CN110781958B - OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation - Google Patents
OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation Download PDFInfo
- Publication number
- CN110781958B CN110781958B CN201911022154.XA CN201911022154A CN110781958B CN 110781958 B CN110781958 B CN 110781958B CN 201911022154 A CN201911022154 A CN 201911022154A CN 110781958 B CN110781958 B CN 110781958B
- Authority
- CN
- China
- Prior art keywords
- flow
- flow direction
- similarity
- spanning tree
- subtree
- 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
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000011218 segmentation Effects 0.000 title claims abstract description 18
- 238000004364 calculation method Methods 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 18
- 238000010276 construction Methods 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 9
- 240000007594 Oryza sativa Species 0.000 claims 1
- 235000007164 Oryza sativa Nutrition 0.000 claims 1
- 235000009566 rice Nutrition 0.000 claims 1
- 238000013461 design Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 235000006629 Prosopis spicigera Nutrition 0.000 description 1
- 240000000037 Prosopis spicigera Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009440 infrastructure construction Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to an OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation, which comprises the following steps: step S1, collecting the track data to be measured; step S2, extracting original flow direction data from the track data and preprocessing the original flow direction data to form an OD flow direction set F; step S3, calculating the similarity value among OD flow directions in the OD flow direction set F, constructing a maximum weight spanning tree MST of the F, and cutting the MST to obtain a plurality of subtree CTs and noise flow directions which have no similarity relation with each other; step S4, carrying out self-similarity judgment on the subtrees, organizing the flow directions in the subtrees into OD flow direction clusters to be output if the self-similarity of the subtrees meets the preset cluster output standard, otherwise, carrying out step S5; step S5: and cutting the subtrees which do not meet the output standard of the cluster by adopting an iterative optimal binary segmentation method until the cut preset subtrees meet the output standard of the cluster and are organized into OD flow direction cluster output. The invention can effectively improve the operation efficiency and reduce the calculation time.
Description
Technical Field
The invention relates to an OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation.
Background
How to mine the travel characteristics of urban residents is important to traffic planning design and management. With the rapid development and popularization of mobile positioning technology, large-data-volume geographic movement data such as human daily activity trajectory data, group migration data, vehicle trajectory data and the like are more and more easily acquired. The OD flow direction data is special mobile data, only position information of origin and Destination is reserved but actual track information is ignored, the flow direction and hot points of resident travel can be identified by clustering the vehicle OD flow direction data with large data volume level, the travel characteristics of urban residents, the space-time connection of group flow among different areas of the city and the change trend of the space-time connection can be accurately grasped, and the OD flow direction data has important significance for traffic planning design and city management.
Disclosure of Invention
In view of the above, the present invention provides an OD flow direction clustering method based on maximum spanning tree and optimal graph partitioning, which combines with the construction and cutting of the maximum weight spanning tree to partition data into a plurality of subtrees without similarity relationship, and each subtree can perform distributed processing respectively, thereby effectively improving the operation efficiency and reducing the computation time.
In order to realize the purpose, the invention adopts the following technical scheme:
an OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation comprises the following steps:
step S1, collecting the track data to be measured;
step S2, extracting the serial number of each track data, the longitude and latitude coordinate information and the time information of the O point and the D point from the track data to be detected to form original flow direction data, preprocessing all the flow direction data to form an OD flow direction set F;
step S3, calculating the similarity value among OD flow directions in the OD flow direction set F, constructing a maximum weight spanning tree MST of the F, and cutting the MST to obtain a plurality of subtree CTs and noise flow directions which have no similarity relation with each other;
step S4, self-similarity judgment is carried out on the subtree, if the self-similarity of the subtree meets the preset cluster output standard, the flow direction in the subtree is organized into OD flow direction clusters to be output, otherwise, the step S5 is carried out;
step S5: and cutting the subtrees which do not meet the output standard of the cluster by adopting an iterative optimal binary segmentation method until the cut preset subtrees meet the output standard of the cluster and are organized into OD flow direction cluster outputs.
Further, the step S3 is specifically:
step S31, calculating the similarity value among the OD flow directions in the OD flow direction set F;
step S32, regarding each flow in the flow set F as a vertex, regarding a connecting line between the two vertices as an edge, and regarding a similarity value between the flow as a weight between the vertices;
step S33, the maximum weight spanning tree MST is constructed and partitioned into subtrees and a noise flow direction.
