CN110941278A - In-station dynamic security analysis method - Google Patents
In-station dynamic security analysis method Download PDFInfo
- Publication number
- CN110941278A CN110941278A CN201911328001.8A CN201911328001A CN110941278A CN 110941278 A CN110941278 A CN 110941278A CN 201911328001 A CN201911328001 A CN 201911328001A CN 110941278 A CN110941278 A CN 110941278A
- Authority
- CN
- China
- Prior art keywords
- track
- preset
- point
- motion
- analysis method
- 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
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0242—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using non-visible light signals, e.g. IR or UV signals
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Electromagnetism (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Alarm Systems (AREA)
Abstract
The embodiment of the invention provides an in-station dynamic security analysis method. The method comprises the steps of acquiring the motion track of each passenger in a station by adopting a location-based service LBS; classifying the motion tracks by adopting a preset classification algorithm according to track models of preset types, and counting the number of tracks corresponding to each track model; if the number of the tracks of a certain track model exceeds a preset alarm threshold value, alarm information is sent according to the statistical information of the motion tracks belonging to the certain track model.
Description
Technical Field
The invention relates to the technical field of rail transit, in particular to an in-station dynamic security analysis method.
Background
With the rapid development of subway construction in China, the complexity of a subway line network is also rapidly improved, and a large number of subway stations for multi-line (3 lines and more) transfer appear, such as three-line transfer stations of western straight gate station in Beijing, Song Jia Zhuang station, Futian station in Shenzhen and the like, and four-line transfer stations of century station in Shanghai, Longyang road station and the like. Inside the multi-line transfer station, because different lines are located in different platforms, the cross of the transfer lines in the station is caused, and great risks are brought to the safe operation of the subway under the impact of large passenger flows during the morning and evening peaks, holidays and major activities.
In order to deal with the risks, the method adopted at present is a manual + camera safety analysis mode, the number or distribution density of passenger groups in a station is detected through a camera, and then workers in the station conduct on-site dispersion and command. However, the installation position of the camera is affected by the space in the station, so that a large amount of shielding is caused, the detection range is limited, and the arrangement of a large amount of cameras brings complicated wiring, installation and debugging work. Whereas the process of manual supervision can only analyze the density of people in a single area.
Therefore, the existing safety analysis method cannot accurately predict the safety risk in the station.
Disclosure of Invention
Because the existing method has the problems, the embodiment of the invention provides an in-station dynamic security analysis method.
In a first aspect, an embodiment of the present invention provides an intra-site dynamic security analysis method, which is characterized by including:
acquiring the motion track of each passenger in the station by adopting a location-based service LBS;
classifying the motion tracks by adopting a preset classification algorithm according to track models of preset types, and counting the number of tracks corresponding to each track model;
and if the track quantity of a certain track model exceeds a preset alarm threshold value, sending alarm information according to the statistical information of the motion track belonging to the certain track model.
Further, if the number of the tracks of a certain track model exceeds a preset alarm threshold, sending alarm information according to the statistical information of the motion tracks belonging to the certain track model, specifically including:
if the track quantity of the certain track model exceeds the alarm threshold value, executing a preset track point clustering algorithm on each motion track belonging to the certain track model to obtain ROI (region of interest) and the track point quantity corresponding to each ROI;
analyzing the number of the track points corresponding to each ROI, and sending alarm information; wherein, the alarm information comprises an alarm reason and an alarm area.
Further, executing a preset track point clustering algorithm on each motion track belonging to the certain track model to obtain an ROI and the number of track points corresponding to each ROI, specifically including:
obtaining an entry point, a departure point and a stop point in each motion track according to a preset interest point analysis method; the entry point is a first track point of the motion track, the exit point is a last track point of the motion track, and the stop point is a track point obtained by screening according to a preset CB-SMoT algorithm;
respectively carrying out a preset DBSCAN algorithm on an entry point, a departure point and a stop point of each motion track to obtain ROIs and corresponding relations between each ROI and each entry point, departure point and stop point; the ROI comprises at least one entry region, at least one exit region and at least one stop region;
and counting to obtain the number of track points corresponding to each ROI according to the corresponding relation between each ROI and each entry point, exit point and stop point.
Further, after classifying each motion trajectory, the intra-station dynamic security analysis method further includes:
counting the motion tracks of the track models which do not belong to the preset types to obtain the number of abnormal tracks;
and judging the in-station confusion degree according to the number of the abnormal tracks.
Further, after the acquiring the motion trajectory of each passenger in the station by using the location based service LBS, the in-station dynamic security analysis method further includes:
executing preset data preprocessing on the motion trail, and eliminating the motion trail which does not meet preset filtering conditions; wherein the preset filtering conditions include:
the spatial distance and the temporal distance of the entry point and the exit point of the motion track exceed preset spatial threshold values and temporal threshold values;
the number of track points contained in the motion track exceeds a preset number threshold.
Further, after sending the alarm information, the in-station dynamic security analysis method further includes:
and obtaining a recommended passing path according to the alarm information.
