Method, device, equipment and medium for determining IMSI owner

Abstract

The present invention provides a method, apparatus, device, and medium for determining an IMSI owner, comprising associating a portrait capture device with an IMSI device to generate a portrait-IMSI-device association table; storing portrait aggregate trajectory data in a database and establishing a portrait aggregate trajectory table partitioned by day, storing IMSI trajectory data in a database and establishing an IMSI trajectory table partitioned by day; generating a personnel trajectory statistics table and an IMSI trajectory statistics table for the previous day daily, and simultaneously generating an IMSI trajectory intermediate table and a portrait trajectory intermediate table for the previous day daily; performing joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table, and portrait trajectory intermediate table to generate a portrait-IMSI association table; determining IMSI owner information based on the confidence level of the portrait-IMSI association table, and inserting the IMSI owner information into the IMSI owner information table. The present invention has the advantage of improving the real-time and reliability of the association between IMSI and portrait.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and medium for determining an IMSI owner. Background Art

[0002] With the development and construction of urban informatization, the coverage rate of portrait capture equipment on the road has gradually increased, and with the continuous development of computer vision and machine learning technologies, technologies such as portrait recognition and portrait aggregation have become more and more mature.

[0003] In some cities, the number of daily profiles of people has reached over 20 million. A wide variety of analytical algorithms and applications exist for extracting valuable information from this data. Furthermore, with the rapid development of mobile communications technology, the number of mobile phone users has increased dramatically. In communications networks, each mobile device has a unique International Mobile Subscriber Identity (IMSI) that identifies and distinguishes different users. The IMSI is important information associated with mobile phone users, but in some cases, it is necessary to more accurately determine the identity of mobile phone users for security management, service personalization, and other needs. Therefore, research based on the association of IMSI and profiles has emerged. Currently, there are two main methods for associating IMSI and personal information:

[0004] The first method is to collect basic IMSI information and then, when checking the IMSI information, obtain the IMSI's trajectory data over a period of time online and perform an online collision with the portrait trajectory data. However, this online trajectory calculation method is slow and may be subject to concurrency issues caused by multiple users' simultaneous requests. The online calculation time range is usually limited to the recent period, which may ignore the information of some people whose historical trajectories highly overlap, resulting in misjudgments and missed detections, reducing the credibility of the association.

[0005] The second method uses a simple spatiotemporal relationship between IMSIs and portraits, saves the association information of each IMSI and portrait, and then performs integral calculations when a certain number of associations are accumulated. However, since a large amount of association data may contain a lot of noise, the computational complexity is huge and it is not suitable for large-scale data processing. In addition, the credibility of the association is low, and the real-time performance is poor.

[0006] As can be seen from the above, the existing IMSI and portrait association solutions need to be improved in terms of real-time performance and reliability. In view of the above problems, the inventors of this case conducted in-depth research on this issue, which led to the creation of this case. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a method, device, equipment and medium for determining the IMSI owner, which can improve the real-time and credibility of the association between the IMSI and the portrait.

[0008] In a first aspect, the present invention provides a method for determining an IMSI owner, the method comprising the following steps:

[0009] Associate the portrait capture device with the IMSI device and generate a portrait IMSI device association table;

[0010] The portrait trajectory data collected by the portrait capture device is stored in the database, and a portrait trajectory table partitioned by day is established; the IMSI trajectory data collected by the IMSI device is stored in the database, and an IMSI trajectory table partitioned by day is established;

[0011] Every day, a personnel trajectory statistics table and an IMSI trajectory statistics table for the previous day are generated. At the same time, an IMSI trajectory intermediate table and a portrait trajectory intermediate table for the previous day are generated. The IMSI trajectory intermediate table is used to count the IMSI trajectory data captured by multiple portrait capture devices on the same day, and the portrait trajectory intermediate table is used to count the portrait aggregate trajectory data associated with IMSI devices on the same day.

[0012] Perform joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table, and portrait trajectory intermediate table to generate a portrait IMSI association table, wherein the table field of the portrait IMSI association table includes confidence;

[0013] The IMSI owner information is determined based on the confidence level of the portrait IMSI association table, and the IMSI owner information is inserted into the IMSI owner information table.

[0014] Furthermore, the portrait capture device is associated with the IMSI device and the portrait IMSI device association table is generated as follows:

[0015] According to the collection range of the IMSI device, the portrait capture device within the preset distance is associated with the IMSI device, and a portrait IMSI device association table is generated. The table fields of the portrait IMSI device association table include at least the device number of the portrait capture device, the device number of the IMSI device, and the distance between the portrait capture device and the IMSI device.

[0016] Furthermore, the table attributes of the portrait archive trajectory table include at least a unique identifier of the person, a device number of the portrait capture device, a thumbnail address, a large image address, a capture time, longitude, and latitude;

[0017] The table attributes of the IMSI track table include at least a mobile user identification code, a device number of an IMSI device, a collection time, a longitude, and a latitude.

[0018] Furthermore, the table attributes of the personnel trajectory statistics table include the personnel unique identifier, the latest time the portrait was captured, the latest time the portrait of the portrait-associated IMSI trajectory was captured, the number of times the portrait of the portrait-associated IMSI trajectory was captured, and the number of IMSI devices or statistical dates on which the portrait of the portrait-associated IMSI trajectory was collected;

[0019] The table attributes of the IMSI track statistics table include mobile user identification code, latest collection time, number of collections, number of collected IMSI devices and statistical date.

[0020] Furthermore, the IMSI trajectory intermediate table is obtained by combining the IMSI trajectory table with the IMSI trajectory statistics table and filtering out the data collected by only one IMSI device on that day; the portrait trajectory intermediate table is generated by associating the portrait aggregate trajectory table with the portrait IMSI device association table through the portrait capture device.

[0021] Furthermore, the joint statistics of the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are performed to generate the portrait IMSI association table, which specifically includes:

[0022] The generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are jointly counted every day to generate a portrait IMSI association table, wherein the table fields of the portrait IMSI association table also include the personnel unique identifier, mobile user identification code, number of travels, number of days when travels are captured, number of IMSI devices traveling together, latest travel time, number of co-occurrences in different locations, number of IMSI device collections, number of IMSI devices collected, latest collection time of IMSI devices, number of days when IMSI devices are collected, number of portraits of portrait-associated IMSI trajectory captured, latest time when portraits of portraits of portrait-associated IMSI trajectory are captured, number of IMSI devices collected for portraits of portrait-associated IMSI trajectory or number of days when portraits of portraits of portrait-associated IMSI trajectory are captured;

[0023] The confidence score is calculated as follows: confidence_score = w1·txscore + w2·(1-abscore) + w3·devscore + w4·dayscore; where:

[0024] w1, w2, w3 and w4 represent weights, and the sum of w1, w2, w3 and w4 is 1;

[0025] txscore indicates the proportion of peers, and txscore = total / (pm_total + imsi_total), where total indicates the number of peers, pm_total indicates the number of times a person's portrait is captured in the portrait-associated IMSI trajectory, and imsi_total indicates the number of times an IMSI device is collected.

