Data mining methods, apparatus, electronic devices and computer-readable storage media

By capturing pictures from the personnel mobile data flow and generating frequent pattern trees, the problem of inefficient peers in the existing technology is solved, and effective peer identification and shopping mall customer storage conversion are achieved.

CN114297254BActive Publication Date: 2024-11-15SHENZHEN XUMI YUNTU SPACE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111499385.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-09
Publication Date
2024-11-15
Estimated Expiration
2041-12-09

AI Technical Summary

Technical Problem

The method of exploring peers in the existing technology is relatively inefficient and it is difficult to effectively capture peers, resulting in low efficiency in shopping mall customer storage conversion.

Method used

By capturing multiple pictures from the personnel mobile data stream, obtaining facial information, generating personnel links, and mining target data based on frequent pattern trees, improving peer recognition efficiency.

Benefits of technology

It improves the efficiency of exploring peers, effectively captures peers and converts them into shopping mall customer storage, and solves the problem of low efficiency in the existing technology.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114297254B_ABST
    Figure CN114297254B_ABST
Patent Text Reader

Abstract

The present disclosure relates to the technical field of data mining, and provides a data mining method, device, electronic device, and computer-readable storage medium. The method comprises: capturing multiple pictures from a personnel mobile data stream, and obtaining the facial information of the personnel included in each of the multiple pictures, to obtain a facial information set, wherein the facial information comprises a facial ID, an ID of a data acquisition device for acquiring the picture, and a capture time point of the picture; generating a personnel link according to the facial information set, wherein the personnel link is a sequence string arranged in sequence according to the facial information; generating a frequent pattern tree according to the personnel link; and mining target data through the frequent pattern tree. The present disclosure solves the problem of low efficiency in mining companions, and achieves the technical effect of improving the efficiency of mining companions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the technical field of data mining, and in particular to a data mining method, device, electronic device, and computer-readable storage medium. Background Art

[0002] In real life, people often go shopping together, but generally only a few people will consume. The companions of individual consumers are often potential valuable users. Since the user trajectory path is relatively complex and the flow is relatively frequent, it is difficult to effectively capture companions, which is a loss for effectively converting the mall's customer reserve. In order to mine the companions of the consumer group, a method for mining companions is proposed.

[0003] At present, there are some methods that can be used to search for faces by face recognition and calculating the degree of time overlap, which has achieved the goal of mining people who are traveling together. However, this method often requires a large amount of calculation, and each person needs to search for people who are traveling together in the entire feature space individually, which is inefficient; in addition, this method often misses a lot of information when querying people who are traveling together by time interval, making it difficult to effectively obtain people who are traveling together.

[0004] It can be seen that no effective solution has been proposed for the problem of low efficiency in mining peers in related technologies. Summary of the invention

[0005] In view of this, the embodiments of the present disclosure provide a data mining method, device, electronic device and computer-readable storage medium to solve the problem of low efficiency in mining peers in the prior art.

[0006] According to a first aspect of an embodiment of the present disclosure, a data mining method is provided, comprising: capturing multiple pictures from a personnel movement data stream, and obtaining facial information of the personnel included in each of the multiple pictures to obtain a facial information set, wherein the facial information includes a facial ID, an ID of a data acquisition device that captures the picture, and a capture time point of the picture; generating a personnel link based on the facial information set, wherein the personnel link is a sequence string arranged in sequence according to the facial information; generating a frequent pattern tree based on the personnel link; and mining target data through the frequent pattern tree.

[0007] According to a second aspect of an embodiment of the present disclosure, a data mining device is provided, comprising: an acquisition module, for capturing multiple pictures from a personnel movement data stream, and acquiring facial information of the persons included in each of the multiple pictures, to obtain a facial information set, wherein the facial information includes a facial ID, an ID of a data acquisition device for capturing the picture, and a capture time point of the picture; a first generation module, for generating a personnel link based on the facial information set, wherein the personnel link is a sequence string arranged in sequence according to the facial information; a second generation module, for generating a frequent pattern tree based on the personnel link; and a mining module, for mining target data through the frequent pattern tree.