Further, the in-station dynamic security analysis method further includes:
and obtaining a track model of the preset type by adopting a preset track clustering algorithm according to a pre-stored first training set containing historical motion tracks.
Further, the obtaining of the preset type of track model by using a preset track clustering algorithm according to a pre-stored first training set containing historical motion tracks specifically includes:
acquiring the first training set, executing preset data preprocessing on each historical motion track, and removing the historical motion tracks which do not meet preset filtering conditions to obtain a second training set;
executing the track point clustering algorithm on each historical motion track in the second training set to obtain an ROI corresponding to the second training set;
according to the ROI corresponding to the second training set, removing historical motion tracks which do not meet the condition of a preset complete track from the second training set to obtain a third training set; wherein the preset complete trajectory condition comprises: the entry point and the exit point of the historical motion track belong to an entry area and an exit area respectively;
executing a preset k-menas clustering algorithm on each historical motion track in the third training set to obtain k track types, and calculating prototype motion tracks corresponding to each track type;
combining the track types meeting preset combination conditions by adopting a preset combination algorithm according to the prototype motion track of each track type to obtain a track model of the preset type; wherein the number of the preset types is less than or equal to k.
Further, the classification algorithm is specifically a K-nearest neighbor KNN classification algorithm.
Further, the preset merging algorithm is specifically a connected subgraph search of an undirected graph.
According to the in-station dynamic security analysis method provided by the embodiment of the invention, the movement track of each passenger is obtained through the LBS technology, the corresponding track model is judged, and when the track quantity of the track model exceeds the alarm threshold value, alarm information is sent, so that the alarm prediction of the in-station security risk can be more accurately carried out.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a flow chart of a method for dynamic security analysis in a station according to an embodiment of the present invention;
FIG. 2 is a flow chart of another method for dynamic security analysis in a station according to an embodiment of the present invention;
FIG. 3 is a flow chart of a method for dynamic security analysis in a station according to another embodiment of the present invention;
fig. 4 is a flowchart of a dynamic security analysis method in a station according to another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart of an intra-site dynamic security analysis method according to an embodiment of the present invention, and as shown in fig. 1, the method includes:
step S01, acquiring the movement locus of each passenger in the station using the location based service LBS.
Location Based Services (LBS) technology is a technology for acquiring Location information of a user terminal through a mobile communication network.
The safety analysis system of the embodiment of the invention obtains the position information of the user terminal of each passenger in the station through the LBS technology, and takes the position information as the track point of the passenger to sequence on the map, thereby obtaining the movement track P ═ P of each passenger in the period from entering the station to leaving the station0,p1,…,pnAnd realizing the full-flow tracking of each single target in the station. Each locus point piThe corresponding location information includes at least an acquisition TimeiAnd three dimensional space coordinates (x)i,yi,zi). Therefore, the motion trail is a three-dimensional space coordinate sequence arranged according to the acquisition time.
And step S02, classifying the motion tracks by adopting a preset classification algorithm according to track models of preset types, and counting the track quantity corresponding to each track model.
The safety analysis system obtains track models of preset types in advance, each track model comprises a typical track corresponding to the track model, and the typical track is composed of a three-dimensional space coordinate sequence.
And matching the movement track of the passenger with the typical track of each track model through a preset classification algorithm. And determining a track model to which the motion track of each passenger belongs according to the matching result.
Further, the classification algorithm is specifically a K-nearest neighbor KNN classification algorithm.
There are many classification algorithms that can be used to match a motion trajectory with a typical trajectory, and only the K-nearest neighbor (KNN) classification algorithm is exemplified here, and the K value takes 1. The distances of the motion track and each typical track are calculated respectively, and the distance measure can be set according to actual needs, for example, the Euclidean distance can be used. And taking a track model corresponding to a typical track with the closest distance to the motion track as the type of the motion track. When calculating the euclidean distance between two tracks, if the number of track points included in the two tracks is different, the shorter track needs to be complemented to the same number by using the value of the last track point.
In order to increase the calculation speed when calculating the euclidean distance, the euclidean distance may be calculated after reducing the three-dimensional space coordinates into two-dimensional space coordinates by projecting the three-dimensional space coordinates in the Z bearing direction.
And after classifying the motion tracks, counting the track number of the motion tracks corresponding to the track models.
And step S03, if the track quantity of a certain track model exceeds a preset alarm threshold, sending alarm information according to the statistical information of the motion track belonging to the certain track model.
If the number of the tracks obtained by statistics of a certain track model exceeds a preset alarm threshold value in a preset period, it is determined that a large number of passengers move in a station along the typical track corresponding to the track model in a short period, and therefore, the area where the typical track corresponding to the track model passes has a safety risk and needs to send corresponding alarm information to related personnel.
The specific content of the alarm information can be determined by statistical information obtained by further performing statistical analysis on all motion tracks belonging to the track model.
The embodiment of the invention acquires the movement track of each passenger through the LBS technology to judge the corresponding track model, and sends alarm information when the track quantity of the track model exceeds the alarm threshold value, thereby more accurately carrying out alarm prediction on the safety risk in the station.