[0026] abscore represents the ratio of co-occurrences in different locations, and abscore = abnormal_total / total, where abnormal_total represents the number of co-occurrences in different locations, and total represents the number of co-occurrences in the same location.

[0027] devscore indicates the proportion of IMSI devices in the same group, and devscore = dev_total / max(imsi_dev_total,pm_dev_total), where dev_total indicates the number of IMSI devices in the same group, imsi_dev_total indicates the number of IMSI devices whose portraits are associated with the IMSI trajectory, and pm_dev_total indicates the number of IMSI devices whose portraits are collected.

[0028] Dayscore represents the proportion of days of traveling together, dayscore = day_total / max(imsi_day,pm_day), where day_total represents the number of days the traveling together was captured, imsi_day represents the number of days the IMSI device was collected, and pm_day represents the number of days the portrait of the person associated with the IMSI trajectory was captured;

[0029] Determine whether the calculated confidence level is greater than or equal to a first preset value; if so, retain the data in the portrait-IMSI association table; if not, delete the data in the portrait-IMSI association table and insert the data in the portrait-IMSI association table into a temporary table, which also includes the old confidence level;

[0030] At the same time, after each preset statistical period, the data within the preset statistical period is re-counted, and the portrait-IMSI association table is supplemented according to the re-counted results.

[0031] Furthermore, the method also includes: incrementally looping and temporarily storing data in a table whose old confidence is greater than or equal to a second preset value, and if the association between the personnel unique identifier and the mobile user identification code exists in the IMSI owner information table, determining whether the personnel unique identifier and the mobile user identification code are manually set; if not, deleting the corresponding IMSI owner information in the IMSI owner information table; if yes, not deleting the corresponding IMSI owner information in the IMSI owner information table.

[0032] In a second aspect, the present invention provides a device for determining an IMSI owner, the device comprising a portrait IMSI-device association module, a data storage module, a trajectory statistics module, a portrait IMSI association statistics module, and an owner information determination module;

[0033] The portrait IMSI device association module is used to associate the portrait capture device with the IMSI device and generate a portrait IMSI device association table;

[0034] The data storage module is used to store the portrait trajectory data collected by the portrait capture device into the database and establish a portrait trajectory table partitioned by day; store the IMSI trajectory data collected by the IMSI device into the database and establish an IMSI trajectory table partitioned by day;

[0035] The trajectory statistics module is used to generate a personnel trajectory statistics table and an IMSI trajectory statistics table for the previous day every day, and at the same time generate an IMSI trajectory intermediate table and a portrait trajectory intermediate table for the previous day every day, wherein the IMSI trajectory intermediate table is used to count the IMSI trajectory data captured by multiple portrait capture devices on the same day, and the portrait trajectory intermediate table is used to count the portrait aggregate trajectory data associated with the IMSI device on the same day;

[0036] The portrait IMSI association statistics module is used to perform joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table, thereby generating a portrait IMSI association table, wherein the table field of the portrait IMSI association table includes confidence;

[0037] The owner information determination module is used to determine the IMSI owner information according to the confidence level of the portrait IMSI association table, and insert the IMSI owner information into the IMSI owner information table.

[0038] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the method described in the first aspect when the program is executed by a processor.

[0040] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages: a portrait IMSI device association table is generated by associating a portrait capture device with an IMSI device; a portrait aggregate trajectory table partitioned by day is established using the portrait aggregate trajectory data collected by the portrait capture device, and an IMSI trajectory table partitioned by day is established using the IMSI trajectory data collected by the IMSI device; a personnel trajectory statistics table and an IMSI trajectory statistics table of the previous day are statistically generated every day, and an IMSI trajectory intermediate table and a portrait trajectory intermediate table of the previous day are statistically generated every day; and the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are jointly statistically generated to generate a portrait IMSI association table; a comprehensive IMSI and portrait association solution is formed from noise filtering, computational efficiency, device imbalance problem, timed association calculation, recalculation strategy, confidence calculation, and data update strategy, which can effectively improve the accuracy, real-timeness and credibility of the IMSI and portrait association.

[0041] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0043] Figure 1 This is a flowchart of a method for determining an IMSI owner in Embodiment 1 of the present invention;

[0044] Figure 2 This is a schematic diagram of the structure of a device for determining an IMSI owner in a second embodiment of the present invention;

[0045] Figure 3 This is a schematic structural diagram of an electronic device in a third embodiment of the present invention;

[0046] Figure 4 Schematic diagram of the structure of the medium in the fourth embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to better understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0048] Example 1

[0049] This embodiment provides a method for determining the IMSI owner, such as Figure 1 As shown, the method includes the following steps:

[0050] S1. Associate the portrait capture device with the IMSI device and generate a portrait IMSI device association table dim.face_imsi_device. In specific use, the portrait IMSI device association table can be used to associate the portrait archive trajectory data with the IMSI trajectory data;

[0051] S2. The portrait trajectory data collected by the portrait capture device is stored in the database, and a portrait trajectory table ods_people.face_track partitioned by day is established. By adopting the partitioning method by day, it can facilitate management and improve analysis performance. The IMSI trajectory data collected by the IMSI device is stored in the database, and an IMSI trajectory table ods_people.imsi_track partitioned by day is established. By adopting the partitioning method by day, it can facilitate management and improve analysis performance.

[0052] S3. Generate daily statistics of the personnel trajectory statistics table dwd_people.face_cnt and the IMSI trajectory statistics table dwd_people.imsi_cnt for the previous day. The personnel trajectory statistics table and the IMSI trajectory statistics table also use a daily partitioning method to save data. At the same time, generate daily statistics of the IMSI trajectory intermediate table ods_people.imsi_track_mulit and the portrait trajectory intermediate table ods_people.face_imsi_track for the previous day. The IMSI trajectory intermediate table is used to count the IMSI trajectory data captured by multiple portrait capture devices on the same day, and the portrait trajectory intermediate table is used to count the portrait aggregate trajectory data associated with the IMSI device on the same day. In the present invention, by generating the IMSI trajectory intermediate table for the previous day on a daily basis, the noise interference and large amount of calculation caused by a large number of non-moving mobile terminal devices being captured multiple times by the same IMSI device on the same day can be effectively solved. By generating the portrait trajectory intermediate table for the previous day on a daily basis, the amount of subsequent calculation can be effectively reduced.

[0053] S4. Perform joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table, and portrait trajectory intermediate table to generate a portrait-IMSI association table dwd_people.imsi_face_tx. The table field of the portrait-IMSI association table includes a confidence score.

[0054] S5. Determine the IMSI owner information based on the confidence level of the portrait-IMSI association table, and insert the IMSI owner information into the IMSI owner information table to facilitate user access and viewing.

