CN107301254A - A kind of road network hot spot region method for digging - Google Patents

A kind of road network hot spot region method for digging Download PDF

Info

Publication number
CN107301254A
CN107301254A CN201710735328.1A CN201710735328A CN107301254A CN 107301254 A CN107301254 A CN 107301254A CN 201710735328 A CN201710735328 A CN 201710735328A CN 107301254 A CN107301254 A CN 107301254A
Authority
CN
China
Prior art keywords
cluster
space
sub
time
spot
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
Application number
CN201710735328.1A
Other languages
Chinese (zh)
Other versions
CN107301254B (en
Inventor
田玲
罗光春
殷光强
陈爱国
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN201710735328.1A priority Critical patent/CN107301254B/en
Publication of CN107301254A publication Critical patent/CN107301254A/en
Application granted granted Critical
Publication of CN107301254B publication Critical patent/CN107301254B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • 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
    • G06F16/28—Databases characterised by their database models, e.g. relational or object models
    • G06F16/284—Relational databases
    • G06F16/285—Clustering or classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Multimedia (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a kind of road network hot spot region method for digging, belong to data mining technology field, solve in the prior art, the problem of during using track space-time similarity measurement and cluster calculation to trajectory clustering.Step 1 of the present invention, to all orbit segments carry out track segmentation, calculate segmentation after two sub-trajectories section between space-time similitude and time-space matrix；Step 2, the space-time similitude according to sub-trajectory, time-space matrix and the DBSCAN algorithms based on dynamic neighbour carry out cluster calculation to all track segment datas in mesh space；Step 3, select in the class cluster of cluster calculation notable class gathering and close, and extracted in the conjunction of notable class gathering and stop spot；Step 4, the track segment number carried according to stop spot draw the high temperature region for stopping spot, the hot spot region in drawing road network in the region where high temperature stops spot.The present invention is used for locus positioning.

