A method and device for analyzing footholds based on multi-source traffic data

Abstract

The present invention discloses a method for analyzing a foothold based on multi-source traffic data. A controller pre-processes and sorts the passing vehicle data, positioning trajectory data and ticket information obtained at each preset time interval, and obtains the traffic trajectory points of a target object within each preset time interval. The traffic trajectory points are clustered and analyzed using a DBSCAN algorithm to obtain the center points of the subsection foothold areas of the target object within each preset time interval. The foothold of the target object is obtained based on the center points of the subsection foothold areas within each preset time interval, and the traffic trajectory points of each preset time interval are fused. The traffic trajectory points are clustered and analyzed using the DBSCAN algorithm to obtain the foothold of the target object. This method greatly reduces the cost of manual confirmation and improves the efficiency of obtaining the foothold of the target object.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for analyzing a foothold based on multi-source traffic data. Background Art

[0002] When the police are solving a case, they usually need to obtain the whereabouts of the suspect. In order to save a lot of personnel investigation costs, artificial intelligence technology can be used to assist the police in solving the case. There are two main types of traditional personnel or vehicle whereabouts analysis systems: the first type usually uses face and vehicle recognition combined with key area monitoring; the second type uses offline big data analysis, manual confirmation, and algorithm tuning.

[0003] The first solution relies on the experience of business personnel. Through local reports or existing data, a number of key locations are screened out, and long-term monitoring is carried out through manual monitoring or face matching to finally determine whether the location is the place where the suspicious person has settled. It has strong limitations and relies heavily on the experience and professionalism of business personnel. In addition, the monitoring statistical analysis requires a lot of manpower and machine computing power. The fault tolerance rate is low. If a region appears for the first time, the progress may be slow due to misjudgment.

[0004] The second solution is to build an algorithm model through sample data when building a data warehouse. After the algorithm model is initially confirmed, offline analysis of massive positioning data is performed, and the algorithm results are manually confirmed to clarify the tuning direction and modify the algorithm parameters. This process is repeated until the offline calculation results meet the business needs. This solution directly performs offline calculations on massive positioning data, which consumes a lot of resources, and the algorithm tuning has a high time cost and trial and error cost. Summary of the invention

[0005] The purpose of the present invention is to provide a method for analyzing a foothold based on multi-source traffic data in order to solve the above-mentioned problem, as described in detail below.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides a method for analyzing a destination point based on multi-source traffic data, comprising real-time acquisition of the passing data of a target vehicle driven by a target object, the positioning trajectory data of the target vehicle, and the ticket information of the target object; preprocessing and sorting the passing data, the positioning trajectory data, and the ticket information acquired at each preset time interval to acquire the traffic trajectory points of the target object within each preset time interval; clustering analysis of the traffic trajectory points using a DBSCAN algorithm to acquire the center point of a segmented destination point area of ​​the target object within each preset time interval; and acquiring the destination point of the target object based on the center point of the segmented destination point area within each preset time interval.

[0008] By adopting the above-mentioned landing point analysis method based on multi-source traffic data, the controller pre-processes and sorts the passing vehicle data, positioning trajectory data and ticketing information obtained at each preset time interval, and obtains the traffic trajectory points of the target object within each preset time interval; the DBSCAN algorithm is used to perform cluster analysis on the traffic trajectory points to obtain the center points of the segmented landing point areas of the target object within each preset time interval; based on the center points of the segmented landing point areas within each preset time interval, the landing point of the target object is obtained, the traffic trajectory points of each preset time interval are fused, and then the DBSCAN algorithm is used to perform cluster analysis on the traffic trajectory points to obtain the landing point of the target object, which greatly reduces the cost of manual confirmation and improves the efficiency of obtaining the landing point of the target object.

[0009] Preferably, the real-time acquisition of the passing data of the target vehicle driven by the target object comprises the following steps: retrieving the comprehensive passing data of the target vehicle from the database of the vehicle management office, the comprehensive passing data including the position information of the checkpoint passed by the target vehicle, the facial capture feature data of the time of passing the checkpoint; matching the facial capture feature data with the facial feature data of the target object, filtering out the target object from the comprehensive passing data, and acquiring the passing data of the target vehicle driven by the target object.