[0055] As a preferred embodiment of the present invention, the steps of associating the portrait capture device with the IMSI device and generating the portrait IMSI device association table are as follows:

[0056] According to the collection range of the IMSI device, the portrait capture device within a preset distance is associated with the IMSI device, and a portrait IMSI device association table is generated. For example, when initializing the data, the portrait capture device within 500 meters of the IMSI device can be associated with the IMSI device. Of course, in the specific implementation of the present invention, the preset distance can be flexibly set according to the collection range of the IMSI device;

[0057] The table fields of the face-IMSI device association table dim.face_imsi_device include at least: the device number (devno) of the face capture device, the device number (imsi_dev) of the IMSI device, and the distance (distance) between the face capture device and the IMSI device. Of course, in the specific implementation of the present invention, the face-IMSI device association table can also be manually adjusted according to actual usage needs.

[0058] As a preferred embodiment of the present invention, in S2, the table attributes of the portrait aggregate track table ods_people.face_track include at least the person's unique identifier (idcard: ID number), the device number of the portrait capture device (camera_id), the small image address (crop_image), the large image address (orig_image), the capture time (time), the longitude (longitude) and the latitude (latitude);

[0059] The table attributes of the IMSI track table ods_people.imsi_track include at least the mobile subscriber identification code (imsi code), the device number (dev_no) of the IMSI device, the acquisition time (time), the longitude (longitude) and the latitude (latitude). Of course, the present invention is not limited to this, and other table attributes may be added according to actual needs during implementation.

[0060] In S2, the portrait tracking data and the IMSI tracking data may be stored in a database, such as an MPP database, for example, a Greenplum database.

[0061] As a preferred embodiment of the present invention, in S3, the table attributes of the personnel trajectory statistics table include a personnel unique identifier (idcard), the latest time when the portrait was captured (maxtime), the latest time when the portrait of the portrait-associated IMSI trajectory was captured (pm_maxtime), the number of times the portrait of the portrait-associated IMSI trajectory was captured (pm_total), the number of IMSI devices whose portraits of the portrait-associated IMSI trajectory were collected (imsi_dev_total) or the statistical date (day); wherein, the latest time when the portrait was captured refers to the latest time when a person was captured by a portrait capture device, the latest time when the portrait of the portrait-associated IMSI trajectory refers to the latest time when a person was captured by a portrait capture device associated with the IMSI trajectory data, the number of times the portrait of the portrait-associated IMSI trajectory was captured refers to the number of times a person was captured by a portrait capture device associated with the IMSI trajectory data, and the number of IMSI devices whose portraits of the portrait-associated IMSI trajectory were collected refers to the number of IMSI devices associated with the portrait capture device that collected the portrait of the portrait-associated IMSI trajectory;

[0062] The table attributes of the IMSI trajectory statistics table dwd_people.imsi_cnt include mobile user identification code (imsi code), latest collection time (maxtime), number of collections (total), number of collected IMSI devices (dev_total) and statistical date (day); among which, the latest collection time refers to the latest time when a person's mobile terminal device is collected by the IMSI device, the number of collections refers to the number of times a person's mobile terminal device is collected by the IMSI device, and the number of collected IMSI devices refers to how many IMSI devices have collected a person's mobile terminal device.

[0063] As a preferred embodiment of the present invention, the IMSI track intermediate table ods_people.imsi_track_mulit is obtained by combining the IMSI track table with the IMSI track statistics table to filter out data collected by only one IMSI device on the same day, and the table fields of the IMSI track intermediate table are the same as the table fields of the IMSI track table. Since a large number of non-moving mobile terminal devices are collected multiple times by the same IMSI device on the same day in actual use, causing noise interference and a large amount of calculation, the present invention combines the IMSI track table with the IMSI track statistics table to filter out data collected by only one IMSI device on the same day, that is, only retaining data collected by multiple IMSI devices on the same day, which can effectively reduce noise interference and reduce the amount of calculation.

[0064] The portrait trajectory intermediate table ods_people.face_imsi_track is generated by associating the portrait cluster trajectory table with the portrait IMSI device association table through the portrait capture device, and compared with the portrait cluster trajectory table, the portrait trajectory intermediate table has an additional field for the device number (imsi_dev_no) of the IMSI device; since the number of IMSI devices is relatively small, and the number of portrait capture devices is about 100 times that of IMSI devices, the portrait cluster trajectory table and the portrait IMSI device association table are associated through the portrait capture device to generate the portrait trajectory intermediate table, which can effectively reduce the amount of calculation.

[0065] As a preferred embodiment of the present invention, the joint statistics of the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are performed to generate the portrait IMSI association table, which specifically includes:

[0066] The generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are jointly counted every day to generate a portrait IMSI association table. The table fields of the portrait IMSI association table also include the personnel unique identifier (idcard), mobile user identification code (imsi), number of trips (total), number of days the trip was captured (day_total), number of IMSI devices in the trip (dev_total), the latest trip time (max_time), number of co-occurrences in different locations (abnormal_total), number of IMSI devices collected (imsi_ total), the number of IMSI devices collected (imsi_dev_total), the latest collection time of the IMSI device (imsi_maxitime), the number of days the IMSI device was collected (imsi_day), the number of times the portrait of the portrait-associated IMSI track was captured (pm_total), the latest portrait capture time of the portrait-associated IMSI track (pm_maxtime), the number of IMSI devices on which the portrait of the portrait-associated IMSI track was collected (pm_dev_total), or the number of days the portrait of the portrait-associated IMSI track was captured (pm_day).

[0067] In the specific implementation of the present invention, in the portrait IMSI association table, when the imsi_dev_no of the portrait trajectory intermediate table is the same as the dev_no of the IMSI trajectory intermediate table, and the time difference between the two trajectories (i.e., the IMSI trajectory and the portrait aggregate trajectory) is within a preset time (such as 5 minutes, which can be set according to actual needs), it represents a trip together. At the same time, the correlation between the IMSI track and the portrait cluster track in the portrait track intermediate table and the IMSI track intermediate table of the previous day is counted regularly every day. The total is the number of times the IMSI track and the portrait cluster track are in the same line; day_total is 1 by default if there is a line in the statistics of each day; abnormal_total is the number of times the two tracks (i.e., the IMSI track and the portrait cluster track) are far apart, i.e., the imsi_dev_no in the portrait track intermediate table is different from the dev_no in the IMSI track intermediate table (i.e., the portrait capture device is not within the collection range of the IMSI device), but the time difference between the two tracks is within the preset time; imsi_dev_total corresponds to the dev_total in the IMSI track statistics table, imsi_day corresponds to the total in the IMSI track statistics table, imsi_maxitime corresponds to the maxtime in the IMSI track statistics table, and there is a track imsi_day in the IMSI track statistics table on the same day. y is 1; pm_total corresponds to pm_total in the personnel trajectory statistics table, pm_dev_total corresponds to imsi_dev_total in the personnel trajectory statistics table, pm_maxtime corresponds to pm_maxtime in the personnel trajectory statistics table. If there is portrait captured data in the portrait-associated IMSI trajectory, the pm_day defaults to 1. At the same time, after daily statistical data is collected, if the person and IMSI are already in the portrait-IMSI association table, the time field is directly replaced, and the numeric field is accumulated (such as the number of days the same person was captured, the number of days the IMSI device was collected, and the number of days the portrait of the portrait-associated IMSI trajectory was captured, etc. need to be accumulated), and the confidence is recalculated. If the confidence is greater than or equal to a first preset value (such as 50%), the data in the portrait-IMSI association table is retained. If the confidence is less than the first preset value, the data in the portrait-IMSI association table is deleted, and the data in the portrait-IMSI association table is inserted into the temporary table.