Fig. 2 is a flowchart of another intra-site dynamic security analysis method according to an embodiment of the present invention, and as shown in fig. 2, the step S03 specifically includes:
step S031, if the number of trajectories of the certain trajectory model exceeds the alarm threshold, a preset trajectory point clustering algorithm is performed on each motion trajectory belonging to the certain trajectory model to obtain an ROI and the number of trajectory points corresponding to each ROI.
When it is determined that the number of the trajectories of a certain trajectory model in the current period exceeds the alarm threshold, further analysis may be performed on all the motion trajectories belonging to the trajectory model obtained by statistics in the current period.
The method comprises the steps of counting track points contained in all motion tracks belonging to the track model, and executing a preset track point clustering algorithm to obtain a plurality of regions of Interest (ROI), wherein the ROI is a Region with more concentrated track points. Meanwhile, the corresponding track points of each ROI are counted to obtain the number of the track points of each ROI.
Step S032, analyzing the number of the track points corresponding to each ROI, and sending alarm information; wherein, the alarm information comprises an alarm reason and an alarm area.
By analyzing the number of the track points corresponding to each ROI and the position of the station corresponding to each ROI, the alarm reason of the alarm, the specific alarm area and other contents are determined and recorded in the alarm information for sending. The ROI corresponding to the inlet a and the outlet B is taken as an example for illustration:
if the number of the track points of the ROI corresponding to the inlet A is large, and the number of the track points of the ROI corresponding to the outlet B is also large, it is indicated that a large number of passengers enter from the inlet A, and meanwhile, a large number of passengers exit from the outlet B, and the alarm reason is that the subway station is in a point-to-point operation mode, and the inlet A and the outlet B need to be subjected to current-limiting dredging;
if the number of the track points of the ROI corresponding to the inlet A is large and the number of the track points of the ROI corresponding to the outlet B is small, the reason that the subway station is in a centralized entering operation mode of a certain entering station port for alarming is explained, and at the moment, current-limiting dredging needs to be performed on the inlet A;
if the number of the track points of the ROI corresponding to the inlet A is small and the number of the track points of the ROI corresponding to the outlet B is large, the reason that the alarm is caused is that the vicinity of the outlet B of the subway station is in an emergent passenger flow state is described, and at the moment, the flow-limiting dredging needs to be performed on the outlet B;
and if the number of the track points of the ROI corresponding to the inlet A is small and the number of the track points of the ROI corresponding to the outlet B is small, the fact that the alarm reason does not exist is shown, and the subway station is in the operation valley.
According to the embodiment of the invention, the ROI and the corresponding number of the track points are obtained by executing the track point clustering algorithm on all the motion tracks belonging to a certain track model, so that the alarm information comprising the subway operation mode and the alarm area is further analyzed and obtained, and the safety risk is more accurately predicted.
Based on the foregoing embodiment, further, step S031 specifically includes:
step 0311, according to a preset interest point analysis method, an entry point, a departure point and a stop point in each motion trajectory are obtained; the entry point is a first track point of the motion track, the exit point is a last track point of the motion track, and the stop point is a track point obtained by screening according to a preset CB-SMoT algorithm.
The specific calculation process of the track point clustering algorithm can be set according to actual needs, and only one example is given in the embodiment of the invention.
An interest point of each motion trajectory is obtained, and the interest point can be divided into an entry point, an exit point and a stop point.
The method for acquiring the entry point and the exit point is simple, and the first track point of all track points contained in the motion track can be directly used as the entry point of the motion track, and the last track point can be used as the exit point of the motion track.
And the stopping point reflects the staying time of the passenger near the track point, and can reflect the efficiency condition of the passenger passing the track point to a certain extent.
The stopping point can be obtained by adopting a track stopping point recognition algorithm CB-SMoT algorithm based on a correlation coefficient. By calculating the ith trace point piThe average passing speed of the track points with the preset number is front and back, and if the average passing speed is smaller than a preset speed threshold value, the track point piCan be called asThe core point is a stopping point. For example, if the preset number is 3, the average passing speed is:
said (x)i-1，yi-1，zi-1) And Timei-1Is a track point pi-1The three-dimensional space coordinates and the acquisition time of (x) are describedi+1，yi+1，zi+1) And Timei+1Is a track point pi+1Three-dimensional space coordinates and acquisition time.
The speed threshold value can be set according to a preset normal walking speed of the person, for example, 30% of the normal walking speed of the person, and 0.3 m/s.
Step 0312, performing a preset DBSCAN algorithm on the entry point, the exit point and the stop point of each motion trajectory to obtain ROIs and corresponding relations between the ROIs and the entry point, the exit point and the stop point; the ROI includes at least one entry region, at least one exit region, and at least one stop region.
Then, the entry points, exit points and stop points corresponding to all the motion trajectories are respectively executed to execute a preset Density-Based Clustering algorithm (DBSCAN), so as to obtain the ROI. Obtaining at least one entry area by performing DBSCAN on all entry points; all departure points are used as DBSCAN, and at least one departure area can be obtained; and DBSCAN is performed for all the stop points, so that at least one stop area can be obtained.