Description

A kind of road network hot spot region method for digging
Technical field
A kind of road network hot spot region method for digging, for locus positioning, belongs to data mining technology field.
Background technology
In recent years, locus location technology fast development application, along with the quick popularization of these technologies, we can be with The positional information of substantially any mobile object is easily tracked, so as to form the huge track using track as the form of expression The deep information of certain motor behavior of mobile object can largely be reflected by containing in database, the track data of these magnanimity. Space-time trajectory data is as one kind of space-time data, and what the locus of its essential record mobile object was changed over time becomes Gesture, and vehicle space-time trajectory data is more special, it is limited in road network, therefore, and conventional data digging method is very It is many to be all not directly applicable during space-time trajectory data is excavated, it is necessary to carry out certain improvement.
Because the research of hot spot region in road network has important actual application value, so, to hotspot path region Research must be directed to track data effective in road network and carry out.It is then to find road that clustering is carried out to track data A kind of conventional means in popular path in net.Trajectory clustering mainly includes two parts：Track space-time similarity measurement and cluster Calculate.Research method the most frequently used at present is mainly based upon mesh space and track is divided in terms of the space-time similarity measurement of track Cut, this method carries out mesh space division to track data first and track is cut, and carries out space-time to the sub-trajectory after segmentation Similitude is added the space-time similitude for drawing track with chronotaxis.This method can be calculated relatively accurately between track Space-time similitude, but this method will calculate its space similarity respectively for the similarity measurement between every a pair of tracks And time similarity, when track data amount is larger, the response time of the algorithm is than larger.And in terms of cluster calculation, due to The shape of track class cluster is often similar " banding ", rather than " spherical ", therefore, and cluster calculation process is usually using the most typical Density clustering algorithm DBSCAN, the algorithm can realize the cluster calculation to arbitrary shape class cluster.But this method is being gathered The value of artificial input two parameters of the radius of neighbourhood and neighborhood density threshold is needed when class is calculated, and the two parameter values is good The bad result that will directly affect cluster, and DBSCAN algorithms are not provided with a kind of determination side of the two parameter values in itself Method.
The content of the invention
It is an object of the invention to：Solve in the prior art, using track space-time similarity measurement and cluster calculation to rail When mark is clustered, space-time similarity measurement is when track data volume is larger, and the response time is than larger；European coordinate can not be expressed accurately Distance between two tracks in road network；And, it is necessary to which artificial is defeated during use density clustering algorithm DBSCAN progress cluster calculations Enter the radius of neighbourhood and neighborhood density threshold, when value is inaccurate, the problem of result of cluster can be directly affected；The invention provides A kind of road network hot spot region method for digging.
The technical solution adopted by the present invention is as follows：
A kind of road network hot spot region method for digging, it is characterised in that following steps：
Step 1, to all orbit segments carry out track segmentation, calculate segmentation after two sub-trajectories section between space-time it is similar Property and time-space matrix；
Step 2, according to the space-time similitude of sub-trajectory, time-space matrix and the DBSCAN algorithms based on dynamic neighbour are to grid All track segment datas in space carry out cluster calculation；
Step 3, select in the class cluster of cluster calculation notable class gathering and close, and extracted in the conjunction of notable class gathering and stop spot Point；
Step 4, the track segment number carried according to stop spot draw the high temperature region for stopping spot, in high temperature Region where stopping spot draws the hot spot region in road network.
Further, the step 1 is comprised the following steps that：
Step 1.1, the division that dynamic grid space is carried out to the area of space where all orbit segments；
Step 1.2, according to breakpoint in mesh space track sets carry out track segmentation；
Step 1.3, calculate track segmentation after two sub-trajectories section between space-time similitude and time-space matrix.
Further, the step 1.1 is comprised the following steps that：
The minimum external square of step 1.11, area of space according to where Minimum Convex Closure principle solving goes out all orbit segments Shape；
Step 1.12, the sampled point number contained on each track segment length and orbit segment is solved, calculate the orbit segment On the average distance that is moved within the two neighboring sampled point time of vehicle；
Step 1.13, using average distance size as mesh space size, dynamic grid is carried out to minimum enclosed rectangle empty Between divide.
Further, the step 1.2 is comprised the following steps that：
Step 1.21, the data for being successively read each sampled point on each orbit segment；
Step 1.22：Compare the longitude and latitude degrees of data of two neighboring orbit segment sampled point position；