[0068] The confidence score is calculated as follows: confidence_score = w1·txscore + w2·(1-abscore) + w3·devscore + w4·dayscore; where:

[0069] w1, w2, w3, and w4 represent weights, and the sum of w1, w2, w3, and w4 is 1. In the specific implementation of the present invention, the specific values ​​of w1, w2, w3, and w4 can be adjusted according to actual business needs and experience. As a specific implementation of the present invention, the values ​​of w1, w2, w3, and w4 are all 0.25;

[0070] txscore indicates the proportion of peers, and txscore = total / (pm_total + imsi_total), where total indicates the number of peers, pm_total indicates the number of times a person's portrait is captured in the portrait-associated IMSI trajectory, and imsi_total indicates the number of times an IMSI device is collected.

[0071] abscore represents the ratio of co-occurrences in different locations, and abscore = abnormal_total / total, where abnormal_total represents the number of co-occurrences in different locations, and total represents the number of co-occurrences in the same location. In the specific implementation of the present invention, when the number of co-occurrences in different locations is greater than the number of co-occurrences in the same location, the confidence level is directly reduced to a negative number.

[0072] devscore indicates the proportion of IMSI devices in the same group, and devscore = dev_total / max(imsi_dev_total,pm_dev_total), where dev_total indicates the number of IMSI devices in the same group, imsi_dev_total indicates the number of IMSI devices whose portraits are associated with the IMSI trajectory, and pm_dev_total indicates the number of IMSI devices whose portraits are collected.

[0073] Dayscore represents the proportion of days of traveling together, dayscore = day_total / max(imsi_day,pm_day), where day_total represents the number of days the traveling together was captured, imsi_day represents the number of days the IMSI device was collected, and pm_day represents the number of days the portrait of the person associated with the IMSI trajectory was captured;

[0074] It should be noted that according to the above calculation formula for confidence, the greater the proportion of co-occurrences, the higher the confidence score; the higher the proportion of co-occurrences in different locations, the lower the confidence score; the higher the proportion of IMSI devices in the same industry, the higher the confidence score; the higher the proportion of co-occurrence days, the higher the confidence score;

[0075] Determine whether the calculated confidence score is greater than or equal to a first preset value (e.g., 50%, which can be adjusted according to actual needs). If so, retain the data in the portrait IMSI association table; if not, delete the data in the portrait IMSI association table and insert the data in the portrait IMSI association table into a temporary table, which also includes the old confidence score. It should be noted that the temporary table has an additional field for the old confidence score (old_confidence_score) compared to the portrait IMSI association table, and the other fields are the same as those in the portrait IMSI association table. The old confidence score is not set. The confidence level specifically refers to the confidence level calculated before the current confidence level. For example, if the current confidence level is 40%, and the previous confidence level before the current confidence level was calculated was 91%, then the current confidence level and the previous confidence level need to be saved together in a temporary table. Of course, in a specific implementation, if the person and IMSI are not in the portrait-IMSI association table, the confidence level can be manually calculated. If the calculated confidence level is greater than or equal to a first preset value, the data is inserted into the portrait-IMSI association table. If the calculated confidence level is less than the first preset value, the data is not inserted into the portrait-IMSI association table.

[0076] At the same time, after every preset statistical period (such as 30 days, the preset statistical period can be set according to actual needs), the data within the preset statistical period is re-counted, and the portrait IMSI association table is supplemented according to the re-counted results.

[0077] The present invention generates a portrait-IMSI association table by performing joint statistics on the people trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table, and portrait trajectory intermediate table generated on a daily basis. Since incremental statistics are performed daily, the data volume is small and the statistical speed is fast, which makes it easy to view the latest situation, thereby ensuring real-time performance. However, since the statistical time width is not long enough, some special mobile terminal devices that do not move or data from multiple devices that do not travel together every day may be missed. Therefore, this solution adds a calculation node that re-counts the data within 30 days every 30 days to supplement the portrait-IMSI association table dwd_people.imsi_face_tx. The calculated statistical items are the same as those calculated on a daily basis, except that the statistical time is extended.

[0078] In some embodiments of the present invention, each IMSI in the generated portrait-IMSI association table may correspond to multiple individuals' data, and in the IMSI owner information table, only the data with the highest confidence level is generally recommended to be saved. Therefore, in S5, determining the IMSI owner information based on the confidence level of the portrait-IMSI association table specifically includes:

[0079] A determination is made as to whether the confidence level is greater than or equal to a third preset value (e.g., 90%, which can be set based on actual needs). If so, the data with the highest confidence level among the IMSI-associated personnel information is inserted into the IMSI owner information table. If not, the data is not inserted into the IMSI owner information table. Alternatively, the IMSI-associated personnel information is manually confirmed, and the confirmed data is inserted into the IMSI owner information table. The IMSI owner information table can be stored in Elasticsearch to facilitate system queries.

[0080] In some embodiments of the present invention, since the confidence level may be lower than a first preset value after the daily scheduled calculation and statistics, it can be considered that the previous recommendation was incorrect or that the IMSI owner has changed. Therefore, the method further includes: incrementally looping and temporarily storing data in the table whose old confidence level is greater than or equal to a second preset value (such as 90%, and the second preset value can be set according to actual needs); if the association between the person's unique identifier and the mobile user identification code exists in the IMSI owner information table, it is determined whether the person's unique identifier and the mobile user identification code are manually set; if not, the corresponding IMSI owner information in the IMSI owner information table is deleted; if so, the corresponding IMSI owner information in the IMSI owner information table is not deleted.

[0081] By adopting the above-mentioned technical scheme of the present invention, at least the following technical effects or advantages are achieved: a portrait capture device and an IMSI device are associated to generate a portrait IMSI device association table; a portrait aggregate trajectory table partitioned by day is established using the portrait aggregate trajectory data collected by the portrait capture device, and an IMSI trajectory table partitioned by day is established using the IMSI trajectory data collected by the IMSI device; a personnel trajectory statistics table and an IMSI trajectory statistics table of the previous day are statistically generated every day, and an IMSI trajectory intermediate table and a portrait trajectory intermediate table of the previous day are statistically generated every day; and the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are jointly statistically generated to generate a portrait IMSI association table; a comprehensive IMSI and portrait association scheme is formed from noise filtering, computational efficiency, equipment imbalance problem, timed association calculation, recalculation strategy, confidence calculation, and data update strategy, which can effectively improve the accuracy, real-timeness and credibility of the IMSI and portrait association.

