CN111259444A - Track data label clustering method fusing privacy protection - Google Patents
Track data label clustering method fusing privacy protection Download PDFInfo
- Publication number
- CN111259444A CN111259444A CN202010048296.XA CN202010048296A CN111259444A CN 111259444 A CN111259444 A CN 111259444A CN 202010048296 A CN202010048296 A CN 202010048296A CN 111259444 A CN111259444 A CN 111259444A
- Authority
- CN
- China
- Prior art keywords
- label
- node
- track data
- generalization
- 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
Images
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
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
- G06F21/6254—Protecting personal data, e.g. for financial or medical purposes by anonymising data, e.g. decorrelating personal data from the owner's identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioethics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a track data tag clustering method fusing privacy protection, which combines vehicle track data mining with privacy protection of interest points in tracks; in the process of label propagation, firstly, carrying out global generalization treatment on the vehicle identification number for the stored label sequence to realize anonymous treatment on a single vehicle track; for longitude and latitude coordinate columns accurate to position points, local generalization processing is carried out on interest points by combining stop point data in vehicle track data, privacy protection on all high-frequency stop points is further realized by adopting a global generalization processing method for other points, vehicle identification numbers and GPS coordinate attribute semantic information in the track data are considered in the clustering process, vehicle track data mining and generalization processing are combined, and sensitive information in the vehicle track data is subjected to concealment processing, so that the privacy information involved in the vehicle track clustering process can be effectively protected.
Description
Technical Field
The invention relates to the field of vehicle track data mining and privacy protection, in particular to a track data tag clustering method fusing privacy protection.
Background
With the development and popularization of vehicle-mounted wireless sensing equipment, vehicle track data information which can be collected is improved. The track clustering method is one of key technologies for analyzing and applying track data. The traditional track clustering method mainly adopts a method based on distance measurement, but has the defects of inaccurate clustering, high estimation cost, lack of privacy protection mechanism and the like. The track clustering method based on semantic analysis proposed in recent years effectively improves clustering precision, but privacy protection is still not considered in the clustering process. Therefore, how to protect privacy sensitive data from being leaked while performing track data mining becomes a problem which needs to be solved urgently.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a track data tag clustering method fusing privacy protection, which can effectively protect privacy information involved in a vehicle track clustering process.
In order to achieve the purpose, the invention adopts the following technical scheme:
a track data tag clustering method fusing privacy protection comprises the following steps:
step 1), data preprocessing: performing dimensionality reduction on the most original track data acquired from the vehicle-mounted sensor;
step 2), conversion of a dual graph: modeling a road section between two intersections in a road network into a node according to the most original track data after the dimension reduction processing, and modeling the intersection between every two nodes into a connecting line between the nodes, thereby converting the road network into a dual graph;
step 3), initializing label information of all nodes to enable each node to have a unique label, setting a sequence for storing historical labels for each node, and then carrying out label propagation on the initialized label information by an SLPA label propagation method to finish label clustering;
step 4), generalizing the sequence of the stored historical labels;
step 5), removing low-frequency tags: according to the label clustering result after generalization processing in the step 3) and the step 4), if the frequency of the label clustering result is lower than a preset threshold value, deleting the label; otherwise, directly outputting the final clustering result.
Further, removing abnormal data from the most original track data acquired by the vehicle-mounted sensor equipment, and simultaneously extracting the GPS coordinate attribute from the most original track data set as a semantic label of the track; and then, carrying out Gaussian projection on the original track data, and converting the longitude and latitude coordinates in the original track into geodetic coordinates, thereby completing the dimension reduction processing of the most original track data.
Further, the original trajectory data is subjected to Gaussian projection by a Gaussian projection coordinate conversion tool.
Further, in step 2), a node is randomly selected and recorded as a monitoring node, the node is set as a current node, and a neighbor node is recorded as a propagation node;
the probability of random selection of each propagation node of the current node is proportional to the probability of occurrence (P) in the current node's stored sequencei) And sending the label to the listening node;
wherein, listiRepresents the number of occurrences of the selected tag in the current stored sequence, Σ listiDenotes the sum of the occurrences of all tags in the current tag sequence, 0<Pi<＝1；
The current node selects the label with the maximum occurrence probability from the label information set transmitted by the transmission node to be added into a label list stored by the monitoring node, and the label with the maximum occurrence probability is used as a new label in the iteration;
and repeating the iteration process to enable the label to be transmitted in the continuous traversal process until convergence or traversal reaches the set times, and ending the iteration.