Step 1.23：If longitude and latitude is unchanged between two neighboring sampled point, at the place of two sampled point centre positions As breakpoint；
Step 2.4：Track segmentation is carried out to initial trace section according to each breakpoint location calculated.
Further, the step 1.3 is comprised the following steps that：
Spatial simlanty between step 1.31, calculating two sub-trajectories section, if spatial simlanty is not zero, calculates two sub- rails Chronotaxis between mark section, otherwise goes to step 1.33, and the formula for calculating spatial simlanty and chronotaxis is：
In formula, Lc(TRi,TRj) represent two tracks in sub-trajectory section space or accumulated time length, L (TRi) table Show sub-trajectory TRiTotal length, L (TRj) represent sub-trajectory TRjTotal length, L (TRi)+L(TRj)-Lc(TRi,TRj) represent It is space or total length of time, the i.e. span of two strip orbit segments, Sim (TRi,TRj) represent two sub-trajectories section between space phase Like property or chronotaxis；
If step 1.32, chronotaxis calculate the space-time similitude between two sub-trajectories section, otherwise gone to be not zero Step 1.33, the formula of calculating space-time similitude is：
STSim(TRi,TRj)=SSim (TRi,TRj)×TSim(TRi,TRj)；
In formula, SSim (TRi,TRj) what is represented is the spatial simlanty between two sub-trajectories section, TSim (TRi,TRj) represent Be chronotaxis between two sub-trajectories, STSim (TRi,TRj) what is represented is that the two sub-trajectories section space-time calculated is similar Property measurement；
Step 1.33, time-space matrix between two sub-trajectories is calculated, computational methods are：
STDist(TRi,TRj)=1-STSim (TRi,TRj)；
In formula, STSim (TRi,TRj) what is represented is the space-time similarity measurement between two sub-trajectories section, STDist (TRi, TRj) what is represented is the time-space matrix between two sub-trajectories section.
Further, the step 2 is comprised the following steps that：
Step 2.1, according to the space-time similitude of sub-trajectory, time-space matrix, neighbour's scale evolution algorithmic and DBSCAN algorithms Calculate neighbour's scale change of each orbit segment up-sampling point；
Each sampled point distance and the distance of other sampled points in step 2.2, calculating orbit segment, for a sampled point, Sampled point away from its ultimate range is labeled as max, and the sampled point away from its minimum range is labeled as min, if max>2min, then should Sampled point is divided into concussion object set, otherwise, is divided into stable object set；
The numbering Cluster_id=1 of cluster in step 2.3, the stable object set of initialization and concussion object set, and will The cluster numbering of stable object set interior joint is defaulted as 0；
Step 2.4, the kernel object v that any one cluster numbering of selection is 0 in object set is stablized, and breadth First searches The reachable object set Reach of its density of rope；
It is minimum in step 2.5, the search kernel object collection Core in object set Reach, search kernel object collection Core Cluster numbering Min_Cluster；
If step 2.6, Min_Cluster=0, object set Reach and kernel object v cluster numbering are labeled as Cluster_id, Cluster_id=Cluster_id+1, otherwise, search are connected with object set Reach and kernel object v density Object set Connect, object set Reach, density connected object set Connect and kernel object v cluster are numbered and marked Class cluster is obtained for Min_Cluster, i.e. cluster；
Step 2.7, judge in stable set object whether to still have kernel object v, if in the presence of, return to step 2.3, If otherwise obtaining all class clusters, step 2.8 is carried out；
Step 2.8, differentiation shake boundary point and noise spot in object set Oscillation, and allocation boundary point is arrived Different cluster in class cluster.
Further, the step 3 is comprised the following steps that：
The number n and the quantity m of all orbit segments of class cluster obtained by step 3.1, Statistical Clustering Analysis；
Step 3.2, make p=m/n；If the quantity containing orbit segment is more than p in step 3.3, cluster gained class cluster, should Class cluster is labeled as notable class cluster, otherwise labeled as non-significant class cluster；
Step 3.4：A notable class cluster C is chosen from cluster result, the orbit segment included in notable class cluster C is originated Point is set to point set K；
Step 3.5：A breakpoint b is randomly selected from point set K, other breakpoints are merged into composition with breakpoint b successively to expand Big point set Q, if the minimum circumscribed circle radius that the breakpoint b added causes point set Q is more than preassigned threshold value beta, needs Breakpoint b is deleted from point set Q；
Step 3.6：The institute for traveling through fixed point collection K a little, if the breakpoint number distribution contained in point set Q is more than threshold alpha, marks Point set Q is stop spot；
Step 3.7：Repeat step 3.5- steps 3.6, spot is stopped until having generated candidates all in notable class cluster C；
Step 3.8：Repeat step 3.4- steps 3.7, until having traveled through the notable class cluster of whole.
Further, the step 4 is comprised the following steps that：
Step 4.1, calculate that each stop spot is corresponding to stop temperature information, computational methods are：
Wherein, hspotTo stop the temperature of spot, nsubtraTo stop the orbit segment number that spot is included, ntraRepresent It is to stop the trace number that spot is included, β is a coefficient；
Step 4.2, from stop temperature information in obtain stop spot high temperature region；
Step 4.3, the region according to where high temperature stops spot draw the hot spot region in road network.
In summary, by adopting the above-described technical solution, the beneficial effects of the invention are as follows：