[0008] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0009] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0010] Compared with the prior art, the embodiments of the present disclosure have the following beneficial effects: the present disclosure obtains a set of face information by capturing multiple pictures from a personnel mobile data stream and obtaining the face information of the personnel included in each of the multiple pictures, wherein the face information includes the face ID, the data acquisition device ID for capturing the picture, and the capture time of the picture; generates a personnel link based on the face information set, wherein the personnel link is a sequence string arranged in order according to the face information; generates a frequent pattern tree based on the personnel link; and mines the target data through the frequent pattern tree. That is to say, the embodiments of the present disclosure introduce the traditional frequent pattern mining frequent pattern tree method into the peer mining scenario, rather than simply calculating the time overlap degree to search for faces or querying peers by time intervals, thereby solving the problem of low efficiency in mining peers in the related technology, and achieving the technical effect of improving the efficiency of mining peers. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, 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 disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 is a schematic diagram of an application scenario of an embodiment of the present disclosure;

[0013] Figure 2 It is a flowchart of a data mining method provided by an embodiment of the present disclosure;

[0014] Figure 3 is a schematic diagram of a data mining method provided by an embodiment of the present disclosure;

[0015] Figure 4 is a structural schematic diagram of a data mining device provided by an embodiment of the present disclosure;

[0016] Figure 5 The structure of a data mining device provided by the embodiment of the present disclosure is shown in FIG. Figure 1 ;

[0017] Figure 6 The structure of a data mining device provided by the embodiment of the present disclosure is shown in FIG. Figure 2 ;

[0018] Figure 7 The structure of a data mining device provided by the embodiment of the present disclosure is shown in FIG. Figure 3 ;

[0019] Figure 8 The structure of a data mining device provided by the embodiment of the present disclosure is shown in FIG. Figure 4 ;

[0020] Fig. 9 The structure of a data mining device provided by the embodiment of the present disclosure is shown in FIG. Figure 5 ;

[0021] Fig.10 It is a structural schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0023] A data mining method and device according to an embodiment of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0024] Figure 1 Schematic diagram of an application scenario of an embodiment of the present disclosure. The application scenario may include terminal devices 1, 2 and 3, a server 4 and a network 5.

[0025] Terminal devices 1, 2 and 3 can be hardware or software. When terminal devices 1, 2 and 3 are hardware, they can be various electronic devices with display screens and supporting communication with server 4, including but not limited to smart phones, tablet computers, laptop portable computers and desktop computers, etc.; when terminal devices 1, 2 and 3 are software, they can be installed in the electronic devices described above. Terminal devices 1, 2 and 3 can be implemented as multiple software or software modules, or as a single software or software module, and the embodiments of the present disclosure are not limited to this. Furthermore, various applications can be installed on terminal devices 1, 2 and 3, such as data processing applications, instant messaging tools, social platform software, search applications, shopping applications, etc.

[0026] The server 4 may be a server that provides various services, for example, a background server that receives a request sent by a terminal device that establishes a communication connection with the server, and the background server may receive and analyze the request sent by the terminal device, and generate a processing result. The server 4 may be a single server, or a server cluster composed of several servers, or a cloud computing service center, which is not limited in the embodiments of the present disclosure.

[0027] It should be noted that the server 4 can be hardware or software. When the server 4 is hardware, it can be various electronic devices that provide various services for the terminal devices 1, 2, and 3. When the server 4 is software, it can be multiple software or software modules that provide various services for the terminal devices 1, 2, and 3, or it can be a single software or software module that provides various services for the terminal devices 1, 2, and 3, and the embodiments of the present disclosure are not limited to this.

[0028] The network 5 can be a wired network connected by coaxial cable, twisted pair and optical fiber, or it can be a wireless network that can interconnect various communication devices without wiring, such as Bluetooth, Near Field Communication (NFC), infrared, etc., which is not limited in the embodiments of the present disclosure.

[0029] The user can establish a communication connection with the server 4 via the network 5 through the terminal devices 1, 2 and 3 to receive or send information, etc. It should be noted that the specific types, quantities and combinations of the terminal devices 1, 2 and 3, the server 4 and the network 5 can be adjusted according to the actual needs of the application scenario, and the embodiments of the present disclosure are not limited to this.

