set
A daylong movement locus of taxi is had recorded in journal file.The complete trajectory of vehicle is divided as unit of day
At several sub-trajectories.Have recorded the data of 28 days and 31 days respectively to Shanghai and Shenzhen.The 20 and 22 days companies in Shanghai and Shenzhen
Continuous track respectively constitutes the set for including 906 and 1945 anonymous tracks, remaining 8 and 9 days, independent using daily track
Constitute the test trails segment of corresponding anonymous track collection.
It should be noted that since the stop of taxi can be divided into on-board and off-board parking, carrying stop off and empty wagons
Parking three categories are needed first to filter out on-board and off-board stop, be easy to according to the information that data include to obtain useful information
On-board and off-board stop can be removed.Fig. 2 is the exhaustive division of stop.
In the present embodiment, removal is stopped caused by blocking up and waiting traffic lights, and has used down time, stop arrives
Three features of interval time of the distance at crossing and two stops.Since traffic congestion causes vehicle to loiter, and parking
Time, two stops interval time have direct relation, equally wait traffic lights and the distance of stop to crossing, down time
Also there is direct relation, therefore use these three attributes that can reach as the characteristic attribute constitutive characteristic vector of stop and distinguish this
The purpose of a little stops.Indicate that down time, the distance of dist expression stops to crossing, inter indicate two and stops with dura
The interval time of vehicle point constructs the feature vector S of stoptp=(dura, dist, inter).Fig. 3 and it is shown in Fig. 4 be to vehicle
Stop carry out statistical result, some stops constitute column figure, this indicates this vehicle one month all often at this
Stopped vehicle in a position.The such a continuous several days stops all in the same position are exactly the feature of track, as the one of edge
A little characteristic points are also the feature that certain tracks distinguish over other tracks.
In the present embodiment, the exception parking point in manned stop is filtered out using mahalanobis distance, sets geneva
The threshold value of distance isStop more than threshold value is considered as abnormal point by (p is degree of freedom).The calculating of mahalanobis distance is public
Formula is：
Wherein StpiIndicate the feature vector of i-th of stop,Indicate one group of parking point set
Feature vector mean value, v then indicates the covariance matrix of stop set feature vector S.
In present embodiment, the traffic congestion in manned stop is identified and filtered out using the SVM classifier trained and is waited
The stop of traffic lights, to obtain required interest stop.By using the manned stop after removal exception parking point
An One-classSVM grader is trained, which can be determined that whether stop is traffic congestion and waits the parking of traffic lights
Point.Due to only interesting stop and non-two class of interest stop in empty wagons stop, is removing all traffic congestions and waiting
After the non-interest stop of traffic lights, the stop of reservation is required interest stop.The result classified using SVM classifier
As shown in Figure 5 and Figure 6.Wherein, the deeper part of color stop, shallower part of color caused by block up, waiting traffic lights
It is just needed interest stop, the track characteristic vector that interest stop is constituted is used for calculating matching accuracy rate.
In present embodiment, track characteristic vector is constructed with the interest stop extracted.Construction feature vector uses
Improved TF-IDF methods, the mode for handling text in this way analyze track.The track that taxi travels is seen
Make a text, section number regards each word as, and it is exactly number that word occurs in the text that the parking in section, which is counted out,.
Weighted value tf-idfI, jCalculation formula be：
tf-idfI, j=tfI, j×idfI, j.
Wherein,
Wherein, NiFor vehicle v in training dataiThe section sum passed through；mjInclude section r for training datajTrack
Number；CiFor vehicle v in training dataiTrack belonging to one kind；mI, jFor CiIn include section rjTrack number；tI, jFor rjIn Ci
The number of middle appearance；tfI, jFor rjIn viTrack in the frequency that occurs.
Obtaining weighted value tf-idfI, jAfterwards, so that it may to form following track characteristic vector：
Wi=(ωI, 1, ωI, 2..., ωI, n).
Wherein ωI, jIndicate rjWeighted value, i.e. ωI, j=tf-idfI, j。
These stops extracted in training data constitute track characteristic, and the stop in test data constitutes test
Track characteristic is matched with complementation string similarity value.By measure two inner product of vectors spaces angle cosine value come
Measure the similitude between them.It is identical to determine whether two vectors are pointed generally in from the cosine value of the angle between two vectors
Direction.In comparison procedure, vectorial scale is not considered, and considers only the pointing direction of vector.Two vectors
Between the form of cosine similarity dot product indicate that its size, formula are as follows：
Wherein, A, B indicate two different track characteristic vectors respectively.