1st, the road hot spot region method for digging that invention is provided, combines the track space-time under mesh space Similarity measurement and the DBSCAN method of trajectory clustering of optimization, road network can not accurately be expressed by preferably overcoming traditional European coordinate In the drawbacks of distance and tradition DBSCAN clusters need artificially to pre-enter relevant parameter between two tracks；
2nd, the method proposed by the present invention that track sets are represented using mesh space coordinate overcome track sampled point because There is deviation and can not accurately calculate the puzzlement of space-time similitude between track in network environment, sample devices etc.；
3. proposed by the present invention excavate the method for road hot spot region for high sample frequency track number based on track of vehicle According to best results, it can not only save the memory space expense of track data, can also improve the execution efficiency of whole system；
4. proposed by the present invention trajectory time similitude is multiplied with spatial simlanty draws the side of track space-time similitude Method, can greatly improve the computational efficiency for calculating track space-time similitude, the response time is than very fast.
Brief description of the drawings
The sub-process figure that Fig. 1 calculates for track space-time similitude in the present invention and time-space matrix；
Fig. 2 is the DBSCAN algorithm flow charts optimized based on dynamic neighbour in the present invention；
Fig. 3 is the sub-process figure based on cluster result progress hot spot region excavation in the present invention；
Fig. 4 is the notable class cluster distribution situation in step 5 of the present invention, wherein, the section of black region covering is notable The poly- region of class gathering；
Fig. 5 is the hot spot region distribution situation in step 7 of the present invention, wherein, black splotch region is high temperature area Domain, gray corrosion region is common temperature region.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, not For limiting the present invention.
The invention provides a kind of road hot spot region method for digging.Optimized by entering Mobile state neighbour to track of vehicle DBSCAN clusters and stopped the calculating of spot temperature, can accurately and effectively road improvement hot spot region mining effect.Pass through DBSCAN clustering algorithms based on dynamic neighbour overcome the drawbacks of cluster result is influenceed larger by artificial input parameter value.And lead to Cross then can relatively accurately describe the distribution situation of hot spot region to stopping spot progress temperature information calculating.
A kind of road network hot spot region method for digging, following steps：
Step 1, to all orbit segments carry out track segmentation, calculate segmentation after two sub-trajectories section between space-time it is similar Property and time-space matrix；Comprise the following steps that：
Step 1.1, the division that dynamic grid space is carried out to the area of space where all orbit segments；Specific steps are such as Under：
The minimum external square of step 1.11, area of space according to where Minimum Convex Closure principle solving goes out all orbit segments Shape；
Step 1.12, the sampled point number contained on each track segment length and orbit segment is solved, calculate the orbit segment On the average distance that is moved within the two neighboring sampled point time of vehicle；
Step 1.13, using average distance size as mesh space size, dynamic grid is carried out to minimum enclosed rectangle empty Between divide.
Step 1.2, according to breakpoint in mesh space track sets carry out track segmentation；Comprise the following steps that：
Step 1.21, the data for being successively read each sampled point on each orbit segment；
Step 1.22：Compare the longitude and latitude degrees of data of two neighboring orbit segment sampled point position；
Step 1.23：If longitude and latitude is unchanged between two neighboring sampled point, at the place of two sampled point centre positions As breakpoint；
Step 2.4：Track segmentation is carried out to initial trace section according to each breakpoint location calculated.
Step 1.3, calculate track segmentation after two sub-trajectories section between space-time similitude and time-space matrix.Step 1.31st, calculate two sub-trajectories section between spatial simlanty, if spatial simlanty is not zero, calculate two sub-trajectories section between when Between similitude, otherwise go to step 1.33, the formula for calculating spatial simlanty and chronotaxis is：
In formula, Lc(TRi,TRj) represent two tracks in sub-trajectory section space or accumulated time length, L (TRi) table Show sub-trajectory TRiTotal length, L (TRj) represent sub-trajectory TRjTotal length, L (TRi)+L(TRj)-Lc(TRi,TRj) represent It is space or total length of time, the i.e. span of two strip orbit segments, Sim (TRi,TRj) represent two sub-trajectories section between space phase Like property or chronotaxis；
If step 1.32, chronotaxis calculate the space-time similitude between two sub-trajectories section, otherwise gone to be not zero Step 1.33, the formula of calculating space-time similitude is：
STSim(TRi,TRj)=SSim (TRi,TRj)×TSim(TRi,TRj)；
In formula, SSim (TRi,TRj) what is represented is the spatial simlanty between two sub-trajectories section, TSim (TRi,TRj) represent Be chronotaxis between two sub-trajectories, STSim (TRi,TRj) what is represented is that the two sub-trajectories section space-time calculated is similar Property measurement；
Step 1.33, time-space matrix between two sub-trajectories is calculated, computational methods are：
STDist(TRi,TRj)=1-STSim (TRi,TRj)；
In formula, STSim (TRi,TRj) what is represented is the space-time similarity measurement between two sub-trajectories section, STDist (TRi, TRj) what is represented is the time-space matrix between two sub-trajectories section.