[0082] Based on the same inventive concept, this application also provides a device corresponding to the method in Example 1, see Example 2 for details.

[0083] Example 2

[0084] In this embodiment, a device for determining the IMSI owner is provided. Figure 2As shown, the device includes a portrait IMSI device association module, a data storage module, a trajectory statistics module, a portrait IMSI association statistics module and an owner information determination module;

[0085] The portrait IMSI device association module is used to associate the portrait capture device with the IMSI device and generate the portrait IMSI device association table dim.face_imsi_device. When used in practice, the portrait archive trajectory data and the IMSI trajectory data can be associated through the portrait IMSI device association table;

[0086] The data storage module is used to store the portrait cluster trajectory data collected by the portrait capture device into the database, and establish the portrait cluster trajectory table ods_people.face_track partitioned by day. By adopting the partitioning method by day, it can facilitate management and improve analysis performance; store the IMSI trajectory data collected by the IMSI device into the database, and establish the IMSI trajectory table ods_people.imsi_track partitioned by day. By adopting the partitioning method by day, it can facilitate management and improve analysis performance;

[0087] The trajectory statistics module is used to generate a personnel trajectory statistics table dwd_people.face_cnt and an IMSI trajectory statistics table dwd_people.imsi_cnt for the previous day every day. The personnel trajectory statistics table and the IMSI trajectory statistics table also use a daily partitioning method to save data. At the same time, the IMSI trajectory intermediate table ods_people.imsi_track_mulit and the portrait trajectory intermediate table ods_people.face_imsi_track for the previous day are generated every day. The IMSI trajectory intermediate table is used to count the IMSI trajectory data captured by multiple portrait capture devices on the same day, and the portrait trajectory intermediate table is used to count the portrait aggregate trajectory data associated with the IMSI device on the same day. In the present invention, by generating the IMSI trajectory intermediate table for the previous day every day, the noise interference and large amount of calculation caused by a large number of non-moving mobile terminal devices being captured multiple times by the same IMSI device on the same day can be effectively solved. By generating the portrait trajectory intermediate table for the previous day every day, the subsequent calculation amount can be effectively reduced.

[0088] The portrait IMSI association statistics module is used to perform joint statistics on the generated people trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table, thereby generating a portrait IMSI association table dwd_people.imsi_face_tx, wherein the table field of the portrait IMSI association table includes a confidence score;

[0089] The owner information determination module is used to determine the IMSI owner information according to the confidence level of the portrait IMSI association table, and insert the IMSI owner information into the IMSI owner information table to facilitate user access and viewing.

[0090] As a preferred embodiment of the present invention, the steps of associating the portrait capture device with the IMSI device and generating the portrait IMSI device association table are as follows:

[0091] According to the collection range of the IMSI device, the portrait capture device within a preset distance is associated with the IMSI device, and a portrait IMSI device association table is generated. For example, when initializing the data, the portrait capture device within 500 meters of the IMSI device can be associated with the IMSI device. Of course, in the specific implementation of the present invention, the preset distance can be flexibly set according to the collection range of the IMSI device;

[0092] The table fields of the face-IMSI device association table dim.face_imsi_device include at least: the device number (devno) of the face capture device, the device number (imsi_dev) of the IMSI device, and the distance (distance) between the face capture device and the IMSI device. Of course, in the specific implementation of the present invention, the face-IMSI device association table can also be manually adjusted according to actual usage needs.

[0093] As a preferred embodiment of the present invention, in the data storage module, the table attributes of the portrait archive track table ods_people.face_track include at least the person's unique identifier (idcard: ID number), the device number of the portrait capture device (camera_id), the small image address (crop_image), the large image address (orig_image), the capture time (time), the longitude (longitude) and the latitude (latitude);

[0094] The table attributes of the IMSI track table ods_people.imsi_track include at least the mobile subscriber identification code (imsi code), the device number (dev_no) of the IMSI device, the acquisition time (time), the longitude (longitude) and the latitude (latitude). Of course, the present invention is not limited to this, and other table attributes may be added according to actual needs during implementation.

[0095] In the data storage module, the portrait cluster trajectory data and the IMSI trajectory data can be stored in the database and stored in the mpp database, such as the greenplum database.

[0096] As a preferred embodiment of the present invention, in the trajectory statistics module, the table attributes of the personnel trajectory statistics table include a personnel unique identifier (idcard), the latest portrait capture time (maxtime), the latest portrait capture time of the portrait-associated IMSI trajectory (pm_maxtime), the number of portrait capture times of the portrait-associated IMSI trajectory (pm_total), the number of IMSI devices whose portraits of the portrait-associated IMSI trajectory are collected (imsi_dev_total) or the statistical date (day); wherein, the latest portrait capture time refers to the latest time when a person is captured by a portrait capture device, the latest portrait capture time of the portrait-associated IMSI trajectory refers to the latest time when a person is captured by a portrait capture device associated with IMSI trajectory data, the number of portrait capture times of the portrait-associated IMSI trajectory refers to the number of times a person is captured by a portrait capture device associated with IMSI trajectory data, and the number of IMSI devices whose portraits of the portrait-associated IMSI trajectory are collected refers to the number of IMSI devices associated with the portrait capture device that collect the portrait of a person;

[0097] The table attributes of the IMSI trajectory statistics table dwd_people.imsi_cnt include mobile user identification code (imsi code), latest collection time (maxtime), number of collections (total), number of collected IMSI devices (dev_total) and statistical date (day); among which, the latest collection time refers to the latest time when a person's mobile terminal device is collected by the IMSI device, the number of collections refers to the number of times a person's mobile terminal device is collected by the IMSI device, and the number of collected IMSI devices refers to how many IMSI devices have collected a person's mobile terminal device.

[0098] As a preferred embodiment of the present invention, the IMSI track intermediate table ods_people.imsi_track_mulit is obtained by combining the IMSI track table with the IMSI track statistics table to filter out data collected by only one IMSI device on the same day, and the table fields of the IMSI track intermediate table are the same as the table fields of the IMSI track table. Since a large number of non-moving mobile terminal devices are collected multiple times by the same IMSI device on the same day in actual use, causing noise interference and a large amount of calculation, the present invention combines the IMSI track table with the IMSI track statistics table to filter out data collected by only one IMSI device on the same day, that is, only retaining data collected by multiple IMSI devices on the same day, which can effectively reduce noise interference and reduce the amount of calculation.

[0099] The portrait trajectory intermediate table ods_people.face_imsi_track is generated by associating the portrait cluster trajectory table with the portrait IMSI device association table through the portrait capture device, and compared with the portrait cluster trajectory table, the portrait trajectory intermediate table has an additional field for the device number (imsi_dev_no) of the IMSI device; since the number of IMSI devices is relatively small, and the number of portrait capture devices is about 100 times that of IMSI devices, the portrait cluster trajectory table and the portrait IMSI device association table are associated through the portrait capture device to generate the portrait trajectory intermediate table, which can effectively reduce the amount of calculation.