The DBSCAN clustering algorithm may find clusters in the data with noisy points. Specifically, the DBSCAN algorithm needs to give two parameters in advance, one is the sample point neighborhood radius Eps, and the other is the minimum point number MinPts in the neighborhood. For the trace point piIn other words, the number of trace points of a point in its neighborhood of Eps (i)>MinPts, then point of trace piReferred to as core points. If Eps (i)<MinPts, and trace point piNot in any other coreIn the Eps neighborhood of the center point, the track point piIs defined as a noise point. For core point piAnd pjAnd if the two track points are in the neighborhood of the other side, clustering is formed by the two core points and the track points in the neighborhood, the clustering relation is transferred along with neighborhood retrieval and is gradually expanded, finally, a region formed by each cluster is an ROI, and the corresponding relation between each ROI and an entry point, an exit point and a stop point is determined by marking corresponding ROI labels on the track points in the region.
The domain radius Eps and the number of minimum points MinPts in the neighborhood may be set according to the size of a specific station.
And step 0313, counting to obtain the number of track points corresponding to each ROI according to the corresponding relation between each ROI and each entry point, exit point and stop point.
And counting the track points with the same ROI label to obtain the number of the track points corresponding to each ROI.
According to the embodiment of the invention, the ROI and the number of the track points corresponding to each ROI are obtained by acquiring the entry point, the exit point and the stop point of each motion track and by the DBSCAN algorithm, and are used for obtaining subsequent alarm information, so that the safety risk can be more accurately predicted.
Based on the above embodiment, further, after classifying each motion trajectory, the intra-station dynamic security analysis method further includes:
counting the motion tracks of the track models which do not belong to the preset types to obtain the number of abnormal tracks;
and judging the in-station confusion degree according to the number of the abnormal tracks.
In the process of classifying the acquired motion trail through a preset classification algorithm, if a trail model matched with the motion trail is not found, the motion trail is classified as an abnormal trail and is used for representing that phenomena such as retrograde motion, transverse motion or fighting possibly occur. And counting the number of abnormal tracks of the abnormal tracks. Therefore, the passing chaotic programs in the station can be judged according to the number of the abnormal tracks, and corresponding alarm is given.
According to the embodiment of the invention, the movement track which does not belong to any track model is judged as the abnormal track, and the number of the abnormal tracks is counted, so that the degree of disorder of traffic in the station can be rapidly determined according to the number of the tracks of the anomalus poverty users, and the safety risk alarm can be timely carried out.
Fig. 3 is a flowchart of another intra-site dynamic security analysis method according to an embodiment of the present invention, and as shown in fig. 3, after step S01, the intra-site dynamic security analysis method further includes:
step S011, executing preset data preprocessing on the motion trail, and eliminating the motion trail which does not meet preset filtering conditions; wherein the preset filtering conditions include:
the spatial distance and the temporal distance of the entry point and the exit point of the motion track exceed preset spatial threshold values and temporal threshold values; the entry point is a first track point of the motion track, and the exit point is a last track point of the motion track;
the number of track points contained in the motion track exceeds a preset number threshold.
In the process of practical application, when the motion trajectory of each passenger is acquired according to the LBS technology, some special abnormal trajectories may be received, for example, the motion trajectory due to signal interruption or mistransmission, so that the data preprocessing may be performed on the motion trajectory through a preset filtering condition, so as to remove the motion trajectory that does not satisfy the filtering condition.
The filtering conditions can be set according to actual requirements, and only some examples are given in the embodiments of the present invention.
When obtaining the motion track P ═ P0,p1,…,pnAfter that, the first trace point p is compared0And the last trace point pnAnd obtaining the space distance and the time distance between the two track points, and comparing the space distance and the time distance with a preset space threshold and a preset time threshold.
If the spatial distance between the two track points is smaller than the spatial threshold, the motion track is too short, does not meet the filtering condition and needs to be removed;
if the time distance between the two track points is smaller than the time threshold, it is indicated that the time spent by the motion track is too short, the filtering condition is not met, and the motion track needs to be removed.
In addition, the number n +1 of the track points contained in the motion track can be counted and compared with a preset number threshold, and if the number of the track points is smaller than the number threshold, it is determined that the track points contained in the motion track are too few, do not meet the filtering condition, and need to be removed.
The spatial threshold, temporal threshold, and number threshold may be set according to the size and scale of the stations.
According to the embodiment of the invention, the acquired motion trail is subjected to data preprocessing to remove the motion trail which does not meet the preset filtering condition, so that the judgment and alarm on the safety risk are accelerated.
Based on the above embodiment, further, the in-station dynamic security analysis method further includes:
and obtaining a recommended passing path according to the alarm information.
According to the embodiment, when the safety analysis system judges that the safety risk exists, corresponding alarm information is sent out to give out specific alarm reasons and alarm areas. Therefore, when a new passenger enters the station, a recommended passage path can be sent to the passenger for avoiding the current alarm area.