Further, the specific steps in the step 4) include: if the attribute column identification of the vehicle identification number is a standard identifier column, carrying out global generalization processing on the vehicle identification number; if the attribute column identification of the vehicle identity identification number is not the quasi-identifier column, judging whether the attribute column identification of the vehicle identity identification number is an interest point, and if the attribute column identification of the vehicle identity identification number is the interest point, performing local generalization processing on the position coordinates of the interest point.
Further, if the attribute column identification of the vehicle identification number is not a quasi-identifier column, the following steps are performed:
4.2.1 counting the number of the values of each position point in the label sequence, and taking out the interest point coordinates of which the counted number is more than or equal to a threshold value 10;
4.2.2 the coordinates of the interest points with the statistical number more than or equal to the threshold value 10 are sorted in descending order according to the size of the statistical number;
4.2.3, selecting the interest point coordinate with the largest statistical number according to the arrangement sequence in the step 4.2.2 to carry out first generalization treatment;
4.2.4 if the tag sequence after the first generalization does not meet the generalization treatment result, skipping to the step 4.2.1, otherwise skipping to the step 4.2.5;
4.2.5, performing iterative computation until all the interest point coordinates which are larger than or equal to the threshold value are completely processed by local generalization, otherwise, skipping to the step 4.2.2;
and 4.3, performing global generalization on the position coordinates of the common points by using generalization processing.
Further, the ordinary point refers to a point other than the point of interest in the position coordinate data.
Further, global generalization refers to performing the same generalization operation on the same group of data at a time.
Further, local generalization refers to performing different generalization operations for different subsets in the same set of data.
Compared with the prior art, the invention has the following beneficial technical effects:
the invention relates to a track data label clustering method integrating privacy protection, which is characterized in that the most original track data acquired from a vehicle-mounted sensor is subjected to dimensionality reduction, vehicle track data mining and privacy protection of interest points in tracks are combined, a road section between two intersections in a road network is modeled into a node according to the most original track data subjected to dimensionality reduction, and the intersection between every two nodes is modeled into a connecting line between the nodes, so that the road network is converted into a dual graph; in the process of label propagation, firstly, carrying out global generalization treatment on the vehicle identification number for the stored label sequence to realize anonymous treatment on a single vehicle track; for longitude and latitude coordinate columns accurate to position points, local generalization processing is carried out on interest points by combining stop point data in vehicle track data, privacy protection on all high-frequency stop points is further realized by adopting a global generalization processing method for other points, vehicle identification numbers and GPS coordinate attribute semantic information in the track data are considered in the clustering process, vehicle track data mining and generalization processing are combined, and sensitive information in the vehicle track data is subjected to concealment processing, so that the privacy information involved in the vehicle track clustering process can be effectively protected.
And for the longitude and latitude coordinate column accurate to the position point, local generalization processing is carried out on the interest point by combining the stopping point data in the vehicle track data, and privacy protection on all high-frequency stopping points is further realized by adopting a global generalization processing method for other points.
Drawings
FIG. 1 is a block flow diagram of a method as described in an example of the invention.
Fig. 2 is a graph of the conversion of a road network into a dual graph according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
as shown in fig. 1 and fig. 2, the track data tag clustering method with privacy protection fused according to the present invention, as shown in fig. 1, specifically includes the following steps,
step 1, preprocessing data;
1.1, cleaning the most original track data acquired from the vehicle-mounted sensor equipment to obtain original track data;
the cleaning treatment specifically refers to removing abnormal data from the most original track data, and simultaneously extracting GPS coordinate attributes from the most original track data set as semantic tags of tracks; the GPS coordinate attributes include vehicle speed, heading, and altitude.
1.2, carrying out Gaussian projection on the original track data (namely the cleaned most original track data) through a Gaussian projection coordinate conversion tool, and converting longitude and latitude coordinates in the original track into geodetic coordinates so as to finish dimension reduction processing on the most original track data;
original trajectory: representing space-time sampling points (V) for a vehicle over a time intervali，Xi_T，Yi_TAnd T) in the sequence. Wherein, Xi_TIndicating vehicle ViLongitude coordinate at time T, Yi_TIndicating vehicle ViLatitude coordinate at time T.