[0010] Preferably, the vehicle passing data, the positioning trajectory data and the ticketing information obtained based on each preset time interval are preprocessed and sorted, comprising the following steps: using a preset sliding window to obtain the vehicle passing data, the positioning trajectory data and the ticketing information obtained at each preset time interval to obtain the traffic travel volume of the target object; the traffic travel volume includes the travel starting point, travel end point, travel mode and travel time period of the target object; cleaning each item of data in the traffic travel volume and sorting them in chronological order to obtain the effective traffic travel volume.

[0011] Preferably, the travel starting point and the travel end point in each travel itinerary are parsed from the effective traffic travel volume; and a travel point set is obtained based on the travel starting point and the travel end point.

[0012] Preferably, the use of the DBSCAN algorithm to perform cluster analysis on the traffic trajectory points to obtain the center point of the target object's stop area in each of the preset time periods includes the following steps: using the DBSCAN algorithm to perform cluster analysis on the travel point set to obtain the first center cluster point; based on the first center cluster point, obtaining the center point of the target object's stop area.

[0013] Preferably, the step of obtaining the target object's landing point based on the center points of the landing point areas within each preset time period comprises the following steps: performing cluster analysis on the N center points of the landing point areas within the preset time period using the DBSCAN algorithm to obtain a second center cluster point; and obtaining the target object's landing point area result based on the second center cluster point.

[0014] Preferably, the method of using a preset sliding window to obtain the vehicle passing data, the positioning trajectory data and the ticketing information obtained at each preset time interval to obtain the traffic volume of the target object includes the following steps: using a first preset sliding window to obtain the vehicle passing data, the positioning trajectory data and the ticketing information within a first preset time interval to obtain the traffic volume of the target object; obtaining the destination of the target object based on the center point of the subsection footprint area within each of the preset time intervals, including the following steps: using a DBSCAN algorithm to perform cluster analysis on the traffic trajectory points to obtain the center point of the subsection footprint area of ​​the target object within the first preset time interval; adjusting the tuning parameters of the DBSCAN algorithm based on the difference between the center point of the subsection footprint area and the actual destination of the target vehicle to obtain a first DBSCAN algorithm; using the first DBSCAN algorithm to obtain the traffic volume obtained by the second preset sliding window to obtain the center point of the subsection footprint area of ​​the target object within the second preset time interval; repeating the above steps until the center point of the subsection footprint area of ​​the target object within the Nth preset time interval is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a schematic diagram of the process of the foothold analysis based on multi-source traffic data of the present invention;

[0017] Figure 2 It is a flow chart of the trajectory point acquisition step of the present invention;

[0018] Figure 3 It is a flow chart of the center point acquisition step of the present invention. DETAILED DESCRIPTION

[0019] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0020] See also Figure 1 As shown, the present invention provides a method for analyzing a foothold based on multi-source traffic data, comprising the following steps:

[0021] S1. Real-time acquisition of the passing data of the target vehicle driven by the target object, the positioning trajectory data of the target vehicle, and the ticketing information of the target object;

[0022] The vehicle passing data refers to the location information of the checkpoints passed by the target vehicle in each trip of the target vehicle and the facial capture feature data when passing the checkpoints.

[0023] The process of obtaining the vehicle passing data of the target vehicle driven by the target object includes: retrieving the vehicle passing data of the target vehicle from the vehicle management office database, wherein the vehicle passing data includes the position information of the checkpoint through which the target vehicle passes, and the facial capture feature data of the time when the target vehicle passes the checkpoint;

[0024] The facial feature data of the captured face is matched with the facial feature data of the target object, and the target object is screened out from the comprehensive vehicle passing data to obtain the vehicle passing data of the target vehicle driven by the target object.

[0025] S2, pre-processing and sorting the vehicle passing data, positioning trajectory data and ticketing information obtained at each preset time interval to obtain the traffic trajectory points of the target object within each preset time interval;

[0026] A preset sliding window is used to obtain vehicle passing data, positioning trajectory data and ticketing information at each preset time interval to obtain the target object's traffic volume; the traffic volume includes the target object's travel starting point, travel end point, travel mode and travel time period; each data in the traffic volume is cleaned and sorted in chronological order to obtain the effective traffic volume.