[0100] As a preferred embodiment of the present invention, the joint statistics of the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are performed to generate the portrait IMSI association table, which specifically includes:

[0101] The generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are jointly counted every day to generate a portrait IMSI association table. The table fields of the portrait IMSI association table also include the personnel unique identifier (idcard), mobile user identification code (imsi), number of trips (total), number of days the trip was captured (day_total), number of IMSI devices in the trip (dev_total), the latest trip time (max_time), number of co-occurrences in different locations (abnormal_total), number of IMSI devices collected (imsi_ total), the number of IMSI devices collected (imsi_dev_total), the latest collection time of the IMSI device (imsi_maxitime), the number of days the IMSI device was collected (imsi_day), the number of times the portrait of the portrait-associated IMSI track was captured (pm_total), the latest portrait capture time of the portrait-associated IMSI track (pm_maxtime), the number of IMSI devices on which the portrait of the portrait-associated IMSI track was collected (pm_dev_total), or the number of days the portrait of the portrait-associated IMSI track was captured (pm_day).

[0102] In the specific implementation of the present invention, in the portrait IMSI association table, when the imsi_dev_no of the portrait trajectory intermediate table is the same as the dev_no of the IMSI trajectory intermediate table, and the time difference between the two trajectories (i.e., the IMSI trajectory and the portrait aggregate trajectory) is within a preset time (such as 5 minutes, which can be set according to actual needs), it represents a trip together. At the same time, the correlation between the IMSI track and the portrait cluster track in the portrait track intermediate table and the IMSI track intermediate table of the previous day is counted regularly every day. The total is the number of times the IMSI track and the portrait cluster track are in the same line; day_total is 1 by default if there is a line in the statistics of each day; abnormal_total is the number of times the two tracks (i.e., the IMSI track and the portrait cluster track) are far apart, i.e., the imsi_dev_no in the portrait track intermediate table is different from the dev_no in the IMSI track intermediate table (i.e., the portrait capture device is not within the collection range of the IMSI device), but the time difference between the two tracks is within the preset time; imsi_dev_total corresponds to the dev_total in the IMSI track statistics table, imsi_day corresponds to the total in the IMSI track statistics table, imsi_maxitime corresponds to the maxtime in the IMSI track statistics table, and there is a track imsi_day in the IMSI track statistics table on the same day. y is 1; pm_total corresponds to pm_total in the personnel trajectory statistics table, pm_dev_total corresponds to imsi_dev_total in the personnel trajectory statistics table, pm_maxtime corresponds to pm_maxtime in the personnel trajectory statistics table. If there is portrait captured data in the portrait-associated IMSI trajectory, the pm_day defaults to 1. At the same time, after daily statistical data is collected, if the person and IMSI are already in the portrait-IMSI association table, the time field is directly replaced, and the numeric field is accumulated (such as the number of days the same person was captured, the number of days the IMSI device was collected, and the number of days the portrait of the portrait-associated IMSI trajectory was captured, etc. need to be accumulated), and the confidence is recalculated. If the confidence is greater than or equal to a first preset value (such as 50%), the data in the portrait-IMSI association table is retained. If the confidence is less than the first preset value, the data in the portrait-IMSI association table is deleted, and the data in the portrait-IMSI association table is inserted into the temporary table.

[0103] The confidence score is calculated as follows: confidence_score = w1·txscore + w2·(1-abscore) + w3·devscore + w4·dayscore; where:

[0104] w1, w2, w3, and w4 represent weights, and the sum of w1, w2, w3, and w4 is 1. In the specific implementation of the present invention, the specific values ​​of w1, w2, w3, and w4 can be adjusted according to actual business needs and experience. As a specific implementation of the present invention, the values ​​of w1, w2, w3, and w4 are all 0.25;

[0105] txscore indicates the proportion of peers, and txscore = total / (pm_total + imsi_total), where total indicates the number of peers, pm_total indicates the number of times a person's portrait is captured in the portrait-associated IMSI trajectory, and imsi_total indicates the number of times an IMSI device is collected.

[0106] abscore represents the ratio of co-occurrences in different locations, and abscore = abnormal_total / total, where abnormal_total represents the number of co-occurrences in different locations, and total represents the number of co-occurrences in the same location. In the specific implementation of the present invention, when the number of co-occurrences in different locations is greater than the number of co-occurrences in the same location, the confidence level is directly reduced to a negative number.

[0107] devscore indicates the proportion of IMSI devices in the same group, and devscore = dev_total / max(imsi_dev_total,pm_dev_total), where dev_total indicates the number of IMSI devices in the same group, imsi_dev_total indicates the number of IMSI devices whose portraits are associated with the IMSI trajectory, and pm_dev_total indicates the number of IMSI devices whose portraits are collected.

[0108] Dayscore represents the proportion of days of traveling together, dayscore = day_total / max(imsi_day,pm_day), where day_total represents the number of days the traveling together was captured, imsi_day represents the number of days the IMSI device was collected, and pm_day represents the number of days the portrait of the person associated with the IMSI trajectory was captured;

[0109] It should be noted that according to the above calculation formula for confidence, the greater the proportion of co-occurrences, the higher the confidence score; the higher the proportion of co-occurrences in different locations, the lower the confidence score; the higher the proportion of IMSI devices in the same industry, the higher the confidence score; the higher the proportion of co-occurrence days, the higher the confidence score;

[0110] Determine whether the calculated confidence score is greater than or equal to a first preset value (e.g., 50%, which can be adjusted according to actual needs). If so, retain the data in the portrait IMSI association table; if not, delete the data in the portrait IMSI association table and insert the data in the portrait IMSI association table into a temporary table, which also includes the old confidence score. It should be noted that the temporary table has an additional field for the old confidence score (old_confidence_score) compared to the portrait IMSI association table, and the other fields are the same as those in the portrait IMSI association table. The old confidence score is not set. The confidence level specifically refers to the confidence level calculated before the current confidence level. For example, if the current confidence level is 40%, and the previous confidence level before the current confidence level was calculated was 91%, then the current confidence level and the previous confidence level need to be saved together in a temporary table. Of course, in a specific implementation, if the person and IMSI are not in the portrait-IMSI association table, the confidence level can be manually calculated. If the calculated confidence level is greater than or equal to a first preset value, the data is inserted into the portrait-IMSI association table. If the calculated confidence level is less than the first preset value, the data is not inserted into the portrait-IMSI association table.

[0111] At the same time, after every preset statistical period (such as 30 days, the preset statistical period can be set according to actual needs), the data within the preset statistical period is re-counted, and the portrait IMSI association table is supplemented according to the re-counted results.