The cosine similarity similarity of test fragment and each wheel paths in anonymous track set is found out first, point
The maximum cosine similarity value for not finding out test trails segment and they, is eventually found maximum value therein, is regarded as and surveys
Examination path segment belongs to a vehicle.
In the present embodiment, when obtaining the feature vector of every test trails, its affiliated vehicle has been understood.Only need by
Compare the affiliated information of vehicles in anonymous track that is obtained and it is true belonging to information of vehicles be compared, it is last accurate just to can know that
Rate.By analyzing accuracy rate, the method effect degree of this extraction feature can be embodied.Specific accuracy rate formula is as follows：
Wherein, ncorrcetIndicate the classification results after test, nallIndicate true classification.
The accuracy rate for removing anonymous methods based on stop is as shown in Figure 7.Accuracy rate is respectively 48.58% and 39.66%.
This method can distinguish 444 i.e. in 906 track of vehicle of Shanghai, and this method can distinguish 720 in 1945 tracks in Shenzhen.
In addition, the increase (meaning more interest stop information) of the path length (i.e. number of days) with attacker's acquisition, goes anonymity
Accuracy rate will significantly rise.As shown in Figure 8 and Figure 9, Shanghai Data and Shenzhen data are risen by 48.58% and 39.66% respectively
To 85.22% and 66.67%.It can be seen that as long as attacker obtains the truthful data of enough users, so that it may with accurately
The original track of victim is allotted, to steal the privacy of user.
It does not hide it is not difficult to find that the present invention realizes to match from anonymous motion track concentration by stop record
The initial trace of name analyzes interest parking point feature caused by driver personal preference, verifies single driver's interest
The stability for point feature of stopping and uniqueness, and disclose anonymous track data security risk.
Claims (6)
1. a kind of removing the mobile trajectory data of vehicle anonymity anonymous method based on parking record, which is characterized in that obtain certain
Several motion track segments of one taxi any time period are known using track is held compared with the track that anonymous track is concentrated
The not track of the taxi, includes the following steps：
(1) all stops in track are extracted, are stopped caused by removing on-board and off-board, parking is divided into manned parking to stop with zero load
Vehicle, wherein manned parking includes stopping caused by stopping and block up caused by waiting traffic lights；
(2) abnormal point of manned parking is removed, and manned parking is modeled；
(3) equal traffic lights and traffic congestion in the unloaded stop of gained model removal are used to stop, reservation interest stop；
(4) feature vector based on TF-IDF is constructed to interest stop, by the feature vector of known trajectory and anonymous track
Feature vector compared to pair, find out in the set of anonymous track with the matched anonymous track of known trajectory.
2. according to claim 1 remove the mobile trajectory data of vehicle anonymity anonymous method based on parking record,
It is characterized in that, the step (1) is specially：The point that screening out position does not change over time is concentrated from track, regards it as stop,
Whether it is on-board and off-board parking according to the manned condition adjudgement before and after the parking for hiring out wheel paths concentration.
3. according to claim 1 remove the mobile trajectory data of vehicle anonymity anonymous method based on parking record,
It is characterized in that, indicates that down time, dist indicate stop to the distance at crossing, inter expressions with dura in the step (2)
The interval time of two stops, construction feature vector Stp=(dura, dist, inter) removes manned stop using mahalanobis distance
The abnormal point of vehicle.
4. according to claim 3 remove the mobile trajectory data of vehicle anonymity anonymous method based on parking record,
It is characterized in that, the abnormal point that manned parking is removed using mahalanobis distance is specially：Set the threshold value of mahalanobis distance asWherein, p is degree of freedom, and the stop more than threshold value is considered as abnormal point；The calculation formula of mahalanobis distance is：Wherein StpiIndicate that the feature vector of i-th of stop, μ indicate one group of stop
The feature vector mean value of set；V then indicates the covariance matrix of stop set feature vector.
5. according to claim 1 remove the mobile trajectory data of vehicle anonymity anonymous method based on parking record,
It is characterized in that, One-class SVM classifiers is trained with the manned stop after removal exception parking point in the step (3),
To distinguish whether stop is traffic congestion and waits the stop of traffic lights；Identified and removed in unloaded stop with grader traffic congestion and
The stop of equal traffic lights, reservation is interest stop.