Further, the step S31 is specifically:
the method for calculating the similarity between OD flow directions is as follows:
sim(Fi，Fj)＝1
-func(ratioO)*func(ratioD)*func(ratioTime)/23
ratioO＝dist(Oi，Oj)/disLimit
ratioD＝dist(Di，Dj)/disLimit
ratioTime＝span(timei，timej)/timeLimit
wherein, dist (O)i,Oj) In the direction of flow FiAnd FjDistance at point O, dist (D)i,Dj) In the direction of flow FiAnd FjDistance at point D, span (time)i,timej) In the direction of flow FiAnd FjThe time difference of getting on the train or getting off the train is manually input time parameters, the unit is minutes or hours, the distance is a spatial similarity parameter, and the calculation method comprises the following steps:
wherein len (F)i) In the direction of flow FiLength of (k)>3. In order to prevent the OD flow length from being too long and the spatial characteristics of the clustering result from being very fuzzy, the invention adds a limiting condition that when the flow length is more than 5000m, the DISLimit is a fixed value of 5000/k (unit: meter)
When sim (F)i，Fj)∈[0，0.875]The flow directions Fi and Fj are similar in space-time, and sim (F)i，Fj) The larger, the flow direction FiAnd FjThe higher the degree of spatiotemporal similarity.
Further, the step S33 is specifically:
s331, selecting any vertex as a construction starting point of a maximum weight spanning tree (MST);
and S332, selecting the edge with the maximum weight between the vertex in the maximum weight spanning tree and the isolated vertex, and adding the edge and the corresponding vertex into the MST.
And S333, repeating the step 2 until all the vertexes are added into the MST, and finishing the construction process of the MST.
And step S334, the edges with the weight less than 0 in the MST are broken to obtain a plurality of subtrees CT without similarity relation and noise flow direction.
Further, the self-similarity judgment is performed on the subtree, and a self-similarity function Css (CT) of the subtree CT is specifically as follows:
wherein, numCTThe number of vertices (flow direction) in the subtree CT.
Further, the step S5 is specifically:
step S51: constructing an undirected graph G from the vertexes in the subtree CT, and weighting values w of the edges of the undirected graphijWhen the similarity value between the vertexes is smaller than 0, no edge connection exists between the vertexes, the weight is set to be 0, and a weight matrix W is as follows:
where n is the number of vertices in the sub-tree CT. For any point V in undirected graph GiIts degree diDefined as the sum of the weights of all edges connected to it, i.e.
With the definition of each vertex degree, a degree matrix D is obtained. The degree matrix D is a diagonal matrix, i.e.
The normalized laplacian matrix L is calculated from the above matrices, and the calculation formula is as follows:
step S52: calculating the eigenvalue of the normalized Laplace matrix L, taking the second smallest eigenvalue λ2And the corresponding feature vector x2＝D-1/2α2；
Step S53: feature vector x is clustered by using k-means algorithm2Into two classes, using feature vectors x2The corresponding vertex is divided into two parts by the classification information of (2), and a subtree CT1 and a subtree CT2 which correspond to the vertex are obtained;
step S54: and respectively judging whether the subtree CT1 and the subtree CT2 meet preset cluster output standards, if so, organizing OD flow directions corresponding to all vertexes in the CT1 or the CT2 into OD flow direction clusters and outputting the OD flow direction clusters, and if not, iteratively executing a step S5.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention combines the construction and the cutting of the maximum weight spanning tree to divide the data into a plurality of subtrees without similarity relation, each subtree can respectively carry out distributed processing, the operation efficiency can be effectively improved, and the calculation time can be reduced
2. The invention combines the optimal binary segmentation method of iteration and the flow direction cluster output standard, ensures that the clustering result has sufficient intra-cluster similarity and inter-cluster heterogeneity, and ensures that the clustering result is more convincing.