Step 2, conversion of a dual graph:
2.1 modeling a road section between two intersections in a road network into a node according to the most original track data after the dimension reduction processing, and modeling the intersection between every two nodes into a connecting line between the nodes;
2.2 according to the modeling mode in step 2.1, converting the road network into a dual graph, as shown in fig. 2;
3.1 initializing label information of all nodes, enabling each node to have a unique label, and setting a sequence for storing historical labels for each node; the tag information refers to track data fused with track data semantic information, and comprises the speed, the course and the height of the vehicle; the tag information may also be a custom semantic tag, including vehicle track number;
3.2, carrying out a label propagation process, which comprises the following specific steps:
3.2.1 randomly selecting a node, recording as a monitoring node, setting the node as a current node, and recording a neighbor node as a propagation node;
3.2.2 random selection probability of each propagating node of the current node is proportional to the probability of occurrence (P) in the current node memory sequencei) And sending the label to the listening node;
wherein, listiRepresents the number of occurrences of the selected tag in the current stored sequence, Σ listiDenotes the sum of the occurrences of all tags in the current tag sequence, 0<Pi<＝1；
3.2.3 the current node selects the label with the maximum occurrence probability from the label information set transmitted by the transmission node to be added into a label list stored by the monitoring node, and takes the label with the maximum occurrence probability as a new label in the iteration;
3.2.4 repeating the iteration process until convergence or traversal reaches the set times (the set times are generated by user predefining), ending the iteration, otherwise, skipping to the step 3.2, and enabling the label to be spread in the continuous traversal process;
and 4, performing k anonymization processing on the sequence of the history tags stored in the step 3.1, wherein the k anonymization processing refers to generalization processing of data, namely, unified representation of different data. Because the vehicle track data has slightly different values in adjacent time periods, different attribute values are represented in a unified way, so that an attacker cannot distinguish interest points from common points, and the purpose of privacy protection is achieved;
4.1, global generalization processing is carried out on the vehicle identification number by using k anonymization processing;
4.1.1 identifying the attribute column of the vehicle identification number as a quasi identifier column;
4.1.2 combining the characteristic that the vehicle identity identification number in the original track data of the vehicle can expose the identity information of the vehicle, carrying out global generalization treatment on the vehicle identity identification number, wherein the global generalization means that the same generalization operation is carried out on the same group of data at one time; step 5 is carried out after the global generalization treatment;
4.2, carrying out local generalization processing on the position coordinates of the interest points by using k anonymization processing; the interest points refer to sampling points which are combined with vehicle track data, and the times of appearance of the interest points in the label sequence at the same position (the longitude coordinate value and the latitude coordinate value are equal) exceed a preset threshold; the threshold is set to 10 for this application.
4.2.1 counting the number of the values of each position point in the label sequence, and taking out the interest point coordinates of which the counted number is more than or equal to a threshold value 10;
4.2.2 the coordinates of the interest points with the statistical number more than or equal to the threshold value 10 are sorted in descending order according to the size of the statistical number;
4.2.3, selecting the interest point coordinate with the largest statistical number according to the arrangement sequence in the step 4.2.2 to carry out first generalization treatment;
4.2.4 detecting the label sequence after the first generalization, if the label sequence does not accord with the generalization processing result (in the embodiment, each recorded data is at least the same as the attribute value of other k-1 data), skipping to the step 4.2.1, otherwise, skipping to the step 4.2.5;
4.2.5, performing iterative computation until all the interest point coordinates which are larger than or equal to the threshold value are completely subjected to local generalization processing, otherwise, skipping to the step 4.2.2;
4.3, global generalization processing is carried out on the position coordinates of the common points by using k anonymization processing, and the step 5 is carried out after the global generalization processing;
step 5, removing the low-frequency label; according to the label clustering result in the label sequence in the step 3, if the frequency of the label is lower than a preset threshold value, deleting the label; otherwise, finishing the algorithm and outputting the final clustering result.
Claims (9)
1. A track data label clustering method fusing privacy protection is characterized by comprising the following steps:
step 1), data preprocessing: performing dimensionality reduction on the most original track data acquired from the vehicle-mounted sensor;
step 2), conversion of a dual graph: modeling a road section between two intersections in a road network into a node according to the most original track data after the dimension reduction processing, and modeling the intersection between every two nodes into a connecting line between the nodes, thereby converting the road network into a dual graph;
step 3), initializing label information of all nodes to enable each node to have a unique label, setting a sequence for storing historical labels for each node, and then carrying out label propagation on the initialized label information by an SLPA label propagation method to finish label clustering;
step 4), generalizing the sequence of the stored historical labels;
step 5), removing low-frequency tags: according to the label clustering result after generalization processing in the step 3) and the step 4), if the frequency of the label clustering result is lower than a preset threshold value, deleting the label; otherwise, directly outputting the final clustering result.