6. according to claim 1 remove the mobile trajectory data of vehicle anonymity anonymous method based on parking record,
It is characterized in that, the step (4) is specially：Regard the track that taxi travels as a text, section number regards each as
Word, it is exactly number that word occurs in the text that the parking in section, which is counted out,；Weighted value tf-idfI, jCalculation formula be：
tf-idfI, j=tfI, j×idfI, j, whereinNiFor vehicle in training data
ViThe section sum passed through；mjInclude section r for training datajTrack number；CiFor vehicle v in training dataiTrack
Affiliated one kind；mI, jFor CiIn include section rjTrack number；tI, jFor rjIn CiThe number of middle appearance；tfI, jFor rjIn viRail
The frequency occurred in mark；The stop extracted in training data constitutes track characteristic, and the stop in test data, which is constituted, to be surveyed
Track characteristic is tried, is matched with complementation string similarity value, the cosine value of the angle by measuring two inner product of vectors spaces
To measure the similitude between them；Determine whether two vectors are pointed generally in phase from the cosine value of the angle between two vectors
Same direction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810385918.0A CN108734008B (en) | 2018-04-26 | 2018-04-26 | Method for removing anonymity of moving track data anonymized to vehicle based on parking record |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810385918.0A CN108734008B (en) | 2018-04-26 | 2018-04-26 | Method for removing anonymity of moving track data anonymized to vehicle based on parking record |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108734008A true CN108734008A (en) | 2018-11-02 |
| CN108734008B CN108734008B (en) | 2021-12-07 |
Family
ID=63939306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810385918.0A Active CN108734008B (en) | 2018-04-26 | 2018-04-26 | Method for removing anonymity of moving track data anonymized to vehicle based on parking record |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108734008B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428604A (en) * | 2019-07-30 | 2019-11-08 | 山东交通学院 | It is a kind of based on the taxi illegal parking of GPS track data and map datum monitoring and method for early warning |
| CN110428621A (en) * | 2019-07-30 | 2019-11-08 | 山东交通学院 | A kind of monitoring of Floating Car dangerous driving behavior and method for early warning based on track data |
| CN110766937A (en) * | 2019-05-22 | 2020-02-07 | 北京嘀嘀无限科技发展有限公司 | Parking spot identification method and device, electronic equipment and readable storage medium |
| CN111224940A (en) * | 2019-11-15 | 2020-06-02 | 中国科学院信息工程研究所 | Anonymous service traffic correlation identification method and system nested in encrypted tunnel |
| DE102020204045A1 (en) | 2020-03-27 | 2021-09-30 | Volkswagen Aktiengesellschaft | Vehicle, method, device and computer program for a vehicle for determining a traffic density from at least one movement profile of a vehicle |
| CN113641887A (en) * | 2021-08-26 | 2021-11-12 | 河南工业大学 | Mobile track de-anonymization method and system based on semantic track mode |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354857A (en) * | 2015-12-07 | 2016-02-24 | 北京航空航天大学 | Matching method for vehicle track shielded by overpass |
| CN105608919A (en) * | 2014-11-21 | 2016-05-25 | 杭州海康威视数字技术股份有限公司 | Method and device for determining station position |
| CN105810006A (en) * | 2016-03-29 | 2016-07-27 | 福建工程学院 | Method and system for recognizing roadside parking place |
| CN107945037A (en) * | 2017-11-27 | 2018-04-20 | 北京工商大学 | A kind of social networks based on node structure feature goes de-identification method |
-
2018
- 2018-04-26 CN CN201810385918.0A patent/CN108734008B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105608919A (en) * | 2014-11-21 | 2016-05-25 | 杭州海康威视数字技术股份有限公司 | Method and device for determining station position |
| CN105354857A (en) * | 2015-12-07 | 2016-02-24 | 北京航空航天大学 | Matching method for vehicle track shielded by overpass |
| CN105810006A (en) * | 2016-03-29 | 2016-07-27 | 福建工程学院 | Method and system for recognizing roadside parking place |
| CN107945037A (en) * | 2017-11-27 | 2018-04-20 | 北京工商大学 | A kind of social networks based on node structure feature goes de-identification method |
Non-Patent Citations (1)
| Title |
|---|
| 钟建友等: "移动轨迹数据去匿名化攻击方法", 《计算机工程》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766937A (en) * | 2019-05-22 | 2020-02-07 | 北京嘀嘀无限科技发展有限公司 | Parking spot identification method and device, electronic equipment and readable storage medium |