Drawings
FIG. 1 is a flow diagram for the construction and cutting of a maximum spanning tree in one embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method according to an embodiment of the present invention;
FIG. 3 illustrates OD flow data in accordance with an embodiment of the present invention;
FIG. 4 shows the spatio-temporal co-clustering results of OD flow directions in an embodiment of the present invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
Referring to fig. 2, the present invention provides an OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation, including the following steps:
step S1, collecting track data to be measured;
in the embodiment, the original flow data is formed by extracting the serial number of each piece of track data, longitude and latitude coordinate information of the O point and the D point and time information from the track data of a drop-out rental car in metropolis of 11 months and 1 days in 2016;
step S2, extracting the serial number of each track data, the longitude and latitude coordinate information and the time information of the O point and the D point from the track data to be detected to form original flow direction data, preprocessing all the flow direction data to form an OD flow direction set F;
in this embodiment, all flow direction data are preprocessed to form an OD flow direction set F; the pretreatment is as follows: and taking a division map of the research area range as a base map, overlapping the original flow direction data with the map, and eliminating the OD flow direction of the O point or the D point outside the research area. The obtained OD flow data is shown in fig. 3.
Step S3, calculating the similarity value among OD flow directions in the OD flow direction set F, constructing a maximum weight spanning tree MST of the F, and cutting the MST to obtain a plurality of subtree CTs and noise flow directions which have no similarity relation with each other;
in this embodiment, the step S3 specifically includes:
step S31, calculating the similarity value among the OD flow directions in the OD flow direction set F;
in this embodiment, the step S31 specifically includes:
the method for calculating the similarity between OD flow directions is as follows:
sim(Fi，Fj)＝1
-func(ratioO)*func(ratioD)*func(ratioTime)/23
ratioO＝dist(Oi，Oj)/disLimit
ratioD＝dist(Di，Dj)/disLimit
ratioTime＝span(timei，timej)/timeLimit
wherein, dist (O)i,Oj) In the direction of flow FiAnd FjDistance at point O, dist (D)i,Dj) In the direction of flow FiAnd FjDistance at point D, span (time)i,timej) In the direction of flow FiAnd FjThe time difference of getting on the train or getting off the train is manually input time parameters, the unit is minutes or hours, the distance is a spatial similarity parameter, and the calculation method comprises the following steps:
wherein len (F)i) In the direction of flow FiLength of (k)>3. In order to prevent the spatial characteristics of the clustering result from being very fuzzy due to the excessively long flow direction length of the OD, the invention adds a limiting condition that when the flow direction length is more than 5000m, the distimit is a fixed value 5000/k (unit: meter)
When sim (F)i，Fj)∈[0，0.875]The flow directions Fi and Fj are similar in space-time, and sim (F)i，Fj) The larger, the flow direction FiAnd FjThe higher the degree of spatiotemporal similarity.
Step S32, regarding each flow in the flow set F as a vertex, regarding a connecting line between the two vertices as an edge, and regarding a similarity value between the flow as a weight between the vertices;
step S33, the maximum weight spanning tree MST is constructed and partitioned into subtrees and a noise flow direction.
In this embodiment, the step S33 specifically includes:
s331, selecting any vertex as a construction starting point of a maximum weight spanning tree (MST);
step S332, selecting the edge with the maximum weight between the vertex in the maximum weight spanning tree and the isolated vertex, and adding the edge and the corresponding vertex into the MST.
And S333, repeating the step 2 until all the vertexes are added into the MST, and finishing the construction process of the MST.
And step S334, the edges with the weight less than 0 in the MST are broken to obtain a plurality of subtrees CT without similarity relation and noise flow direction.
Step S4, self-similarity judgment is carried out on the subtree, if the self-similarity of the subtree meets the preset cluster output standard, the flow direction in the subtree is organized into OD flow direction clusters to be output, otherwise, the step S5 is carried out;
in this embodiment, the Self-Similarity of the subtree is determined by using a Self-Similarity determination function (Child Tree Self-Similarity Criterion), which is as follows:
wherein numCTThe number of vertices (flow direction) in the subtree CT.
The self-similarity Css (CT) of the sub-tree CT being equal to 1 means that CT satisfies the cluster-like output criterion, organizes the flow direction within CT into OD flow direction and outputs, and if not, performs step S5.
Step S5: and cutting the subtrees which do not meet the output standard of the cluster by adopting an iterative optimal binary segmentation method until the cut preset subtrees meet the output standard of the cluster and are organized into OD flow direction cluster output.