According to the embodiment of the invention, the recommended passage path is obtained for the passenger according to the alarm information, so that the passenger can more quickly pass through the platform, and the further improvement of the safety risk is avoided.
Fig. 4 is a flowchart of a further intra-site dynamic security analysis method according to an embodiment of the present invention, and as shown in fig. 4, the intra-site dynamic security analysis method further includes:
and step S00, obtaining a track model of the preset type by adopting a preset track clustering algorithm according to a pre-stored first training set containing historical motion tracks.
According to the embodiment, before the obtained motion trajectories are classified, the preset types of trajectory models for classification are obtained. Specifically, the motion estimation method can be obtained by performing a preset trajectory clustering algorithm on a pre-stored first training set composed of a large number of historical motion trajectories.
In the practical application process, as the transportation develops, the tracked model also varies with the change of time or the surrounding environment. Thus, the trajectory model may be updated periodically.
Further, the step S00 specifically includes:
and S001, acquiring the first training set, executing preset data preprocessing on each historical motion track, and removing the historical motion tracks which do not meet preset filtering conditions to obtain a second training set.
S002, executing the track point clustering algorithm on each historical motion track in the second training set to obtain an ROI corresponding to the second training set;
s003, according to the ROI corresponding to the second training set, removing historical motion tracks which do not meet the condition of a preset complete track from the second training set to obtain a third training set; wherein the preset complete trajectory condition comprises: the entry point and the exit point of the historical motion trail belong to an entry area and an exit area respectively.
In order to ensure the integrity of the obtained track model as much as possible, therefore, each historical motion track needs to be cleaned and screened. And according to preset data preprocessing, removing the historical motion trail which does not meet the preset filtering condition from the first training set to obtain the first training set.
And executing a preset track point clustering algorithm on all historical motion tracks in the second training set to obtain the ROI corresponding to the second training set, wherein the ROI at least comprises an entering region and a leaving region.
And according to the corresponding relation between the entry point and the exit point of each historical motion track and the entry area and the exit area, if the entry point of the historical motion track does not belong to any entry area or the exit point does not belong to any exit area, judging that the historical motion track does not meet the preset complete track condition, and removing the historical motion track from the second training set. Thereby combining the remaining historical motion trajectories with the complete trajectory condition into a third training set.
And step S004, executing a preset k-menas clustering algorithm on each historical motion track in the third training set to obtain k track types, and calculating prototype motion tracks corresponding to each track type.
And executing a preset K-means clustering algorithm on each historical motion track in the third training set to obtain K track types. The specific method comprises the following steps:
firstly, the number of track points contained in all historical motion tracks in the first training set is complemented to the same length. Obtaining the similarity w between the historical motion tracks by calculating the Euclidean distance between the historical motion tracksij。
Of course, the similarity between the historical motion trajectories may also be performed by Dynamic Time Warping (DTW) or longest common subsequence algorithm (lcs), which is not specifically limited herein.
Constructing an adjacency matrix W ═ W according to the similarity between historical motion tracksijAnd then calculate the laplacian matrix:
Lsym＝I-D-1/2WD-1/2，
wherein the Laplac phase matrix LsymHas the following properties:
matrix LsymIs of definite symmetry;
matrix LsymHas a minimum feature value of 9, and the corresponding feature vector is
The feature vector is a column vector.
Matrix LsymWith M non-negative real eigenvalues, 0 ═ λ1≤λ2…≤λM。
To Laplace matrix LsymAnd (4) decomposing the eigenvalues, arranging the eigenvalues from small to large, selecting the first k eigenvalues and corresponding eigenvectors thereof, and forming an M multiplied by k matrix U. The rows of the matrix U can be regarded as new feature representation of original data after transformation, and finally k categories can be obtained by clustering the rows of the matrix U by using k-means, wherein the row number of the matrix U is equal to the total number of historical motion tracks, so that the category number of the rows of the matrix U is equal to the track type number of the historical motion tracks.
The k value in the k-means algorithm is the same as the column number k of the U matrix, the selection method of k is the same as the common rule of the k-means algorithm, and a relatively large value can be selected.
At this time, each historical motion track in the third training set is assigned to a corresponding track type, a historical motion track corresponding to each track type is extracted, included track points are averaged, and a prototype motion track of the track type is obtained through calculation.
Step S005, combining the track types meeting preset combination conditions by adopting a preset combination algorithm according to the prototype motion track of each track type to obtain a track model of the preset type; wherein the number of the preset types is less than or equal to k.
Because the k value is larger, the track types meeting the preset combination condition can be combined by adopting a preset combination algorithm according to the prototype motion track corresponding to each track type to reduce the number of classification.
Further, the preset merging algorithm is specifically a connected subgraph search of an undirected graph.
The preset merging algorithms are many, and the embodiment of the present invention only gives an example.
Abstracting a merging algorithm to solve the problem of connected subgraph search of an undirected graph: according to the prototype motion track of each track type, calculating a similarity matrix S ═ SijAnd (f) setting each prototype motion track as a node V of (V, E) in the graph GiIf two clusterings are presentV of typeiAnd vjSimilarity of (S)ijAbove a certain threshold ηeThen, the nodes are considered to be connected with the nodes, i.e. the corresponding trace types can be merged. After the combination, the track models of the preset types can be obtained.