Step 2, according to the space-time similitude of sub-trajectory, time-space matrix and the DBSCAN algorithms based on dynamic neighbour are to grid All track segment datas in space carry out cluster calculation；Comprise the following steps that：
Comprise the following steps that：
Step 2.1, according to the space-time similitude of sub-trajectory, time-space matrix, neighbour's scale evolution algorithmic and DBSCAN algorithms Calculate neighbour's scale change of each orbit segment up-sampling point；
Each sampled point distance and the distance of other sampled points in step 2.2, calculating orbit segment, for a sampled point, Sampled point away from its ultimate range is labeled as max, and the sampled point away from its minimum range is labeled as min, if max>2min, then should Sampled point is divided into concussion object set, otherwise, is divided into stable object set；
The numbering Cluster_id=1 of cluster in step 2.3, the stable object set of initialization and concussion object set, and will The cluster numbering of stable object set interior joint is defaulted as 0；
Step 2.4, the kernel object v that any one cluster numbering of selection is 0 in object set is stablized, and breadth First searches The reachable object set Reach of its density of rope, the whether reachable standard of density herein is exactly that whether there is up to road between object Footpath, exist it is then reachable, in the absence of then unreachable；
It is minimum in step 2.5, the search kernel object collection Core in object set Reach, search kernel object collection Core Cluster numbering Min_Cluster；
If step 2.6, Min_Cluster=0, object set Reach and kernel object v cluster numbering are labeled as Cluster_id, Cluster_id=Cluster_id+1, otherwise, search are connected with object set Reach and kernel object v density Object set Connect, object set Reach, density connected object set Connect and kernel object v cluster are numbered and marked Class cluster is obtained for Min_Cluster, i.e. cluster；
Step 2.7, judge in stable set object whether to still have kernel object v, if in the presence of, return to step 2.3, If otherwise obtaining all class clusters, step 2.8 is carried out；
Step 2.8, differentiation shake boundary point and noise spot in object set Oscillation, and allocation boundary point is arrived Different cluster in class cluster.
Step 3, select in the class cluster of cluster calculation notable class gathering and close, and extracted in the conjunction of notable class gathering and stop spot Point；Comprise the following steps that：
The number n and the quantity m of all orbit segments of class cluster obtained by step 3.1, Statistical Clustering Analysis；
Step 3.2, make p=m/n；If the quantity containing orbit segment is more than p in step 3.3, cluster gained class cluster, should Class cluster is labeled as notable class cluster, otherwise labeled as non-significant class cluster；
Step 3.4：A notable class cluster C is chosen from cluster result, the orbit segment included in notable class cluster C is originated Point is set to point set K；
Step 3.5：A breakpoint b is randomly selected from point set K, other breakpoints are merged into composition with breakpoint b successively to expand Big point set Q, if the minimum circumscribed circle radius that the breakpoint b added causes point set Q is more than preassigned threshold value beta, needs Breakpoint b is deleted from point set Q；
Step 3.6：The institute for traveling through fixed point collection K a little, if the breakpoint number distribution contained in point set Q is more than threshold alpha, marks Point set Q is stop spot；
Step 3.7：Repeat step 3.5- steps 3.6, spot is stopped until having generated candidates all in notable class cluster C；
Step 3.8：Repeat step 3.4- steps 3.7, until having traveled through the notable class cluster of whole.
Step 4, the track segment number carried according to stop spot draw the high temperature region for stopping spot, in high temperature Region where stopping spot draws the hot spot region in road network；Comprise the following steps that：
Step 4.1, calculate that each stop spot is corresponding to stop temperature information, computational methods are：
Wherein, hspotTo stop the temperature of spot, nsubtraTo stop the orbit segment number that spot is included, ntraRepresent It is to stop the trace number that spot is included, β is a coefficient, is set in testWherein, any two orbit segments No matter whether identical is all different tracks section, but if two orbit segments are identical, then they are identical strip path curves.That is orbit segment Number is more than (during containing same trajectories) or equal to (when all orbit segments are differed) trace number.
Step 4.2, from stop temperature information in obtain stop spot high temperature region；
Step 4.3, the region according to where high temperature stops spot draw the hot spot region in road network.
Compared with prior art, road hot spot region provided by the present invention method for digging, is combined under mesh space Track space-time similarity measurement and the DBSCAN method of trajectory clustering of optimization, preferably overcoming traditional European coordinate can not be accurate The drawbacks of distance and tradition DBSCAN clusters need artificially to pre-enter relevant parameter between two tracks in expression road network.Together When, the method proposed by the present invention that track sets are represented using mesh space coordinate overcomes track sampled point because of network rings There is deviation and can not accurately calculate the puzzlement of space-time similitude between track in border, sample devices etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention Any modifications, equivalent substitutions and improvements made within refreshing and principle etc., should be included in the scope of the protection.