| CN110766937B (en) * | 2019-05-22 | 2020-10-20 | 滴图(北京)科技有限公司 | Parking spot identification method and device, electronic equipment and readable storage medium |
| CN110428604A (en) * | 2019-07-30 | 2019-11-08 | 山东交通学院 | It is a kind of based on the taxi illegal parking of GPS track data and map datum monitoring and method for early warning |
| CN110428621A (en) * | 2019-07-30 | 2019-11-08 | 山东交通学院 | A kind of monitoring of Floating Car dangerous driving behavior and method for early warning based on track data |
| CN110428621B (en) * | 2019-07-30 | 2022-07-15 | 山东交通学院 | Track data-based monitoring and early warning method for dangerous driving behavior of floating car |
| CN111224940A (en) * | 2019-11-15 | 2020-06-02 | 中国科学院信息工程研究所 | Anonymous service traffic correlation identification method and system nested in encrypted tunnel |
| DE102020204045A1 (en) | 2020-03-27 | 2021-09-30 | Volkswagen Aktiengesellschaft | Vehicle, method, device and computer program for a vehicle for determining a traffic density from at least one movement profile of a vehicle |
| WO2021191337A1 (en) | 2020-03-27 | 2021-09-30 | Volkswagen Aktiengesellschaft | Vehicle, method, device and computer program for a vehicle for determining traffic density from at least one movement profile of a vehicle |
| CN113641887A (en) * | 2021-08-26 | 2021-11-12 | 河南工业大学 | Mobile track de-anonymization method and system based on semantic track mode |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108734008B (en) | 2021-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108734008A (en) | Anonymous method is gone to the mobile trajectory data of vehicle anonymity based on parking record | |
| Munoz-Organero et al. | Automatic detection of traffic lights, street crossings and urban roundabouts combining outlier detection and deep learning classification techniques based on GPS traces while driving | |
| Zhang et al. | Data-driven intelligent transportation systems: A survey | |
| CN109459045B (en) | Improved interactive voting matching method for low-frequency GPS track | |
| CN105741553A (en) | Method for identifying parking road segment in vehicle track based on dynamic threshold | |
| US20200273328A1 (en) | Method, apparatus, and system for combinining discontinuous road closures detected in a road network | |
| CN110705484B (en) | Method for recognizing continuous lane change illegal behaviors by utilizing driving track | |
| US11756417B2 (en) | Method, apparatus, and system for detecting road incidents | |
| US20220413502A1 (en) | Method, apparatus, and system for biasing a machine learning model toward potential risks for controlling a vehicle or robot | |
| CN113011331A (en) | Method and device for detecting whether motor vehicle gives way to pedestrians, electronic equipment and medium | |
| Hernández Sánchez et al. | Deep neural networks for driver identification using accelerometer signals from smartphones | |
| Li et al. | Driving performances assessment based on speed variation using dedicated route truck GPS data | |
| Muñoz-Organero et al. | Detecting different road infrastructural elements based on the stochastic characterization of speed patterns | |
| CN107622660A (en) | A kind of road restricted driving time determines method and device | |
| Das et al. | Why slammed the brakes on? auto-annotating driving behaviors from adaptive causal modeling | |
| CN115631082A (en) | Multi-scene path restoration method based on expressway network topological structure | |
| CN110491157A (en) | A kind of vehicle correlating method based on parking data and bayonet data | |
| CN109859492A (en) | A kind of system and method identifying fake-licensed car | |
| US20220246028A1 (en) | Method and apparatus for preventing traffic over-reporting via identifying misleading probe data | |
| CN111582019B (en) | Unmanned vehicle lane level scene judging method, system, terminal and storage medium | |
| CN113689705B (en) | Method and device for detecting red light running of vehicle, computer equipment and storage medium | |
| Kahaki et al. | Incident and traffic-bottleneck detection algorithm in high-resolution remote sensing imagery | |
| CN109448377B (en) | Method for evaluating vehicle driving safety by using satellite positioning data | |
| Sun et al. | Heavy vehicle classification through deep learning | |
| CN114091581A (en) | Vehicle operation behavior type identification method based on sparse track |
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 |