[0112] The present invention generates a portrait-IMSI association table by performing joint statistics on the personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table, and portrait trajectory intermediate table generated on a daily basis. Since incremental statistics are performed daily, the data volume is small and the statistical speed is fast, which makes it easy to view the latest situation, thereby ensuring real-time performance. However, since the statistical time width is not long enough, some special mobile terminal devices that have not moved or data of multiple devices not traveling together every day may be missed. Therefore, this solution adds a calculation node, which is to re-count the data within 30 days every 30 days to supplement the portrait-IMSI association table dwd_people.imsi_face_tx. The calculated statistical items are the same as those calculated on a daily basis, except that the statistical time is extended.

[0113] In some embodiments of the present invention, each IMSI in the generated person-portrait / IMSI association table may correspond to multiple person data, and in the IMSI owner information table, only the data with the highest confidence level is generally recommended to be saved; therefore, in the owner information determination module, determining the IMSI owner information based on the confidence level of the person-portrait / IMSI association table specifically includes:

[0114] A determination is made as to whether the confidence level is greater than or equal to a third preset value (e.g., 90%, which can be set based on actual needs). If so, the data with the highest confidence level among the IMSI-associated personnel information is inserted into the IMSI owner information table. If not, the data is not inserted into the IMSI owner information table. Alternatively, the IMSI-associated personnel information is manually confirmed, and the confirmed data is inserted into the IMSI owner information table. The IMSI owner information table can be stored in Elasticsearch to facilitate system queries.

[0115] In some embodiments of the present invention, since the confidence level may be lower than a first preset value after the daily scheduled calculation and statistics, it can be considered that the previous recommendation was incorrect or that the IMSI owner has changed. Therefore, the method further includes: incrementally looping and temporarily storing data in the table whose old confidence level is greater than or equal to a second preset value (such as 90%, and the second preset value can be set according to actual needs); if the association between the person's unique identifier and the mobile user identification code exists in the IMSI owner information table, it is determined whether the person's unique identifier and the mobile user identification code are manually set; if not, the corresponding IMSI owner information in the IMSI owner information table is deleted; if so, the corresponding IMSI owner information in the IMSI owner information table is not deleted.

[0116] By adopting the above-mentioned technical scheme of the present invention, at least the following technical effects or advantages are achieved: a portrait capture device and an IMSI device are associated to generate a portrait IMSI device association table; a portrait aggregate trajectory table partitioned by day is established using the portrait aggregate trajectory data collected by the portrait capture device, and an IMSI trajectory table partitioned by day is established using the IMSI trajectory data collected by the IMSI device; a personnel trajectory statistics table and an IMSI trajectory statistics table of the previous day are statistically generated every day, and an IMSI trajectory intermediate table and a portrait trajectory intermediate table of the previous day are statistically generated every day; and the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table are jointly statistically generated to generate a portrait IMSI association table; a comprehensive IMSI and portrait association scheme is formed from noise filtering, computational efficiency, equipment imbalance problem, timed association calculation, recalculation strategy, confidence calculation, and data update strategy, which can effectively improve the accuracy, real-timeness and credibility of the IMSI and portrait association.

[0117] Based on the same inventive concept, this application provides an electronic device embodiment corresponding to the first embodiment, see the third embodiment for details.

[0118] Example 3

[0119] This embodiment provides an electronic device, such as Figure 3As shown, it includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any implementation method in the first embodiment can be implemented.

[0120] Since the electronic device described in this embodiment is the device used to implement the method in Example 1 of this application, based on the method described in Example 1 of this application, those skilled in the art will be able to understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of this application will not be described in detail here. As long as the device used by those skilled in the art to implement the method in the embodiment of this application falls within the scope of protection to be provided by this application.

[0121] Based on the same inventive concept, this application provides a storage medium corresponding to Example 1, see Example 4 for details.

[0122] Example 4

[0123] This embodiment provides a computer-readable storage medium, such as Figure 4 As shown, a computer program is stored thereon, and when the computer program is executed by a processor, any implementation method in Example 1 can be implemented.

[0124] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatuses, and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0126] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0128] Although the specific embodiments of the present invention are described above, those skilled in the art should understand that the specific embodiments described are merely illustrative and are not intended to limit the scope of the present invention. Equivalent modifications and changes made by those skilled in the art in accordance with the spirit of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for determining an IMSI owner, characterized by: The method comprises the following steps: Associate the portrait capture device with the IMSI device and generate a portrait IMSI device association table; The portrait cluster trajectory data collected by the portrait capture device is stored in the database, and a portrait cluster trajectory table partitioned by day is established; The IMSI track data collected by the IMSI device is stored in the database, and an IMSI track table partitioned by day is established; Every day, a personnel trajectory statistics table and an IMSI trajectory statistics table for the previous day are generated. At the same time, an IMSI trajectory intermediate table and a portrait trajectory intermediate table for the previous day are generated. The IMSI trajectory intermediate table is used to count the IMSI trajectory data captured by multiple portrait capture devices on the same day, and the portrait trajectory intermediate table is used to count the portrait aggregate trajectory data associated with IMSI devices on the same day. Performing joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table to generate a portrait IMSI association table, wherein the table field of the portrait IMSI association table includes a confidence level, specifically comprising: performing joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table on a daily basis to generate a portrait IMSI association table, wherein the table fields of the portrait IMSI association table also include a personnel unique identifier, a mobile user identification code, the number of travel trips, the number of days travel trips were captured, the number of IMSI devices traveled together, the latest travel time, the number of different-location co-occurrences, the number of IMSI device collections, the number of IMSI devices collected, the latest collection time of the IMSI device, the number of days the IMSI device was collected, the number of times the portrait of the portrait-associated IMSI trajectory was captured, the latest capture time of the portrait of the portrait-associated IMSI trajectory, the number of IMSI devices collected for the portrait of the portrait-associated IMSI trajectory, or the number of days the portrait of the portrait-associated IMSI trajectory was captured; The confidence score is calculated as follows: confidence_score = w1⋅txscore + w2⋅(1-abscore) + w3⋅devscore + w4⋅dayscore; where: w1, w2, w3 and w4 represent weights, and the sum of w1, w2, w3 and w4 is 1; txscore indicates the proportion of peers, and txscore = total / (pm_total + imsi_total), where total indicates the number of peers, pm_total indicates the number of times a person's portrait is captured in the portrait-associated IMSI trajectory, and imsi_total indicates the number of times an IMSI device is collected. abscore represents the ratio of co-occurrences in different locations, and abscore = abnormal_total / total, where abnormal_total represents the number of co-occurrences in different locations, and total represents the number of co-occurrences in the same location. devscore indicates the proportion of IMSI devices in the same group, and devscore = dev_total / max(imsi_dev_total,pm_dev_total). dev_total indicates the number of IMSI devices in the same group, imsi_dev_total indicates the number of IMSI devices whose portraits are associated with the IMSI trajectory, and pm_dev_total indicates the number of IMSI devices whose portraits are collected. Dayscore indicates the proportion of days spent together. Dayscore = day_total / max(imsi_day,pm_day). Day_total indicates the number of days spent together, imsi_day indicates the number of days the IMSI device was collected, and pm_day indicates the number of days the portrait associated with the IMSI trajectory was captured. Determine whether the calculated confidence level is greater than or equal to a first preset value; if so, retain the data in the portrait-IMSI association table; if not, delete the data in the portrait-IMSI association table and insert the data in the portrait-IMSI association table into a temporary table, which also includes the old confidence level; At the same time, after each preset statistical period, the data within the preset statistical period is re-counted, and the portrait IMSI association table is supplemented according to the re-counted results; The IMSI owner information is determined based on the confidence level of the portrait IMSI association table, and the IMSI owner information is inserted into the IMSI owner information table.