According to the embodiment of the invention, the track model of the preset type is obtained by executing the preset track clustering algorithm on the pre-stored historical motion track and is used for classifying the motion track of each passenger, so that the safety risk can be predicted more quickly and accurately.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. An intra-site dynamic security analysis method, comprising:
acquiring the motion track of each passenger in the station by adopting a location-based service LBS;
classifying the motion tracks by adopting a preset classification algorithm according to track models of preset types, and counting the number of tracks corresponding to each track model;
and if the track quantity of a certain track model exceeds a preset alarm threshold value, sending alarm information according to the statistical information of the motion track belonging to the certain track model.
2. The in-station dynamic security analysis method according to claim 1, wherein if the number of trajectories of a certain trajectory model exceeds a preset alarm threshold, sending alarm information according to statistical information of a motion trajectory belonging to the certain trajectory model, specifically comprising:
if the track quantity of the certain track model exceeds the alarm threshold value, executing a preset track point clustering algorithm on each motion track belonging to the certain track model to obtain ROI (region of interest) and the track point quantity corresponding to each ROI;
analyzing the number of the track points corresponding to each ROI, and sending the alarm information; wherein, the alarm information comprises an alarm reason and an alarm area.
3. The intra-station dynamic security analysis method according to claim 2, wherein the step of performing a preset track point clustering algorithm on each motion track belonging to the certain track model to obtain ROIs of interest and the number of track points corresponding to each ROI specifically comprises:
obtaining an entry point, a departure point and a stop point in each motion track according to a preset interest point analysis method; the entry point is a first track point of the motion track, the exit point is a last track point of the motion track, and the stop point is a track point obtained by screening according to a preset CB-SMoT algorithm;
respectively carrying out a preset DBSCAN algorithm on an entry point, a departure point and a stop point of each motion track to obtain each ROI and corresponding relations between each ROI and each entry point, departure point and stop point; the ROI comprises at least one entry region, at least one exit region and at least one stop region;
and counting to obtain the number of track points corresponding to each ROI according to the corresponding relation between each ROI and each entry point, exit point and stop point.
4. The in-station dynamic security analysis method according to claim 3, wherein after classifying each motion trajectory, the in-station dynamic security analysis method further comprises:
counting the motion tracks of the track models which do not belong to the preset types to obtain the number of abnormal tracks;
and judging the in-station confusion degree according to the number of the abnormal tracks.
5. The in-station dynamic security analysis method according to claim 4, wherein after the acquiring a motion trajectory of each passenger in a station using the location based service LBS, the in-station dynamic security analysis method further comprises:
executing preset data preprocessing on the motion trail, and eliminating the motion trail which does not meet preset filtering conditions; wherein the preset filtering conditions include:
the spatial distance and the temporal distance of the entry point and the exit point of the motion track exceed preset spatial threshold values and temporal threshold values;
the number of track points contained in the motion track exceeds a preset number threshold.
6. The in-station dynamic security analysis method according to claim 5, wherein after sending the alarm information, the in-station dynamic security analysis method further comprises:
and obtaining a recommended passing path according to the alarm information.
7. The on-site dynamic security analysis method according to claim 6, further comprising:
and obtaining a track model of the preset type by adopting a preset track clustering algorithm according to a pre-stored first training set containing historical motion tracks.
8. The intra-station dynamic security analysis method according to claim 7, wherein the obtaining of the track model of the preset type by using a preset track clustering algorithm according to a pre-stored first training set containing historical motion tracks specifically comprises:
acquiring the first training set, executing preset data preprocessing on each historical motion track, and removing the historical motion tracks which do not meet preset filtering conditions to obtain a second training set;
executing the track point clustering algorithm on each historical motion track in the second training set to obtain an ROI corresponding to the second training set;
according to the ROI corresponding to the second training set, removing historical motion tracks which do not meet the condition of a preset complete track from the second training set to obtain a third training set; wherein the preset complete trajectory condition comprises: the entry point and the exit point of the historical motion track belong to an entry area and an exit area respectively;
executing a preset k-menas clustering algorithm on each historical motion track in the third training set to obtain k track types, and calculating prototype motion tracks corresponding to each track type;
combining the track types meeting preset combination conditions by adopting a preset combination algorithm according to the prototype motion track of each track type to obtain a track model of the preset type; wherein the number of the preset types is less than or equal to k.
9. The intra-station dynamic security analysis method according to claim 8, wherein the classification algorithm is specifically a K-nearest neighbor KNN classification algorithm.