Claims (8)

1. a kind of road network hot spot region method for digging, it is characterised in that following steps：
Step 1, to all orbit segments carry out track segmentation, calculate segmentation after two sub-trajectories section between space-time similitude and Time-space matrix；
Step 2, according to the space-time similitude of sub-trajectory, time-space matrix and the DBSCAN algorithms based on dynamic neighbour are to mesh space In all track segment datas carry out cluster calculation；
Step 3, select in the class cluster of cluster calculation notable class gathering and close, and extracted in the conjunction of notable class gathering and stop spot；
Step 4, the track segment number carried according to stop spot draw the high temperature region for stopping spot, are stopped in high temperature Region where spot draws the hot spot region in road network.
2. a kind of road network hot spot region method for digging according to claim 1, it is characterised in that the step 1 it is specific Step is as follows：
Step 1.1, the division that dynamic grid space is carried out to the area of space where all orbit segments；
Step 1.2, according to breakpoint in mesh space track sets carry out track segmentation；
Step 1.3, calculate track segmentation after two sub-trajectories section between space-time similitude and time-space matrix.
3. a kind of road network hot spot region method for digging according to claim 2, it is characterised in that the tool of the step 1.1 Body step is as follows：
The minimum enclosed rectangle of step 1.11, area of space according to where Minimum Convex Closure principle solving goes out all orbit segments；
Step 1.12, the sampled point number contained on each track segment length and orbit segment is solved, calculated on the orbit segment The average distance that vehicle is moved within the two neighboring sampled point time；
Step 1.13, using average distance size as mesh space size, dynamic grid space is carried out to minimum enclosed rectangle and drawn Point.
4. a kind of road network hot spot region method for digging according to claim 2,3, it is characterised in that the step 1.2 Comprise the following steps that：
Step 1.21, the data for being successively read each sampled point on each orbit segment；
Step 1.22：Compare the longitude and latitude degrees of data of two neighboring orbit segment sampled point position；
Step 1.23：If longitude and latitude is unchanged between two neighboring sampled point, two sampled point centre positions are at place Breakpoint；
Step 2.4：Track segmentation is carried out to initial trace section according to each breakpoint location calculated.
5. a kind of road network hot spot region method for digging according to claim 1, it is characterised in that the tool of the step 1.3 Body step is as follows：
Spatial simlanty between step 1.31, calculating two sub-trajectories section, if spatial simlanty is not zero, calculates two sub-trajectories section Between chronotaxis, otherwise go to step 1.33, the formula for calculating spatial simlanty and chronotaxis is：
In formula, Lc(TRi,TRj) represent two tracks in sub-trajectory section space or accumulated time length, L (TRi) represent sub- rail Mark TRiTotal length, L (TRj) represent sub-trajectory TRjTotal length, L (TRi)+L(TRj)-Lc(TRi,TRj) what is represented is two The space of sub-trajectory section or total length of time, i.e. span, Sim (TRi,TRj) represent two sub-trajectories section between spatial simlanty or Chronotaxis；
If step 1.32, chronotaxis are not zero, the space-time similitude between two sub-trajectories section is calculated, step is otherwise gone to 1.33, calculate space-time similitude formula be：
STSim(TRi,TRj)=SSim (TRi,TRj)×TSim(TRi,TRj)；
In formula, SSim (TRi,TRj) what is represented is the spatial simlanty between two sub-trajectories section, TSim (TRi,TRj) represent be Chronotaxis between two sub-trajectories, STSim (TRi,TRj) what is represented is the two sub-trajectories section space-time similarity measurements calculated Amount；
Step 1.33, time-space matrix between two sub-trajectories is calculated, computational methods are：
STDist(TRi,TRj)=1-STSim (TRi,TRj)；
In formula, STSim (TRi,TRj) what is represented is the space-time similarity measurement between two sub-trajectories section, STDist (TRi,TRj) table What is shown is the time-space matrix between two sub-trajectories section.
6. a kind of road network hot spot region method for digging according to claim 1, it is characterised in that the step 2 it is specific Step is as follows：
Step 2.1, according to the space-time similitude of sub-trajectory, time-space matrix, neighbour's scale evolution algorithmic and DBSCAN algorithms calculate Neighbour's scale change of each orbit segment up-sampling point；
Each sampled point distance and the distance of other sampled points in step 2.2, calculating orbit segment, for a sampled point, away from it The sampled point of ultimate range is labeled as max, and the sampled point away from its minimum range is labeled as min, if max>2min, then sample this Point is divided into concussion object set, otherwise, is divided into stable object set；
The numbering Cluster_id=1 of cluster in step 2.3, the stable object set of initialization and concussion object set, and will be stable The cluster numbering of object set interior joint is defaulted as 0；
Step 2.4, the kernel object v that any one cluster numbering of selection is 0 in object set is stablized, and BFS its The reachable object set Reach of density；
Minimum cluster is compiled in step 2.5, the search kernel object collection Core in object set Reach, search kernel object collection Core Number Min_Cluster；
If step 2.6, Min_Cluster=0, object set Reach and kernel object v cluster numbering are labeled as Cluster_ Id, Cluster_id=Cluster_id+1, otherwise, search for the object set being connected with object set Reach and kernel object v density Connect, Min_ is labeled as by object set Reach, density connected object set Connect and kernel object v cluster numbering Cluster, i.e. cluster obtain class cluster；
Step 2.7, judge in stable set object whether to still have kernel object v, if in the presence of, return to step 2.3, if not All class clusters are then obtained, step 2.8 is carried out；
Step 2.8, the boundary point and noise spot distinguished in concussion object set Oscillation, and allocation boundary point is to class cluster Middle different cluster.
7. a kind of road network hot spot region method for digging according to claim 1, it is characterised in that the step 3 it is specific Step is as follows：
The number n and the quantity m of all orbit segments of class cluster obtained by step 3.1, Statistical Clustering Analysis；
Step 3.2, make p=m/n；If the quantity containing orbit segment is more than p in step 3.3, cluster gained class cluster, by such cluster Labeled as notable class cluster, otherwise labeled as non-significant class cluster；
Step 3.4：A notable class cluster C is chosen from cluster result, the orbit segment starting point included in notable class cluster C is determined For point set K；
Step 3.5：A breakpoint b is randomly selected from point set K, it is expandable that other breakpoints are merged into composition with breakpoint b successively Point set Q, if the minimum circumscribed circle radius that the breakpoint b added causes point set Q is more than preassigned threshold value beta, needs from point Breakpoint b is deleted in collection Q；
Step 3.6：The institute for traveling through fixed point collection K a little, if the breakpoint number distribution contained in point set Q is more than threshold alpha, marks point set Q is stop spot；
Step 3.7：Repeat step 3.5- steps 3.6, spot is stopped until having generated candidates all in notable class cluster C；
Step 3.8：Repeat step 3.4- steps 3.7, until having traveled through the notable class cluster of whole.
8. a kind of road network hot spot region method for digging according to claim 1, it is characterised in that the step 4 it is specific Step is as follows：
Step 4.1, calculate that each stop spot is corresponding to stop temperature information, computational methods are：
Wherein, hspotTo stop the temperature of spot, nsubtraTo stop the orbit segment number that spot is included, ntraWhat is represented is to stop The trace number that spot is included is stayed, β is a coefficient；
Step 4.2, from stop temperature information in obtain stop spot high temperature region；
Step 4.3, the region according to where high temperature stops spot draw the hot spot region in road network.
CN201710735328.1A 2017-08-24 2017-08-24 Road network hot spot area mining method Active CN107301254B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710735328.1A CN107301254B (en) 2017-08-24 2017-08-24 Road network hot spot area mining method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710735328.1A CN107301254B (en) 2017-08-24 2017-08-24 Road network hot spot area mining method