[0030] Figure 2 It is a flowchart of a data mining method provided by an embodiment of the present disclosure. Figure 2 The data mining method can be Figure 1 The terminal device or server executes. Figure 2 As shown, the data mining method includes:

[0031] S201, capturing multiple pictures from a personnel movement data stream, and obtaining facial information of a person included in each of the multiple pictures to obtain a facial information set, wherein the facial information includes a facial ID, an ID of a data acquisition device that captures the picture, and a capture time point of the picture;

[0032] It should be noted that the application scenarios of the embodiments of the present disclosure include but are not limited to the following application scenarios:

[0033] Application scenario one: mining companions of a shopping mall consumer group. For example, the data mining method provided by the embodiment of the present disclosure can be used to capture multiple pictures from a consumer group, and obtain facial information of the consumer group included in each of the multiple pictures to obtain a facial information set; based on the facial information set, a personnel link is generated; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, target data is mined. In other words, introducing the traditional frequent pattern mining method of the frequent pattern tree into the mining scenario of companions of a consumer group can effectively capture companions and effectively convert the companions into shopping mall customers;

[0034] Application scenario two: mining the number of users received by service personnel. For example, the data mining method provided by the embodiment of the present disclosure can be used to capture multiple pictures from all the reception user data streams of the service personnel, and obtain the facial information of the received users included in each of the multiple pictures to obtain a facial information set; based on the facial information set, a personnel link is generated; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, the target data is mined. In other words, the traditional frequent pattern mining method of the frequent pattern tree is introduced into the mining scenario of the reception personnel, so that the number of users received by the service personnel, the frequency of receiving users, etc. can be mined.

[0035] Application scenario three: in security scenarios or airport scenarios, identify whether different pedestrians are in a walking relationship. For example, in a security scenario, the data mining method provided by the embodiment of the present disclosure is used to capture multiple pictures from the pedestrian data stream, and obtain the facial information of the pedestrians included in each of the multiple pictures to obtain a facial information set; based on the facial information set, a personnel link is generated; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, the target data is mined. In other words, introducing the traditional frequent pattern mining method of the frequent pattern tree into the mining scenario of dangerous people can clarify the social relationship between pedestrian targets, discover potential dangers and take timely precautions; on the other hand, when a danger occurs, pedestrian information, as well as information such as the time and place of the danger, can be obtained to provide important clues for investigation.

[0036] In addition, it should be noted that the above-mentioned personnel movement data stream may be the personnel data stream of a certain shopping mall channel on a certain day or during a certain morning or evening peak period, and the personnel movement data stream includes people traveling together and people traveling alone. The disclosed embodiment aims to mine people traveling together from the above-mentioned personnel movement data stream as target data.

[0037] The above data collection device can be a camera, a camera, a mobile phone, etc. For example, taking a camera as an example, the layout requirements of the camera include but are not limited to: the camera is arranged at a height of 2-3 meters, with a depression angle of 15-20 degrees, and is generally deployed at important passages and gates, facing the face of people. For another example, taking a mobile phone as an example, the mobile phone needs to be able to clearly capture all faces in the entire picture.

[0038] Optionally, the data acquisition device captures multiple pictures from the personnel movement data stream in a manner including but not limited to the following manners, which are not limited in any way herein:

[0039] Method 1: capture multiple pictures from the above-mentioned personnel movement data stream every 1 minute, and select the picture with the highest definition from the multiple pictures;

[0040] Method 2: When the flow of mobile personnel is large, multiple pictures are captured from the above-mentioned mobile personnel data stream every 1 minute. When the flow of mobile personnel is small, multiple pictures are captured from the above-mentioned mobile personnel data stream every half minute, and then the picture with the highest definition is selected from the pictures.

[0041] Optionally, the method of obtaining the above-mentioned face ID includes but is not limited to: first detecting and identifying the face in the image, and then numbering the face, for example, personid1, personid2, personid3, etc.

[0042] The above-mentioned face recognition method includes but is not limited to the following methods, which are not limited in any way:

[0043] Method 1: extract two-dimensional face key point information and three-dimensional face key point information from the image frame in the face video, calculate the face posture angle based on the two-dimensional face key point information and the three-dimensional face key point information, and judge whether the face posture angle meets the preset conditions. When the face posture angle meets the preset conditions, extract the face image area in the image frame, and judge whether the face image in the face image area is clear based on a preset clear face classifier. When the face image in the face image area is clear, output the face image area.