In this embodiment, the iterative optimal binary segmentation step is as follows:
1. constructing an undirected graph G from the vertexes in the subtree CT, and weighting values w of the edges of the undirected graphijFor similarity between corresponding vertices i and jAnd when the similarity value between the vertexes is less than 0, no edge is connected between the vertexes, and the weight is set to be 0. The weight matrix W is:
where n is the number of vertices in the sub-tree CT. For any point V in undirected graph GiIts degree diDefined as the sum of the weights of all edges connected to it, i.e.
With the definition of each vertex degree, a degree matrix D is obtained. The degree matrix D is a diagonal matrix, i.e.
The normalized laplacian matrix L is calculated from the above matrices, and the calculation formula is as follows:
2. calculating the eigenvalue of the normalized Laplace matrix L, taking the second smallest eigenvalue λ2And the corresponding feature vector x2＝D-1/2α2。
3. Feature vector x is clustered by using k-means algorithm2Into two classes, using feature vectors x2The corresponding vertex is divided into two parts by the classification information of (1), and a subtree CT1 and a subtree CT2 are obtained corresponding to the vertex.
4. And respectively judging whether the subtree CT1 and the subtree CT2 meet cluster output standards, if so, organizing OD flow directions corresponding to all vertexes in the CT1 or the CT2 into OD flow direction clusters and outputting the OD flow direction clusters, and if not, iteratively executing a step S4.
The above specific implementation steps and parameter settings are followed to obtain the spatial-temporal joint clustering result of the OD flow direction, showing the clustering result at all times, as shown in fig. 4. Each flow in fig. 4 has a time attribute, the time attribute of the class cluster is represented by color, and the number attribute of the class cluster is represented by thickness.
By combining the specific implementation mode and the case, the method can effectively implement space-time combined clustering on the OD flow direction data, so that the travel modes of urban residents at different moments can be extracted from disordered OD data, and reasonable suggestions are provided for urban traffic allocation and infrastructure construction.
The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (6)
1. An OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation is characterized by comprising the following steps:
step S1, collecting the track data to be measured;
step S2, extracting the serial number of each track data, the longitude and latitude coordinate information and the time information of the O point and the D point from the track data to be detected to form original flow direction data, preprocessing all the flow direction data to form an OD flow direction set F;
step S3, calculating the similarity value among OD flow directions in the OD flow direction set F, constructing a maximum weight spanning tree MST of the F, and cutting the MST to obtain a plurality of subtree CTs and noise flow directions which have no similarity relation with each other;
step S4, self-similarity judgment is carried out on the subtree, if the self-similarity of the subtree meets the preset cluster output standard, the flow direction in the subtree is organized into OD flow direction clusters to be output, otherwise, the step S5 is carried out;
step S5: and cutting the subtrees which do not meet the output standard of the cluster by adopting an iterative optimal binary segmentation method until the cut preset subtrees meet the output standard of the cluster and are organized into OD flow direction cluster outputs.
2. The OD flow clustering method based on the maximum spanning tree and the optimal graph segmentation as claimed in claim 1, wherein the step S3 specifically comprises:
step S31, calculating the similarity value among the OD flow directions in the OD flow direction set F;
step S32, regarding each flow in the flow set F as a vertex, regarding a connecting line between the two vertices as an edge, and regarding a similarity value between the flow as a weight between the vertices;
step S33, the maximum weight spanning tree MST is constructed and partitioned into subtrees and a noise flow direction.
3. The OD flow clustering method based on the maximum spanning tree and the optimal graph segmentation as claimed in claim 2, wherein the step S31 specifically comprises:
the method for calculating the similarity between OD flow directions is as follows:
sim(Fi，Fj)＝1-func(ratioO)*func(ratioO)*func(ratioTime)/23
ratioO＝dist(Oi，Oj)/disLimit
ratioD＝dist(Di，Dj)/disLimit
ratioTime＝span(timei，timej)/timeLimit
wherein, dist (O)i,Oj) In the direction of flow FiAnd FjDistance at point O, dist (D)i,Dj) In the direction of flow FiAnd FjDistance at point D, span (time)i,timej) In the direction of flow FiAnd FjThe time difference of getting on the train or getting off the train is manually input time parameters, the unit is minutes or hours, the distance is a spatial similarity parameter, and the calculation method comprises the following steps:
wherein len (F)i) In the direction of flow FiLength of (k)>3; in order to prevent the OD flow length from being too long, the spatial characteristics of the clustering result will be very fuzzy, and a limiting condition is added, that is, when the flow length is greater than 5000m, the distimit is a fixed value of 5000/k, unit: rice;
when sim (F)i，Fj)∈[0，0.875]The flow directions Fi and Fj are similar in space-time, and sim (F)i，Fj) The larger, the flow direction is to FiAnd FjThe higher the degree of spatiotemporal similarity.