2. The track data tag clustering method integrating privacy protection is characterized in that abnormal data are removed from the most original track data acquired from vehicle-mounted sensor equipment, and simultaneously GPS coordinate attributes are extracted from the most original track data set to serve as semantic tags of tracks; and then, carrying out Gaussian projection on the original track data, and converting the longitude and latitude coordinates in the original track into geodetic coordinates, thereby completing the dimension reduction processing of the most original track data.
3. The method for clustering trajectory data tags fusing privacy protection as claimed in claim 2, wherein the original trajectory data is subjected to gaussian projection by a gaussian projection coordinate transformation tool.
4. The track data tag clustering method fusing the privacy protection as claimed in claim 1, wherein in step 2), a node is randomly selected and marked as a monitoring node, the node is set as a current node, and a neighbor node is marked as a propagation node;
the probability of random selection of each propagation node of the current node is proportional to the probability of random selection of the current nodeProbability of occurrence P in node storage sequenceiAnd sending the label to the listening node;
wherein, listiRepresents the number of occurrences of the selected tag in the current stored sequence, Σ listiDenotes the sum of the occurrences of all tags in the current tag sequence, 0<Pi<＝1；
The current node selects the label with the maximum occurrence probability from the label information set transmitted by the transmission node to be added into a label list stored by the monitoring node, and the label with the maximum occurrence probability is used as a new label in the iteration;
and repeating the iteration process to enable the label to be transmitted in the continuous traversal process until convergence or traversal reaches the set times, and ending the iteration.
5. The track data tag clustering method fusing privacy protection as claimed in claim 1, wherein the specific steps in step 4) include: if the attribute column identification of the vehicle identification number is a standard identifier column, carrying out global generalization processing on the vehicle identification number; if the attribute column identification of the vehicle identity identification number is not the quasi-identifier column, judging whether the attribute column identification of the vehicle identity identification number is an interest point, and if the attribute column identification of the vehicle identity identification number is the interest point, performing local generalization processing on the position coordinates of the interest point.
6. The method for clustering trajectory data tags fusing privacy protection according to claim 5, wherein if the attribute column identification of the vehicle identification number is not a quasi-identifier column, the following steps are performed:
4.2.1 counting the number of the values of each position point in the label sequence, and taking out the interest point coordinates of which the counted number is more than or equal to a threshold value 10;
4.2.2 the coordinates of the interest points with the statistical number more than or equal to the threshold value 10 are sorted in descending order according to the size of the statistical number;
4.2.3, selecting the interest point coordinate with the largest statistical number according to the arrangement sequence in the step 4.2.2 to carry out first generalization treatment;
4.2.4 if the tag sequence after the first generalization does not meet the generalization treatment result, skipping to the step 4.2.1, otherwise skipping to the step 4.2.5;
4.2.5, performing iterative computation until all the interest point coordinates which are larger than or equal to the threshold value are completely processed by local generalization, otherwise, skipping to the step 4.2.2;
and 4.3, performing global generalization on the position coordinates of the common points by using generalization processing.
7. The method of claim 6, wherein the common points refer to points in the position coordinate data other than the interest points.
8. The method for clustering trajectory data labels fusing privacy protection as claimed in claim 6, wherein the global generalization is to perform the same generalization operation on the same group of data at a time.