Publications (2)

Publication Number Publication Date
CN107301254A true CN107301254A (en) 2017-10-27
CN107301254B CN107301254B (en) 2020-07-10

Family

ID=60132049

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710735328.1A Active CN107301254B (en) 2017-08-24 2017-08-24 Road network hot spot area mining method

Country Status (1)

Country Link
CN (1) CN107301254B (en)

Cited By (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108133172A (en) * 2017-11-16 2018-06-08 北京华道兴科技有限公司 Method, the analysis method of vehicle flowrate and the device that Moving Objects are classified in video
CN108427965A (en) * 2018-03-05 2018-08-21 重庆邮电大学 A kind of hot spot region method for digging based on road network cluster
CN108804563A (en) * 2018-05-22 2018-11-13 阿里巴巴集团控股有限公司 A kind of data mask method, device and equipment
CN109033011A (en) * 2018-06-19 2018-12-18 东软集团股份有限公司 Calculate method, apparatus, storage medium and the electronic equipment of track frequency
CN109615857A (en) * 2018-12-20 2019-04-12 首都师范大学 The deployment of roadside unit and dispatching method and device in city vehicle-mounted net
CN109686085A (en) * 2018-12-17 2019-04-26 北京交通大学 Activity On the Node kind identification method is stopped based on GPS data vehicle transport dangerous goods
CN109711451A (en) * 2018-12-20 2019-05-03 成都四方伟业软件股份有限公司 A kind of data processing method, device, electronic equipment and storage medium
CN109767615A (en) * 2018-10-19 2019-05-17 江苏智通交通科技有限公司 Road network traffic flow key flow direction and critical path analysis method
CN109977109A (en) * 2019-04-03 2019-07-05 深圳市甲易科技有限公司 A kind of track data cleaning method and adjoint analysis method
CN110174115A (en) * 2019-06-05 2019-08-27 武汉中海庭数据技术有限公司 A kind of method and device automatically generating high accuracy positioning map based on perception data
CN110457315A (en) * 2019-07-19 2019-11-15 国家计算机网络与信息安全管理中心 A kind of group's accumulation mode analysis method and system based on user trajectory data
CN110827540A (en) * 2019-11-04 2020-02-21 黄传明 Motor vehicle movement mode recognition method and system based on multi-mode data fusion
CN110909037A (en) * 2019-10-09 2020-03-24 中国人民解放军战略支援部队信息工程大学 Frequent track mode mining method and device
CN111275963A (en) * 2020-01-14 2020-06-12 北京百度网讯科技有限公司 Method and device for mining hot spot area, electronic equipment and storage medium
CN111372188A (en) * 2020-05-27 2020-07-03 腾讯科技（深圳）有限公司 Method and device for determining hot spot track in area, storage medium and electronic device
CN111380541A (en) * 2018-12-29 2020-07-07 沈阳美行科技有限公司 Interest point determination method and device, computer equipment and storage medium
CN111488417A (en) * 2019-01-28 2020-08-04 阿里巴巴集团控股有限公司 Information processing method, system, device, equipment and computer storage medium
CN111667392A (en) * 2020-06-12 2020-09-15 成都国铁电气设备有限公司 Railway contact net defect hot spot area early warning method based on space-time clustering
CN111797295A (en) * 2020-06-19 2020-10-20 云从科技集团股份有限公司 Multi-dimensional space-time trajectory fusion method and device, machine readable medium and equipment
CN111881370A (en) * 2020-05-21 2020-11-03 北京嘀嘀无限科技发展有限公司 Method and system for describing contour of interest area
CN111897805A (en) * 2020-06-24 2020-11-06 东南大学 Hot spot path mining method based on longest common sub-track density clustering
CN113129328A (en) * 2021-04-22 2021-07-16 中国电子科技集团公司第二十九研究所 Target hotspot area fine analysis method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103559213A (en) * 2013-10-10 2014-02-05 河南大学 Efficient spatial nearest neighbor query method for highway networks
CN103578265A (en) * 2012-07-18 2014-02-12 北京掌城科技有限公司 Method for acquiring taxi-hailing hot spot based on taxi GPS data
CN106383868A (en) * 2016-09-05 2017-02-08 电子科技大学 Road network-based spatio-temporal trajectory clustering method
WO2017107800A1 (en) * 2015-12-24 2017-06-29 阿里巴巴集团控股有限公司 Method of acquiring route hotspot of traffic road and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103578265A (en) * 2012-07-18 2014-02-12 北京掌城科技有限公司 Method for acquiring taxi-hailing hot spot based on taxi GPS data
CN103559213A (en) * 2013-10-10 2014-02-05 河南大学 Efficient spatial nearest neighbor query method for highway networks
WO2017107800A1 (en) * 2015-12-24 2017-06-29 阿里巴巴集团控股有限公司 Method of acquiring route hotspot of traffic road and device
CN106383868A (en) * 2016-09-05 2017-02-08 电子科技大学 Road network-based spatio-temporal trajectory clustering method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LORENZ SCHAUER: "Discovering Hotspots:A Placement Strategy for Wi-Fi based Trajectory Monitoring within Buildings", 《IEEE CONFERENCE》 *
YUHAN DONG，ET AL: "A Novel Passenger Hotspots Searching Algorithm for Taxis in Urban Area", 《IEEE COMPUTER SOCIETY》 *