[0027] S3, using DBSCAN algorithm to perform cluster analysis on traffic trajectory points, and obtain the center point of the target object's segmented foothold area within each preset time length;

[0028] The starting point and destination of each trip are parsed from the effective traffic volume; based on the starting point and destination, a travel point set is obtained; the DBSCAN algorithm is used to perform cluster analysis on the travel point set to obtain the first central cluster point; based on the first central cluster point, the central point of the target object's destination area is obtained.

[0029] S4. Based on the center point of the sub-paragraph foothold area within each preset time length, the foothold of the target object is obtained.

[0030] The DBSCAN algorithm is used to perform cluster analysis on the center points of the landing point areas within N preset time lengths to obtain the second center cluster points; based on the second center cluster points, the landing point area results of the target object are obtained.

[0031] By adopting the above-mentioned landing point analysis method based on multi-source traffic data, the controller pre-processes and sorts the passing vehicle data, positioning trajectory data and ticketing information obtained at each preset time interval, and obtains the traffic trajectory points of the target object within each preset time interval; the DBSCAN algorithm is used to perform cluster analysis on the traffic trajectory points to obtain the center points of the segmented landing point areas of the target object within each preset time interval; based on the center points of the segmented landing point areas within each preset time interval, the landing point of the target object is obtained, the traffic trajectory points of each preset time interval are fused, and then the DBSCAN algorithm is used to perform cluster analysis on the traffic trajectory points to obtain the landing point of the target object, which greatly reduces the cost of manual confirmation and improves the efficiency of obtaining the landing point of the target object.

[0032] In one embodiment of the application, the vehicle passing data, the positioning trajectory data, and the ticketing information acquired based on each preset time interval are pre-processed and sorted, including the following steps:

[0033] S201, using a preset sliding window to obtain vehicle passing data, positioning trajectory data, and ticketing information at each preset time interval to obtain the traffic volume of the target object;

[0034] Among them, the traffic travel volume includes the travel starting point, travel destination, travel mode and travel time period of the target object; the preset sliding window can be set to a 48-hour time window, through which the target object's passing vehicle data, positioning trajectory data and ticketing information are obtained every 48 hours.

[0035] Specifically, the traffic volume of the target object within a certain period of time is determined according to the vehicle passing data, the positioning trajectory data and the ticketing information, and the traffic volume within the same period of time is integrated to obtain the total traffic volume of the target object.

[0036] S202: Clean various data on traffic volume and sort them in chronological order to obtain effective traffic volume.

[0037] Specifically, the data middleware is used to clean up the empty data, disordered data, and expired data in all traffic volumes to obtain valid traffic volumes.

[0038] In the embodiment of the present application, setting a preset sliding window can prevent calculation errors caused by device data upload delays, reorder minute data, and ensure that the expiration window and the sorting window are consistent.

[0039] In one embodiment of the application, step S3 includes:

[0040] S301, using the DBSCAN algorithm to perform cluster analysis on the travel point set to obtain the first central cluster point;

[0041] The starting point and the end point of each trip are analyzed from the effective traffic volume; and a set of travel points is obtained based on the starting point and the end point.

[0042] DBSCAN searches for clusters by checking the Eps neighborhood of each point in the dataset. If the Eps neighborhood of point p contains more than MinPts points, a cluster with p as the core object is created; then, DBSCAN iteratively clusters objects that are directly density-reachable from these core objects. This process may involve the merging of some density-reachable clusters; the process ends when no new points are added to any cluster.

[0043] Specifically, the DBSCAN algorithm is used to perform cluster analysis on the travel point set, and classification clusters are generated after noise points are deleted. GIS area outlining is performed, the area center point is calculated, and the address of the area center point is reversed as the classification cluster name, which is also the name of the destination area.

[0044] S302: Based on the second central cluster point, obtain the landing point area result of the target object.

[0045] The DBSCAN algorithm is used to perform secondary cluster analysis on the regional center points of each classification cluster obtained in S301 to obtain the landing point of the target object.