[0044] Method 2: First, by searching the Haar features of the upper body of the human body, determine that the moving object is a human, and then screen and cut out the color image of the head larger than 39*39 pixels; then, through the local weight sharing convolution formula and the local weight sharing sampling formula, determine the five feature areas of the face, the left eye, the right eye, the tip of the nose, the left corner of the mouth and the right corner of the mouth, and find the center point in the feature area; extract the deep relationship value of the feature area, and convert all the relationship values ​​into a matrix; finally, compare the values ​​in the matrix with the values ​​in the database, and perform probability analysis through the Gaussian model to obtain the forward or reverse value to determine the recognition of the face.

[0045] S202, generating a personnel link according to the face information set, wherein the personnel link is a sequence string arranged in sequence according to the face information;

[0046] For example, based on the time period, the people who appear in each camera can be arranged in order to form a personnel link<personid1,personid2、personid3...> .

[0047] S203, generating a frequent pattern tree according to the personnel link;

[0048] Optionally, in the disclosed embodiment, before generating a frequent pattern tree based on the personnel link, it also includes: deleting facial information with a support value less than a minimum support value, wherein the support value is the number of times the person appears in a time relationship when being continuously captured by the same data acquisition device; or, merging repeated facial information. Because sometimes the image acquisition interval is short, a large amount of duplicate data is generated, such as when a person is continuously photographed with the same camera, directly using this part of the data will seriously affect the program processing time. Therefore, the adjacent and continuous data of the same person under the same camera are merged into one, and the time is recorded as the time of the first appearance during this period under this camera, thereby improving the mining efficiency of the target data.

[0049] Optionally, the generation of the frequent pattern tree according to the personnel link includes but is not limited to the following methods:

[0050] S301, traverse the personnel link, obtain the frequency of each face ID, and obtain a frequency table;

[0051] S302, sorting each face ID in the personnel link according to the frequency of occurrence according to the frequency table, and deleting the face ID that does not appear in the frequency table to obtain a filter table;

[0052] S303, reading the face IDs in the filter table one by one to generate the frequent pattern tree.

[0053] For example, in the embodiment of the present disclosure, the personnel link is<personid1、personid2、personid3、personid4> According to statistics, personid1 appears 5 times, personid2 appears 3 times, personid3 appears 7 times, and personid4 appears 1 time. Sort personid1, personid2, personid3, and personid4 by their frequencies to get a frequency table. Then re-sort them by their frequencies, and delete the personnel information with a frequency less than 2 to get a filter table.<personid3、personid1、personid2> ,Finally, the data in the filter table is read into the frequent pattern tree in sequence to generate a frequent pattern tree.

[0054] For another example, in the embodiment of the present disclosure, the personnel link is<personid1、personid2、personid3、personid4> According to statistics, personid1 appears 5 times, personid2 appears 3 times, personid3 appears 7 times, and personid4 appears 4 times. Sort personid1, personid2, personid3, and personid4 by their frequencies to get a frequency table. Then re-sort them by their frequencies, and delete the personnel information with a frequency less than 2 to get a filter table.<personid3、personid1、personid4、personid2> ,Finally, the data in the filter table is read into the frequent pattern tree in sequence to generate a frequent pattern tree.

[0055] For another example, in the embodiment of the present disclosure, the personnel link is<personid1、personid2、personid3、personid4> According to statistics, personid1 appears 5 times, personid2 appears 3 times, personid3 appears 7 times, and personid4 appears 5 times (personid1 and personid4 have the same frequency). Sort personid1, personid2, personid3, and personid4 by their frequency of occurrence to get a frequency table. Then re-sort them by their frequency of occurrence, and delete the personnel information with a frequency of less than 2 to get a filter table.<personid3 / personid4、personid1、personid2> In this case, the embodiment of the present disclosure can add time sequence information. For example, the time point of personid1 is 9:40, and the time point of personid4 is 9:45. Then the new filter table is<personid4、personid3、personid1、personid2> Finally, the data in the filter table are read into the frequent pattern tree in sequence to generate a frequent pattern tree.

[0056] Through the above steps S301 to S303, useless and repeated data are filtered out, thereby improving the generation efficiency of the frequent pattern tree.