Cited By (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108133172A (en) * 2017-11-16 2018-06-08 北京华道兴科技有限公司 Method, the analysis method of vehicle flowrate and the device that Moving Objects are classified in video
CN108133172B (en) * 2017-11-16 2022-04-05 北京华道兴科技有限公司 Method for classifying moving objects in video and method and device for analyzing traffic flow
CN108427965B (en) * 2018-03-05 2022-08-23 重庆邮电大学 Hot spot area mining method based on road network clustering
CN108427965A (en) * 2018-03-05 2018-08-21 重庆邮电大学 A kind of hot spot region method for digging based on road network cluster
CN108804563A (en) * 2018-05-22 2018-11-13 阿里巴巴集团控股有限公司 A kind of data mask method, device and equipment
CN108804563B (en) * 2018-05-22 2021-11-19 创新先进技术有限公司 Data labeling method, device and equipment
CN109033011B (en) * 2018-06-19 2022-06-21 东软集团股份有限公司 Method and device for calculating track frequency, storage medium and electronic equipment
CN109033011A (en) * 2018-06-19 2018-12-18 东软集团股份有限公司 Calculate method, apparatus, storage medium and the electronic equipment of track frequency
CN109767615A (en) * 2018-10-19 2019-05-17 江苏智通交通科技有限公司 Road network traffic flow key flow direction and critical path analysis method
CN109767615B (en) * 2018-10-19 2021-05-18 江苏智通交通科技有限公司 Method for analyzing key flow direction and key path of road network traffic flow
CN109686085A (en) * 2018-12-17 2019-04-26 北京交通大学 Activity On the Node kind identification method is stopped based on GPS data vehicle transport dangerous goods
CN109711451A (en) * 2018-12-20 2019-05-03 成都四方伟业软件股份有限公司 A kind of data processing method, device, electronic equipment and storage medium
CN109615857A (en) * 2018-12-20 2019-04-12 首都师范大学 The deployment of roadside unit and dispatching method and device in city vehicle-mounted net
CN111380541A (en) * 2018-12-29 2020-07-07 沈阳美行科技有限公司 Interest point determination method and device, computer equipment and storage medium
CN111488417A (en) * 2019-01-28 2020-08-04 阿里巴巴集团控股有限公司 Information processing method, system, device, equipment and computer storage medium
CN111488417B (en) * 2019-01-28 2023-10-24 阿里巴巴集团控股有限公司 Information processing method, system, device, equipment and computer storage medium
CN109977109B (en) * 2019-04-03 2021-04-27 深圳市甲易科技有限公司 Track data accompanying analysis method
CN109977109A (en) * 2019-04-03 2019-07-05 深圳市甲易科技有限公司 A kind of track data cleaning method and adjoint analysis method
CN110174115A (en) * 2019-06-05 2019-08-27 武汉中海庭数据技术有限公司 A kind of method and device automatically generating high accuracy positioning map based on perception data
CN110457315A (en) * 2019-07-19 2019-11-15 国家计算机网络与信息安全管理中心 A kind of group's accumulation mode analysis method and system based on user trajectory data
CN110909037B (en) * 2019-10-09 2024-02-13 中国人民解放军战略支援部队信息工程大学 Frequent track mode mining method and device
CN110909037A (en) * 2019-10-09 2020-03-24 中国人民解放军战略支援部队信息工程大学 Frequent track mode mining method and device
CN110827540A (en) * 2019-11-04 2020-02-21 黄传明 Motor vehicle movement mode recognition method and system based on multi-mode data fusion
CN110827540B (en) * 2019-11-04 2021-03-12 黄传明 Motor vehicle movement mode recognition method and system based on multi-mode data fusion
CN111275963A (en) * 2020-01-14 2020-06-12 北京百度网讯科技有限公司 Method and device for mining hot spot area, electronic equipment and storage medium
CN111881370A (en) * 2020-05-21 2020-11-03 北京嘀嘀无限科技发展有限公司 Method and system for describing contour of interest area
CN111372188A (en) * 2020-05-27 2020-07-03 腾讯科技（深圳）有限公司 Method and device for determining hot spot track in area, storage medium and electronic device
CN111667392A (en) * 2020-06-12 2020-09-15 成都国铁电气设备有限公司 Railway contact net defect hot spot area early warning method based on space-time clustering
CN111667392B (en) * 2020-06-12 2023-06-16 成都国铁电气设备有限公司 Railway contact net defect hot spot area early warning method based on space-time clustering
CN111797295A (en) * 2020-06-19 2020-10-20 云从科技集团股份有限公司 Multi-dimensional space-time trajectory fusion method and device, machine readable medium and equipment
CN111797295B (en) * 2020-06-19 2021-04-02 云从科技集团股份有限公司 Multi-dimensional space-time trajectory fusion method and device, machine readable medium and equipment
CN111897805A (en) * 2020-06-24 2020-11-06 东南大学 Hot spot path mining method based on longest common sub-track density clustering
CN113129328A (en) * 2021-04-22 2021-07-16 中国电子科技集团公司第二十九研究所 Target hotspot area fine analysis method
CN113129328B (en) * 2021-04-22 2022-05-17 中国电子科技集团公司第二十九研究所 Target hotspot area fine analysis method

Also Published As

Publication number Publication date
CN107301254B (en) 2020-07-10

Similar Documents

Publication Publication Date Title
CN107301254A (en) A kind of road network hot spot region method for digging
CN106225791B (en) A kind of GPS positioning and road matching method based on grid dividing
CN103593430B (en) Clustering method based on mobile object spatiotemporal information trajectory subsections
CN103533501B (en) A kind of geography fence generation method
CN108320501B (en) Bus route identification method based on user mobile phone signaling
CN102521973B (en) A kind of mobile phone switches the road matching method of location
CN109614458B (en) Urban community structure mining method and device based on navigation data
CN103390169B (en) A kind of city terrain classification method of Vehicle-borne Laser Scanning cloud data
CN101630463B (en) Method for automatic vectorization of road network digital raster map
CN106912018A (en) Map-matching method and system based on signaling track
CN109697221A (en) Method for digging, device, storage medium and the electronic equipment of track rule
CN111966097B (en) Map building method, system and terminal based on grid map regional exploration
US11009362B2 (en) Generating trail network maps
CN110609824B (en) Hot spot area detection method based on dynamic space network model under urban road network environment
CN107679498A (en) A kind of airborne laser point cloud downtown roads recognition methods
CN106931974A (en) The method that personal Commuting Distance is calculated based on mobile terminal GPS location data record
CN101900565A (en) Path determining method and device
CN106920387A (en) Obtain the method and device of route temperature in traffic route
CN106604291A (en) RTK mobile reference station site selection method and device
CN104615881A (en) User normal track analysis method based on movable position application
KR101846294B1 (en) Rainfall center tracking method based on weather radar
CN107944089B (en) Land parcel height limit analysis system based on current situation vision corridor and analysis method thereof
CN109800903A (en) A kind of profit route planning method based on taxi track data
CN117573794A (en) Asian elephant road identification and road obstruction evaluation method and device
CN117516558A (en) Road network generation method, device, computer equipment and computer readable storage medium

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