10. The intra-site dynamic security analysis method according to claim 8, wherein the preset merging algorithm is a connected subgraph search of an undirected graph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911328001.8A CN110941278B (en) | 2019-12-20 | 2019-12-20 | Dynamic safety analysis method in station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911328001.8A CN110941278B (en) | 2019-12-20 | 2019-12-20 | Dynamic safety analysis method in station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110941278A true CN110941278A (en) | 2020-03-31 |
| CN110941278B CN110941278B (en) | 2023-05-23 |
Family
ID=69912647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911328001.8A Active CN110941278B (en) | 2019-12-20 | 2019-12-20 | Dynamic safety analysis method in station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110941278B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111523577A (en) * | 2020-04-13 | 2020-08-11 | 南京烽火星空通信发展有限公司 | Mass trajectory similarity calculation method based on improved LCSS algorithm |
| CN112268563A (en) * | 2020-10-23 | 2021-01-26 | 重庆越致科技有限公司 | Method and device for detecting traffic state and track of pedestrian on track |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080248815A1 (en) * | 2007-04-08 | 2008-10-09 | James David Busch | Systems and Methods to Target Predictive Location Based Content and Track Conversions |
| CN103327091A (en) * | 2013-06-17 | 2013-09-25 | 北京交通大学 | System and method for obtaining passenger track and behavioral parameter |
| CN104036139A (en) * | 2014-06-12 | 2014-09-10 | 中国科学院软件研究所 | Moving object trajectory monitoring method |
| CN104657424A (en) * | 2015-01-21 | 2015-05-27 | 段炼 | Clustering method for interest point tracks under multiple temporal and spatial characteristic fusion |
| CN106454716A (en) * | 2015-08-13 | 2017-02-22 | 株式会社理光 | Object analyzing method, object analyzing equipment and object analyzing system |
| CN106529815A (en) * | 2016-11-15 | 2017-03-22 | 同济大学 | Estimation method of passenger trip spatial-temporal trajectory of urban rail transit network and application thereof |
| GB201711408D0 (en) * | 2016-12-30 | 2017-08-30 | Maxu Tech Inc | Early entry |
| CN109445304A (en) * | 2018-09-25 | 2019-03-08 | 交控科技股份有限公司 | A kind of intelligent fault analysis system and method based on cab signal |
| WO2019061656A1 (en) * | 2017-09-30 | 2019-04-04 | 平安科技（深圳）有限公司 | Electronic apparatus, service place recommendation method based on lbs data, and storage medium |
| CN109784970A (en) * | 2018-12-13 | 2019-05-21 | 交控科技股份有限公司 | It is a kind of to be ridden the service recommendation method and device of data based on AFC passenger |
| CN109829119A (en) * | 2018-12-26 | 2019-05-31 | 成都熊谷油气科技有限公司 | Information method for early warning based on LBS big data in wisdom pipe network |
| CN109961106A (en) * | 2019-04-18 | 2019-07-02 | 北京百度网讯科技有限公司 | The training method and device of track disaggregated model, electronic equipment |
| CN110111566A (en) * | 2019-04-19 | 2019-08-09 | 腾讯科技（深圳）有限公司 | Trajectory predictions method, apparatus and storage medium |
| CN110309383A (en) * | 2019-06-17 | 2019-10-08 | 武汉科技大学 | Ship trajectory clustering analysis method based on improved DBSCAN algorithm |
| CN110517495A (en) * | 2019-09-05 | 2019-11-29 | 四川东方网力科技有限公司 | Confirmation method, device, equipment and the storage medium of track of vehicle classification |
| CN110570658A (en) * | 2019-10-23 | 2019-12-13 | 江苏智通交通科技有限公司 | Method for identifying and analyzing abnormal vehicle track at intersection based on hierarchical clustering |
-
2019
- 2019-12-20 CN CN201911328001.8A patent/CN110941278B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080248815A1 (en) * | 2007-04-08 | 2008-10-09 | James David Busch | Systems and Methods to Target Predictive Location Based Content and Track Conversions |
| CN103327091A (en) * | 2013-06-17 | 2013-09-25 | 北京交通大学 | System and method for obtaining passenger track and behavioral parameter |
| CN104036139A (en) * | 2014-06-12 | 2014-09-10 | 中国科学院软件研究所 | Moving object trajectory monitoring method |
| CN104657424A (en) * | 2015-01-21 | 2015-05-27 | 段炼 | Clustering method for interest point tracks under multiple temporal and spatial characteristic fusion |
| CN106454716A (en) * | 2015-08-13 | 2017-02-22 | 株式会社理光 | Object analyzing method, object analyzing equipment and object analyzing system |
| CN106529815A (en) * | 2016-11-15 | 2017-03-22 | 同济大学 | Estimation method of passenger trip spatial-temporal trajectory of urban rail transit network and application thereof |
| GB201711408D0 (en) * | 2016-12-30 | 2017-08-30 | Maxu Tech Inc | Early entry |
| WO2019061656A1 (en) * | 2017-09-30 | 2019-04-04 | 平安科技（深圳）有限公司 | Electronic apparatus, service place recommendation method based on lbs data, and storage medium |
| CN109445304A (en) * | 2018-09-25 | 2019-03-08 | 交控科技股份有限公司 | A kind of intelligent fault analysis system and method based on cab signal |