9. The method for clustering trajectory data labels fusing privacy protection as claimed in claim 6, wherein the local generalization specifically performs different generalization operations for different subsets in the same group of data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010048296.XA CN111259444B (en) | 2020-01-16 | 2020-01-16 | Track data label clustering method fusing privacy protection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010048296.XA CN111259444B (en) | 2020-01-16 | 2020-01-16 | Track data label clustering method fusing privacy protection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111259444A true CN111259444A (en) | 2020-06-09 |
| CN111259444B CN111259444B (en) | 2022-09-16 |
Family
ID=70948865
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010048296.XA Active CN111259444B (en) | 2020-01-16 | 2020-01-16 | Track data label clustering method fusing privacy protection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111259444B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112559593A (en) * | 2020-12-14 | 2021-03-26 | 长安大学 | Localized differential privacy protection method based on label clustering |
| CN112668040A (en) * | 2020-12-14 | 2021-04-16 | 长安大学 | Track clustering privacy protection method based on semantics |
| CN112801131A (en) * | 2020-12-17 | 2021-05-14 | 长安大学 | Semantic track anonymous region construction method based on density clustering |
| CN112818402A (en) * | 2021-02-26 | 2021-05-18 | 华南理工大学 | Method for realizing k anonymity of track data release based on point density segmentation track |
| CN113946867A (en) * | 2021-10-21 | 2022-01-18 | 福建工程学院 | Position privacy protection method based on space influence |
| CN113946867B (en) * | 2021-10-21 | 2024-05-31 | 福建工程学院 | Position privacy protection method based on space influence |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160180705A1 (en) * | 2014-12-18 | 2016-06-23 | Jing Liu | Origin destination estimation based on vehicle trajectory data |
| CN106383868A (en) * | 2016-09-05 | 2017-02-08 | 电子科技大学 | Road network-based spatio-temporal trajectory clustering method |
| CN106650486A (en) * | 2016-09-28 | 2017-05-10 | 河北经贸大学 | Trajectory privacy protection method in road network environment |
-
2020
- 2020-01-16 CN CN202010048296.XA patent/CN111259444B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160180705A1 (en) * | 2014-12-18 | 2016-06-23 | Jing Liu | Origin destination estimation based on vehicle trajectory data |
| CN106383868A (en) * | 2016-09-05 | 2017-02-08 | 电子科技大学 | Road network-based spatio-temporal trajectory clustering method |
| CN106650486A (en) * | 2016-09-28 | 2017-05-10 | 河北经贸大学 | Trajectory privacy protection method in road network environment |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112559593A (en) * | 2020-12-14 | 2021-03-26 | 长安大学 | Localized differential privacy protection method based on label clustering |
| CN112668040A (en) * | 2020-12-14 | 2021-04-16 | 长安大学 | Track clustering privacy protection method based on semantics |
| CN112801131A (en) * | 2020-12-17 | 2021-05-14 | 长安大学 | Semantic track anonymous region construction method based on density clustering |
| CN112818402A (en) * | 2021-02-26 | 2021-05-18 | 华南理工大学 | Method for realizing k anonymity of track data release based on point density segmentation track |
| CN113946867A (en) * | 2021-10-21 | 2022-01-18 | 福建工程学院 | Position privacy protection method based on space influence |
| CN113946867B (en) * | 2021-10-21 | 2024-05-31 | 福建工程学院 | Position privacy protection method based on space influence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111259444B (en) | 2022-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111259444B (en) | Track data label clustering method fusing privacy protection | |
| CN109936582B (en) | Method and device for constructing malicious traffic detection model based on PU learning | |
| CN105701123B (en) | The recognition methods of man-vehicle interface and device | |
| CN107977734B (en) | Prediction method based on mobile Markov model under space-time big data | |
| CN111506599B (en) | Industrial control equipment identification method and system based on rule matching and deep learning | |
| Chen et al. | T-DBSCAN: A Spatiotemporal Density Clustering for GPS Trajectory Segmentation. | |
| CN112309126B (en) | License plate detection method and device, electronic equipment and computer readable storage medium | |
| CN113052235B (en) | Method and device for detecting abnormal stay of ship | |
| CN104111073A (en) | Method and device for identifying inaccurate paths in map data | |
| CN110162997B (en) | Anonymous privacy protection method based on interpolation points | |
| Karagiorgou et al. | A layered approach for more robust generation of road network maps from vehicle tracking data | |
| CN108961082A (en) | A kind of vehicle insurance loss assessment system and method based on AI image recognition | |
| CN112885099A (en) | Method, device and equipment for determining key points of vehicle track | |
| Moosavi et al. | Characterizing driving context from driver behavior | |
| CN110727740B (en) | Correlation analysis method and device, computer equipment and readable medium | |
| CN110553658B (en) | Navigation path recommendation method, navigation server, computer device and readable medium | |
| CN112559593A (en) | Localized differential privacy protection method based on label clustering | |
| CN115392937A (en) | User fraud risk identification method and device, electronic equipment and storage medium | |
| CN102855635A (en) | Method and device for determining human body action cycles and recognizing human body actions | |
| CN111540202B (en) | Similar bayonet determining method and device, electronic equipment and readable storage medium | |
| CN117455237A (en) | Road traffic accident risk prediction method based on multi-source data | |
| CN104121917A (en) | Method and device for automatically discovering new bridge | |
| Wang et al. | Accurate Detection of Road Network Anomaly by Understanding Crowd's Driving Strategies from Human Mobility | |
| CN114398462A (en) | Destination recommendation method and system based on multi-source heterogeneous information network | |
| CN114003674A (en) | Double-recording address determination method, device, equipment and 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 |