[0046] In one embodiment of the application, the method of using a preset sliding window to obtain the vehicle passing data, the positioning trajectory data, and the ticketing information obtained at each preset time interval to obtain the traffic volume of the target object includes the following steps:

[0047] Using a first preset sliding window to obtain the vehicle passing data, the positioning trajectory data, and the ticketing information within a first preset time period, and obtain the traffic volume of the target object;

[0048] The step of obtaining the landing point of the target object based on the center point of the landing point area of ​​each segment within the preset time length includes the following steps:

[0049] The DBSCAN algorithm is used to perform cluster analysis on traffic trajectory points to obtain the center point of the segmented footpath area of ​​the target object within the first preset time length; based on the difference between the center point of the segmented footpath area and the actual footpath of the target vehicle, the tuning parameters of the DBSCAN algorithm are adjusted to obtain the first DBSCAN algorithm; the first DBSCAN algorithm is used to obtain the traffic volume obtained by the second preset sliding window to obtain the center point of the segmented footpath area of ​​the target object within the second preset time length;

[0050] Repeat the above steps until the center point of the target object's paragraph footer area within the Nth preset duration is obtained.

[0051] In the embodiment of the present application, the accuracy of the landing point calculation is improved by continuously tuning the clustering algorithm, and there is no need for business personnel to conduct macroscopic observation, which greatly reduces the manpower cost.

[0052] In an application embodiment, a device for analyzing a foothold based on multi-source traffic data is proposed, wherein the device comprises:

[0053] A data acquisition module, which is used to acquire in real time the passing data of the target vehicle driven by the target object, the positioning trajectory data of the target vehicle, and the ticketing information of the target object;

[0054] A trajectory point acquisition module, which is used to pre-process and sort the vehicle passing data, the positioning trajectory data and the ticketing information acquired at each preset time interval, and acquire the traffic trajectory points of the target object within each preset time interval;

[0055] A center point acquisition module, which is used to perform cluster analysis on the traffic trajectory points using the DBSCAN algorithm to obtain the center point of the target object in each of the preset time periods;

[0056] A foothold acquisition module is used to acquire the foothold of the target object based on the center point of the foothold area of ​​each segment within the preset time length.

[0057] By adopting the above-mentioned destination analysis device based on multi-source traffic data, the controller pre-processes and sorts the passing vehicle data, positioning trajectory data and ticketing information obtained at each preset time interval, and obtains the traffic trajectory points of the target object within each preset time interval; the DBSCAN algorithm is used to perform cluster analysis on the traffic trajectory points to obtain the center points of the segmented destination areas of the target object within each preset time interval; based on the center points of the segmented destination areas within each preset time interval, the destination of the target object is obtained, the traffic trajectory points of each preset time interval are fused, and then the DBSCAN algorithm is used to perform cluster analysis on the traffic trajectory points to obtain the destination of the target object, which greatly reduces the cost of manual confirmation and improves the efficiency of obtaining the destination of the target object.

[0058] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A foothold analysis method based on multi-source traffic data, characterized in that: The method comprises the following steps: Real-time acquisition of the passing data of the target vehicle driven by the target object, the positioning trajectory data of the target vehicle, and the ticketing information of the target object; Preprocess and sort the vehicle passing data, the positioning trajectory data and the ticketing information obtained at each preset time interval to obtain the traffic trajectory points of the target object within each preset time interval; Using the DBSCAN algorithm to perform cluster analysis on the traffic trajectory points, obtaining the center point of the target object's segmented footsteps area within each of the preset time lengths; Based on the center point of each segment foothold area within the preset time length, obtaining the foothold of the target object; The real-time acquisition of the passing data of the target vehicle driven by the target object, the positioning trajectory data of the target vehicle and the ticket information of the target object comprises the following steps: retrieving the passing data of the target vehicle from the database of the vehicle management office, wherein the passing data comprises the position information of the checkpoint through which the target vehicle passes and the facial capture feature data of the time when the target vehicle passes the checkpoint; matching the facial capture feature data with the facial feature data of the target object, filtering out the target object from the passing data to obtain the passing data of the target vehicle driven by the target object; The method of obtaining the target object's landing point based on the center points of the landing point areas within each preset time period includes the following steps: using the DBSCAN algorithm to perform cluster analysis on the N center points of the landing point areas within the preset time period to obtain a second center cluster point; based on the second center cluster point, obtaining the landing point area result of the target object.