4. The OD flow clustering method based on the maximum spanning tree and the optimal graph segmentation as claimed in claim 2, wherein the step S33 specifically comprises:
s331, selecting any vertex as a construction starting point of a maximum weight spanning tree (MST);
s332, selecting the edge with the maximum weight between the peak in the maximum weight spanning tree and the isolated peak, and adding the edge and the corresponding peak into the MST;
step S333, repeating the step 2 until all the vertexes are added into the MST, and finishing the construction process of the MST;
and step S334, the edges with the weight less than 0 in the MST are broken to obtain a plurality of subtrees CT without similarity relation and noise flow direction.
5. The maximum spanning tree and optimal graph segmentation based OD flow direction clustering method of claim 4, wherein the self-similarity judgment is performed on the sub-tree, and the self-similarity function CSS (CT) of the sub-tree CT is adopted as follows:
wherein, numCTThe number of vertices in the subtree CT.
6. The OD flow clustering method based on the maximum spanning tree and the optimal graph segmentation as claimed in claim 1, wherein the step S5 specifically comprises:
step S51: constructing an undirected graph from the vertexes in the subtree CT, wherein the weight value w of the edge of the undirected graphijWhen the similarity value between the vertexes is less than 0, the vertexes are connected without edges, and the weight is set to be 0; the weight matrix W is:
wherein n is the number of vertexes in the subtree CT; for any point V in the undirected graphiIts degree diDefined as the sum of the weights of all edges connected to it, i.e.
Obtaining a degree matrix D by using the definition of each vertex degree; the degree matrix D is a diagonal matrix and,
namely, it is
The normalized laplacian matrix L is calculated from the above matrices, and the calculation formula is as follows:
step S52: calculating the eigenvalue of the normalized Laplace matrix L, and taking the second smallest eigenvalue lambda2And the corresponding feature vector x2＝D-1/2α2；
Step S53: using k-means clusteringNormal feature vector x2Into two classes, using feature vectors x2The corresponding vertex is divided into two parts by the classification information of (2), and a subtree CT1 and a subtree CT2 which correspond to the vertex are obtained;
step S54: and respectively judging whether the sub-tree CT1 and the sub-tree CT2 meet preset cluster output standards, if so, organizing OD flow directions corresponding to vertexes in the CT1 or the CT2 into OD flow direction clusters and outputting the OD flow directions, and if not, iteratively executing a step S5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911022154.XA CN110781958B (en) | 2019-10-25 | 2019-10-25 | OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911022154.XA CN110781958B (en) | 2019-10-25 | 2019-10-25 | OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110781958A CN110781958A (en) | 2020-02-11 |
| CN110781958B true CN110781958B (en) | 2022-06-17 |
Family
ID=69387810
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911022154.XA Expired - Fee Related CN110781958B (en) | 2019-10-25 | 2019-10-25 | OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110781958B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105933091A (en) * | 2016-04-18 | 2016-09-07 | 青岛海尔智能家电科技有限公司 | Minimum weight triangulation method and device based on spatial network coding |
| CN106875314A (en) * | 2017-01-31 | 2017-06-20 | 东南大学 | A kind of Urban Rail Transit passenger flow OD method for dynamic estimation |
| CN107748896A (en) * | 2017-10-30 | 2018-03-02 | 陕西师范大学 | A kind of multi-level main body of Urban population flows to generation method |
| CN110111575A (en) * | 2019-05-16 | 2019-08-09 | 北京航空航天大学 | A kind of Forecast of Urban Traffic Flow network analysis method based on Complex Networks Theory |
| CN110232421A (en) * | 2019-06-21 | 2019-09-13 | 福州大学 | One kind merging OD step by step and flows to space-time joint clustering method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7646895B2 (en) * | 2005-04-05 | 2010-01-12 | 3Vr Security, Inc. | Grouping items in video stream images into events |