[0057] In some embodiments, the above-mentioned reading the face IDs in the filter table one by one to generate the frequent pattern tree includes but is not limited to:

[0058] S401, obtaining all personnel relationship records according to the face ID in the filter table, wherein the personnel relationship record is a sequence string sorted by time by the data acquisition device;

[0059] S402 , constructing nodes respectively with the personnel relationship records as tags until all the personnel relationship records are mapped into the frequent pattern tree, wherein the personnel relationship records are respectively mapped into a path of the frequent pattern tree.

[0060] For example, read the first record<v1、v2、v3> , and build nodes with v1, v2, and v3 as labels, to generate the path null->v1->v2->v3, and add 1 to the frequency between nodes on the path, which are initially 0. During the construction process, if there are the same prefix nodes between the paths (for example,<a,b,c,d> and<a,b,k,f> The prefix is<a,b> ), the nodes need to be merged; but if there are common nodes between paths, no additional processing is done. Until all records are built into the frequent tree, and all relationships are mapped to a path of the FP-tree. (such as Figure 3 As shown, by constructing a tree, multiple path strings can be formed, including the paths ABDC, ABDH, ABDI, ACEI, ACEJ).

[0061] For example, read the first record<v1、v2、v3、v4、v5> , and build nodes with v1, v2, v3, v4, and v5 as labels, to generate the path null->v1->v2->v3->v4->v5, and add 1 to the frequency between nodes on the path, which are all initially 0. During the construction process, if there are the same prefix nodes between the paths (for example,<a,b,c,d,e> and<a,b,c,k,f> The prefix is<a,b,c> ), the nodes need to be merged; but if there are common nodes between paths, no additional processing is done. This process continues until all records are built into the frequent tree and all relationships are mapped to a path in the FP-tree.

[0062] The above steps S401 to S402 further improve the generation efficiency of the frequent pattern tree.

[0063] S204: mining target data through the frequent pattern tree.

[0064] Optionally, mining the target data through the frequent pattern tree includes repeatedly performing the following steps until all nodes of the filter table are processed, and finally obtaining the target data:

[0065] S501, traversing the first node in the frequent pattern tree from back to front in reverse order according to the sequence of nodes in the filter table, to obtain a conditional base corresponding to the first node;

[0066] The above conditional base can be expressed as:

[0067] E p ={[(A,count A )],[(B,count B )], ..., [(X, count X )]}

[0068] S502, according to the conditional base, calculate the support of the second node other than the first node, and delete the personnel with less than the minimum support, wherein the second node is a node other than the first node, and the second node includes one or more nodes.

[0069] In other words, the correlation between person t and other nodes is directly obtained through the number of occurrences of the prefix path ending with node p on the FP-tree, thereby mining and returning the frequency information of each person appearing with other people at the same time. The target data is identified by mining through the generated FP tree, and finally the frequent pattern relationship that meets the minimum support condition is obtained.

[0070] In some embodiments, after mining the target data through the frequent pattern tree, the method further includes:

[0071] S601, setting a time decreasing function;

[0072] S602: Calculate the correlation degree of the target data that decays with time according to the time decreasing function.

[0073] Optionally, the above time decreasing function is determined by the following formula:

[0074] T i =1 / (1+a(T0-t i ))

[0075] Among them, t i is the average appearance time of the companions of person i, T0 is the current time, and a is the decay parameter over time.

[0076] Through the above S601 to S602, the correlation degree of the target data that decays with time can be calculated.

[0077] In summary, the disclosed embodiment captures multiple pictures from the personnel movement data stream, and obtains the facial information of the personnel included in each of the multiple pictures, thereby obtaining a facial information set, wherein the facial information includes the facial ID, the ID of the data acquisition device that captured the picture, and the capture time of the picture; based on the facial information set, a personnel link is generated, wherein the personnel link is a sequence string arranged in order according to the facial information; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, the target data is mined. That is to say, in the disclosed embodiment, the traditional frequent pattern mining frequent pattern tree method is introduced into the peer mining scenario, rather than simply calculating the time overlap degree to search for faces or querying peers by time intervals, thereby solving the problem of low efficiency in the method of mining peers in the related technology, and achieving the technical effect of improving the efficiency of mining peers.

[0078] The following uses peer mining as an example to illustrate the embodiments of the present disclosure.