2. The method for analyzing the foothold based on multi-source traffic data according to claim 1, characterized in that: The preprocessing and sorting of the vehicle passing data, the positioning trajectory data and the ticketing information acquired based on each preset time interval includes the following steps: The preset sliding window is used to obtain the vehicle passing data, the positioning trajectory data and the ticketing information obtained at each preset time interval to obtain the traffic volume of the target object; the traffic volume includes the travel starting point, travel end point, travel mode and travel time period of the target object; The various data in the traffic volume are cleaned and sorted in chronological order to obtain the effective traffic volume.

3. The method for analyzing the foothold based on multi-source traffic data according to claim 2, characterized in that: The method further comprises: Analyze the starting point and the end point of each trip from the effective traffic volume; A travel point set is obtained based on the travel starting point and the travel end point.

4. The method for analyzing the foothold based on multi-source traffic data according to claim 3 is characterized in that: The method of using the DBSCAN algorithm to perform cluster analysis on the traffic trajectory points to obtain the center point of the target object in each of the preset time periods includes the following steps: Perform cluster analysis on the travel point set using the DBSCAN algorithm to obtain a first central cluster point; Based on the first central cluster point, a central point of a landing area of ​​the target object is obtained.

5. The method for analyzing the foothold based on multi-source traffic data according to claim 3 is characterized in that: The method of using a preset sliding window to obtain the vehicle passing data, the positioning trajectory data, and the ticketing information obtained at each preset time interval to obtain the traffic volume of the target object includes the following steps: Using a first preset sliding window to obtain the vehicle passing data, the positioning trajectory data, and the ticketing information within a first preset time period, and obtain the traffic volume of the target object; The step of obtaining the landing point of the target object based on the center point of the landing point area of ​​each segment within the preset time length includes the following steps: Using the DBSCAN algorithm to perform cluster analysis on the traffic trajectory points, obtaining the center point of the segmented footstep area of ​​the target object within the first preset time length; Based on the difference between the center point of the segment foothold area and the actual foothold of the target vehicle, adjusting the tuning parameters of the DBSCAN algorithm to obtain a first DBSCAN algorithm; Using the first DBSCAN algorithm to obtain the traffic volume obtained by the second preset sliding window, obtain the center point of the segmented foothold area of ​​the target object within the second preset time length; Repeat the above steps until the center point of the paragraph footer area of ​​the target object within the Nth preset time length is obtained.

6. A landing point analysis device based on multi-source traffic data, characterized in that: The device comprises: A data acquisition module is used to acquire in real time the passing data of a target vehicle driven by a target object, the positioning trajectory data of the target vehicle, and the ticketing information of the target object; the real-time acquisition of the passing data of a target vehicle driven by a target object, the positioning trajectory data of the target vehicle, and the ticketing information of the target object comprises the following steps: retrieving the passing data of the target vehicle from a database of a vehicle administration office, the passing data comprising the position information of the checkpoint through which the target vehicle passes, and the facial capture feature data of the time when the target vehicle passes the checkpoint; Matching the facial feature data with the facial feature data of the target object, filtering out the target object from the vehicle passing data to obtain vehicle passing data of the target vehicle driven by the target object; A trajectory point acquisition module, which is used to pre-process and sort the vehicle passing data, the positioning trajectory data and the ticketing information acquired at each preset time interval, and acquire the traffic trajectory points of the target object within each preset time interval; A center point acquisition module, which is used to perform cluster analysis on the traffic trajectory points using the DBSCAN algorithm to obtain the center point of the target object in each of the preset time periods; A foothold acquisition module is used to acquire the foothold of the target object based on the center points of the foothold areas of each segment within the preset time length; the foothold of the target object based on the center points of the foothold areas within the preset time length includes the following steps: using the DBSCAN algorithm to perform cluster analysis on the center points of the foothold areas within the preset time lengths to acquire a second center cluster point; based on the second center cluster point, acquiring the foothold area result of the target object.

7. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images