| CN109784970A (en) * | 2018-12-13 | 2019-05-21 | 交控科技股份有限公司 | It is a kind of to be ridden the service recommendation method and device of data based on AFC passenger |
| CN109829119A (en) * | 2018-12-26 | 2019-05-31 | 成都熊谷油气科技有限公司 | Information method for early warning based on LBS big data in wisdom pipe network |
| CN109961106A (en) * | 2019-04-18 | 2019-07-02 | 北京百度网讯科技有限公司 | The training method and device of track disaggregated model, electronic equipment |
| CN110111566A (en) * | 2019-04-19 | 2019-08-09 | 腾讯科技（深圳）有限公司 | Trajectory predictions method, apparatus and storage medium |
| CN110309383A (en) * | 2019-06-17 | 2019-10-08 | 武汉科技大学 | Ship trajectory clustering analysis method based on improved DBSCAN algorithm |
| CN110517495A (en) * | 2019-09-05 | 2019-11-29 | 四川东方网力科技有限公司 | Confirmation method, device, equipment and the storage medium of track of vehicle classification |
| CN110570658A (en) * | 2019-10-23 | 2019-12-13 | 江苏智通交通科技有限公司 | Method for identifying and analyzing abnormal vehicle track at intersection based on hierarchical clustering |
Non-Patent Citations (7)
| Title |
|---|
| LIU ZIHAN等: "Indoor Positioning System Based on the Improved", 《2017 IEEE 2ND ADVANCED INFORMATION TECHNOLOGY,ELECTRONIC AND AUTOMATION CONTROL CONFERENCE(IAEAC)》 * |
| 崔邓: "基于智能手机轨迹提取停留点的时空聚类算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 毛江云等: "路网空间下基于马尔可夫决策过程的异常车辆轨迹检测算法", 《计算机学报》 * |
| 王晓腾: "移动对象不确定性轨迹预测模型研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 章梦杰等: "多维空间中基于模式的移动对象轨迹预测", 《计算机系统应用》 * |
| 贾若然;刘曙光;孙启龙;: "基于位置轨迹数据的用户相似性分析" * |
| 高翔等: "基于LBS的定位系统的研究与设计", 《赤峰学院学报（自然科学版）》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111523577A (en) * | 2020-04-13 | 2020-08-11 | 南京烽火星空通信发展有限公司 | Mass trajectory similarity calculation method based on improved LCSS algorithm |
| CN112268563A (en) * | 2020-10-23 | 2021-01-26 | 重庆越致科技有限公司 | Method and device for detecting traffic state and track of pedestrian on track |
| CN112268563B (en) * | 2020-10-23 | 2022-03-04 | 重庆越致科技有限公司 | Method and device for detecting traffic state and track of pedestrian on track |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110941278B (en) | 2023-05-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Patrikar et al. | Anomaly detection using edge computing in video surveillance system | |
| Park et al. | Real-time prediction of secondary incident occurrences using vehicle probe data | |
| Saunier et al. | Probabilistic collision prediction for vision-based automated road safety analysis | |
| Morris et al. | A survey of vision-based trajectory learning and analysis for surveillance | |
| Xiang et al. | Video behavior profiling for anomaly detection | |
| CN111652912B (en) | Vehicle counting method and system, data processing equipment and intelligent shooting equipment | |
| CN111462488B (en) | Intersection safety risk assessment method based on deep convolutional neural network and intersection behavior characteristic model | |
| CN105469029A (en) | System and method for object re-identification | |
| CN113011367A (en) | Abnormal behavior analysis method based on target track | |
| CN110941278B (en) | Dynamic safety analysis method in station | |
| CN111582380A (en) | Ship track density clustering method and device based on space-time characteristics | |
| Shirazi et al. | A typical video-based framework for counting, behavior and safety analysis at intersections | |
| Abdullah et al. | Semantic segmentation based crowd tracking and anomaly detection via neuro-fuzzy classifier in smart surveillance system | |
| CN114612860A (en) | Computer vision-based passenger flow identification and prediction method in rail transit station | |
| CN114463390A (en) | Multi-twin-countermeasure network cross-camera vehicle tracking method with coupled motorcade following strengthening | |
| Lumentut et al. | Evaluation of recursive background subtraction algorithms for real-time passenger counting at bus rapid transit system | |
| CN111353342B (en) | Shoulder recognition model training method and device, and people counting method and device | |
| Chen et al. | Comogcn: Coherent motion aware trajectory prediction with graph representation | |
| CN114926422A (en) | Method and system for detecting boarding and alighting passenger flow | |
| CN111723664A (en) | Pedestrian counting method and system for open type area | |
| Kaviani et al. | A new method for traffic density estimation based on topic model | |
| CN111062294B (en) | Passenger flow queuing time detection method, device and system | |
| Liu et al. | Metro passenger flow statistics based on yolov3 | |
| CN115310682A (en) | Method, device, equipment and medium for dividing ship traffic in complex water area | |
| Wickramasinghe et al. | Pedestrian Detection, Tracking, Counting, Waiting Time Calculation and Trajectory Detection for Pedestrian Crossings Traffic light systems |
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 |