| US7773538B2 (en) * | 2008-05-16 | 2010-08-10 | At&T Intellectual Property I, L.P. | Estimating origin-destination flow entropy |
| CN106845530B (en) * | 2016-12-30 | 2018-09-11 | 百度在线网络技术（北京）有限公司 | character detection method and device |
-
2019
- 2019-10-25 CN CN201911022154.XA patent/CN110781958B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105933091A (en) * | 2016-04-18 | 2016-09-07 | 青岛海尔智能家电科技有限公司 | Minimum weight triangulation method and device based on spatial network coding |
| CN106875314A (en) * | 2017-01-31 | 2017-06-20 | 东南大学 | A kind of Urban Rail Transit passenger flow OD method for dynamic estimation |
| CN107748896A (en) * | 2017-10-30 | 2018-03-02 | 陕西师范大学 | A kind of multi-level main body of Urban population flows to generation method |
| CN110111575A (en) * | 2019-05-16 | 2019-08-09 | 北京航空航天大学 | A kind of Forecast of Urban Traffic Flow network analysis method based on Complex Networks Theory |
| CN110232421A (en) * | 2019-06-21 | 2019-09-13 | 福州大学 | One kind merging OD step by step and flows to space-time joint clustering method |
Non-Patent Citations (1)
| Title |
|---|
| "基于多源异构数据的城市路网动态车流OD估计";周东;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑 C034-657》;20190515;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110781958A (en) | 2020-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107292276B (en) | Vehicle-mounted point cloud clustering method and system | |
| CN106650809B (en) | A kind of vehicle-mounted laser point cloud objective classification method and system | |
| CN101551905B (en) | Method for segmenting multi-dimensional texture image on basis of fuzzy C-means clustering and spatial information | |
| CN110443810B (en) | Point cloud plane segmentation method based on quick adjacent voxel query | |
| CN111091618B (en) | Combined indoor layout method and system based on house type data driving | |
| CN110399819A (en) | A kind of remote sensing image residential block extraction method based on deep learning | |
| CN109508585A (en) | A method of urban function region is extracted based on POI and high-resolution remote sensing image | |
| CN103956042B (en) | A kind of intelligence of the public bicycles dispatcher-controlled territory based on graph theory division methods | |
| CN110263717A (en) | It is a kind of incorporate streetscape image land used status determine method | |
| CN107182036A (en) | The adaptive location fingerprint positioning method merged based on multidimensional characteristic | |
| CN101901343A (en) | Remote sensing image road extracting method based on stereo constraint | |
| CN104820718A (en) | Image classification and searching method based on geographic position characteristics and overall situation vision characteristics | |
| CN109214422B (en) | Parking data repairing method, device, equipment and storage medium based on DCGAN | |
| CN103258001B (en) | A kind of radio frequency map that embeds algorithm based on local linear is without supervised classification method | |
| CN110363178B (en) | Airborne laser point cloud classification method based on local and global depth feature embedding | |
| CN107396374A (en) | A kind of covering method based on fictitious force and Thiessen polygon | |
| CN109493344A (en) | A kind of semantic segmentation method of large-scale city three-dimensional scenic | |
| CN116129118B (en) | Urban scene laser LiDAR point cloud semantic segmentation method based on graph convolution | |
| CN111985389A (en) | Basin similarity discrimination method based on basin attribute distance | |
| CN111536973A (en) | Indoor navigation network extraction method | |
| CN103500278B (en) | The similar Typhoon Analysis method in path based on shortest path first | |
| CN108717642B (en) | City mall site selection method based on GIS | |
| CN111611960B (en) | Large-area ground surface coverage classification method based on multilayer perceptive neural network | |
| CN108898244A (en) | A kind of digital signage position recommended method coupling multi-source element | |
| CN110781958B (en) | OD flow direction clustering method based on maximum spanning tree and optimal graph segmentation |
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 | ||
| 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: 20220617 |