2. The method for determining the IMSI owner according to claim 1, wherein: The steps of associating the portrait capture device with the IMSI device and generating the portrait IMSI device association table are as follows: According to the collection range of the IMSI device, the portrait capture device within the preset distance is associated with the IMSI device, and a portrait IMSI device association table is generated. The table fields of the portrait IMSI device association table include at least the device number of the portrait capture device, the device number of the IMSI device, and the distance between the portrait capture device and the IMSI device.

3. The method for determining the IMSI owner according to claim 1, wherein: The table attributes of the portrait archive trajectory table include at least a unique identifier of the person, a device number of the portrait capture device, a thumbnail address, a large image address, a capture time, a longitude, and a latitude; The table attributes of the IMSI track table include at least a mobile user identification code, a device number of an IMSI device, a collection time, a longitude, and a latitude.

4. The method for determining the IMSI owner according to claim 1, wherein: The table attributes of the personnel trajectory statistics table include the personnel unique identifier, the latest time the portrait was captured, the latest time the portrait of the portrait-associated IMSI trajectory was captured, the number of times the portrait of the portrait-associated IMSI trajectory was captured, and the number of IMSI devices or statistical dates on which the portrait of the portrait-associated IMSI trajectory was collected; The table attributes of the IMSI track statistics table include mobile user identification code, latest collection time, number of collections, number of collected IMSI devices and statistical date.

5. The method for determining the IMSI owner according to claim 1, wherein: The IMSI trajectory intermediate table is obtained by combining the IMSI trajectory table with the IMSI trajectory statistics table and filtering out the data collected by only one IMSI device on that day; the portrait trajectory intermediate table is generated by associating the portrait aggregate trajectory table with the portrait IMSI device association table through the portrait capture device.

6. The method for determining the IMSI owner according to claim 1, wherein: The method further includes: incrementally looping through data in a temporary table with an old confidence level greater than or equal to a second preset value; and if the association between the person's unique identifier and the mobile user identification code exists in the IMSI owner information table, determining whether the person's unique identifier and the mobile user identification code are manually set; if not, deleting the corresponding IMSI owner information in the IMSI owner information table; and if so, not deleting the corresponding IMSI owner information in the IMSI owner information table.

7. A device for determining an IMSI owner, characterized by: The device includes a portrait IMSI device association module, a data storage module, a trajectory statistics module, a portrait IMSI association statistics module and an owner information determination module; The portrait IMSI device association module is used to associate the portrait capture device with the IMSI device and generate a portrait IMSI device association table; The data storage module is used to store the portrait archive trajectory data collected by the portrait capture device into the database and establish a portrait archive trajectory table partitioned by day; The IMSI track data collected by the IMSI device is stored in the database, and an IMSI track table partitioned by day is established; The trajectory statistics module is used to generate a personnel trajectory statistics table and an IMSI trajectory statistics table for the previous day every day, and at the same time generate an IMSI trajectory intermediate table and a portrait trajectory intermediate table for the previous day every day, wherein the IMSI trajectory intermediate table is used to count the IMSI trajectory data captured by multiple portrait capture devices on the same day, and the portrait trajectory intermediate table is used to count the portrait aggregate trajectory data associated with the IMSI device on the same day; The portrait IMSI association statistics module is used to perform joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table, so as to generate a portrait IMSI association table, wherein the table field of the portrait IMSI association table includes a confidence level, specifically comprising: performing joint statistics on the generated personnel trajectory statistics table, IMSI trajectory statistics table, IMSI trajectory intermediate table and portrait trajectory intermediate table on a regular basis every day to generate a portrait IMSI association table, wherein the table fields of the portrait IMSI association table also include a unique identification of the person, a mobile user identification code, the number of travel trips, the number of days the travel trips were captured, the number of IMSI devices of the travel trips, the latest travel time, the number of co-occurrences in different locations, the number of IMSI devices collected, the number of IMSI devices collected, the latest time the IMSI devices were collected, the number of days the IMSI devices were collected, the number of times the portrait of the portrait-associated IMSI trajectory was captured, the latest time the portrait of the portrait-associated IMSI trajectory was captured, the number of IMSI devices collected for the portrait of the portrait-associated IMSI trajectory or the number of days the portrait of the portrait-associated IMSI trajectory was captured; The confidence score is calculated as follows: confidence_score = w1⋅txscore + w2⋅(1-abscore) + w3⋅devscore + w4⋅dayscore; where: w1, w2, w3 and w4 represent weights, and the sum of w1, w2, w3 and w4 is 1; txscore indicates the proportion of peers, and txscore = total / (pm_total + imsi_total), where total indicates the number of peers, pm_total indicates the number of times a person's portrait is captured in the portrait-associated IMSI trajectory, and imsi_total indicates the number of times an IMSI device is collected. abscore represents the ratio of co-occurrences in different locations, and abscore = abnormal_total / total, where abnormal_total represents the number of co-occurrences in different locations, and total represents the number of co-occurrences in the same location. devscore indicates the proportion of IMSI devices in the same group, and devscore = dev_total / max(imsi_dev_total,pm_dev_total). dev_total indicates the number of IMSI devices in the same group, imsi_dev_total indicates the number of IMSI devices whose portraits are associated with the IMSI trajectory, and pm_dev_total indicates the number of IMSI devices whose portraits are collected. Dayscore indicates the proportion of days spent together. Dayscore = day_total / max(imsi_day,pm_day). Day_total indicates the number of days spent together, imsi_day indicates the number of days the IMSI device was collected, and pm_day indicates the number of days the portrait associated with the IMSI trajectory was captured. Determine whether the calculated confidence level is greater than or equal to a first preset value; if so, retain the data in the portrait-IMSI association table; if not, delete the data in the portrait-IMSI association table and insert the data in the portrait-IMSI association table into a temporary table, which also includes the old confidence level; At the same time, after each preset statistical period, the data within the preset statistical period is re-counted, and the portrait IMSI association table is supplemented according to the re-counted results; The owner information determination module is used to determine the IMSI owner information according to the confidence level of the portrait IMSI association table, and insert the IMSI owner information into the IMSI owner information table.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images