[0079] After obtaining basic snapshot information and identifying the face ID in the image, the data is cleaned (abnormal and redundant data is cleaned). Then, the frequent pattern tree is generated using the relational data. After scanning the stored data, the frequent pattern relationship tree is dynamically constructed, and the time intervals between adjacent nodes are stored. Finally, the companions of the person can be mined according to the defined time interval. Specifically:

[0080] S701, using each camera to take a photo of a human body, performing face detection and recognition on the photo, and generating a personid for the face;

[0081] S702, obtaining the face personid, cameraid, and capture time of each picture;

[0082] S703: Based on a time period (such as a day), the personnel appearing in each camera are arranged in order to form a personnel link<personid1,personid3...> ;

[0083] S704, data preprocessing: Delete the data groups with less than the minimum support (generally set to 2, indicating two associations), where the support represents the number of times a certain relationship appears, and the minimum support represents the minimum number of times a relationship appears in the set frequent pattern. Sometimes the image collection interval is short, resulting in a large amount of duplicate data. For example, if a person is photographed multiple times in succession under the same camera, directly using this part of the data will seriously affect the program processing time. Therefore, the adjacent and continuous data of the same person under the same camera are merged into one, and the time is recorded as the time of the first appearance of this person under this camera during this period.

[0084] S705, generating a frequent pattern tree based on the personnel link. In the personnel link frequent pattern tree, by storing the attribute information of the nodes in the tree and the time difference between adjacent nodes, the frequent pattern tree can be dynamically constructed by scanning the storage table once. The rules are as follows:

[0085] (1) Traverse all the data, obtain the frequency of occurrence of each personid, and remove the elements that do not meet the minimum support, and finally obtain the frequency table.

[0086] (2) Based on the frequency table, reorder each element in the data according to its frequency of occurrence, and delete the elements that do not appear in the frequency table to generate a filter table.

[0087] (3) By reading the data in the filter table one by one, the personnel relationship is mapped to the path of the FP-tree. Start reading the personnel relationship records, for example, read the first record<v1,v2> , and build nodes with v1 and v2 as labels, which can generate the path null->v1->v2, and add 1 to the frequency between nodes on the path, which are all initially 0. During the construction process, if there are the same prefix nodes between paths, the nodes need to be merged; but if there are the same nodes between paths, no additional processing is done. This continues until all records are built into the frequent tree and all relationships are mapped to a path of the FP-tree.

[0088] S706, frequent pattern mining, the association between person t and other nodes is directly obtained through the number of occurrences of the prefix path ending with node p on the FP-tree, thereby mining and returning the frequency information of each person appearing with other people at the same time. The identification of fellow travelers is mined through the generated FP tree, and finally the frequent pattern relationship that meets the minimum support condition is obtained.

[0089] (1) According to the order of the nodes in the filter table, traverse the node p from back to front in reverse order to obtain the conditional base corresponding to the node: According to the hierarchical structure of the FP-tree, traverse from the root node to a certain node, and change the count value count of the traversed node at the same time t Finally, the conditional pattern base of p is obtained, which is expressed as:

[0090] E p ={[(A, count A )],[(B,count B )], ..., [(X, count X )]}

[0091] (2) Calculate the support of all nodes except node x in the above formula, and then remove the elements with less than the minimum support;

[0092] (3) Repeat operations (1) and (2) until all nodes in the filter table are processed, and finally obtain the frequent item sets between people.

[0093] S707. The relationship between the companions will change over time. The latest companions can better reflect the importance of the user than the past companions. Add a function T that decreases over time. i =1 / (1+a(T0-t i )), where t i is the average appearance time of peers of user i, T0 is the current time, and a is the decay parameter over time.

[0094] Each person's companions are returned in order of frequency from high to low, and a list of each user's companions can be obtained. At the same time, combined with the time decay function, the degree of association of companions that decays over time can be calculated.

[0095] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0096] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.

[0097] Figure 4 Schematic diagram of a data mining device provided by an embodiment of the present disclosure. Figure 4 As shown, the data mining device includes:

[0098] The acquisition module 401 is configured to capture multiple pictures from the personnel movement data stream, and obtain the face information of the person included in each of the multiple pictures to obtain a face information set, wherein the face information includes a face ID, an ID of a data acquisition device that captures the picture, and a capture time point of the picture;

[0099] It should be noted that the application scenarios of the embodiments of the present disclosure include but are not limited to the following application scenarios:

[0100] Application scenario one: mining companions of a shopping mall consumer group. For example, the data mining method provided by the embodiment of the present disclosure can be used to capture multiple pictures from a consumer group, and obtain facial information of the consumer group included in each of the multiple pictures to obtain a facial information set; based on the facial information set, a personnel link is generated; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, target data is mined. In other words, introducing the traditional frequent pattern mining method of the frequent pattern tree into the mining scenario of companions of a consumer group can effectively capture companions and effectively convert the companions into shopping mall customers;

[0101] Application scenario two: mining the number of users received by service personnel. For example, the data mining method provided by the embodiment of the present disclosure can be used to capture multiple pictures from all the reception user data streams of the service personnel, and obtain the facial information of the received users included in each of the multiple pictures to obtain a facial information set; based on the facial information set, a personnel link is generated; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, the target data is mined. In other words, the traditional frequent pattern mining method of the frequent pattern tree is introduced into the mining scenario of the reception personnel, so that the number of users received by the service personnel, the frequency of receiving users, etc. can be mined.

[0102] Application scenario three: in security scenarios or airport scenarios, identify whether different pedestrians are in a walking relationship. For example, in a security scenario, the data mining method provided by the embodiment of the present disclosure is used to capture multiple pictures from the pedestrian data stream, and obtain the facial information of the pedestrians included in each of the multiple pictures to obtain a facial information set; based on the facial information set, a personnel link is generated; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, the target data is mined. In other words, introducing the traditional frequent pattern mining method of the frequent pattern tree into the mining scenario of dangerous people can clarify the social relationship between pedestrian targets, discover potential dangers and take timely precautions; on the other hand, when a danger occurs, pedestrian information, as well as information such as the time and place of the danger, can be obtained to provide important clues for investigation.

[0103] In addition, it should be noted that the above-mentioned personnel movement data stream may be the personnel data stream of a certain shopping mall channel on a certain day or during a certain morning or evening peak period, and the personnel movement data stream includes people traveling together and people traveling alone. The disclosed embodiment aims to mine people traveling together from the above-mentioned personnel movement data stream as target data.

[0104] The above data collection device can be a camera, a camera, a mobile phone, etc. For example, taking a camera as an example, the layout requirements of the camera include but are not limited to: the camera is arranged at a height of 2-3 meters, with a depression angle of 15-20 degrees, and is generally deployed at important passages and gates, facing the face of people. For another example, taking a mobile phone as an example, the mobile phone needs to be able to clearly capture all faces in the entire picture.

[0105] Optionally, the data acquisition device captures multiple pictures from the personnel movement data stream in a manner including but not limited to the following manners, which are not limited in any way herein:

[0106] Method 1: capture multiple pictures from the above-mentioned personnel movement data stream every 1 minute, and select the picture with the highest definition from the multiple pictures;

[0107] Method 2: When the flow of mobile personnel is large, multiple pictures are captured from the above-mentioned mobile personnel data stream every 1 minute. When the flow of mobile personnel is small, multiple pictures are captured from the above-mentioned mobile personnel data stream every half minute, and then the picture with the highest definition is selected from the pictures.

[0108] Optionally, the method of obtaining the above-mentioned face ID includes but is not limited to: first detecting and identifying the face in the image, and then numbering the face, for example, personid1, personid2, personid3, etc.

[0109] The above-mentioned face recognition method includes but is not limited to the following methods, which are not limited in any way:

[0110] Method 1: extract two-dimensional face key point information and three-dimensional face key point information from the image frame in the face video, calculate the face posture angle based on the two-dimensional face key point information and the three-dimensional face key point information, and judge whether the face posture angle meets the preset conditions. When the face posture angle meets the preset conditions, extract the face image area in the image frame, and judge whether the face image in the face image area is clear based on a preset clear face classifier. When the face image in the face image area is clear, output the face image area.

[0111] Method 2: First, by searching the Haar features of the upper body of the human body, determine that the moving object is a human, and then screen and cut out the color image of the head larger than 39*39 pixels; then, through the local weight sharing convolution formula and the local weight sharing sampling formula, determine the five feature areas of the face, the left eye, the right eye, the tip of the nose, the left corner of the mouth and the right corner of the mouth, and find the center point in the feature area; extract the deep relationship value of the feature area, and convert all the relationship values ​​into a matrix; finally, compare the values ​​in the matrix with the values ​​in the database, and perform probability analysis through the Gaussian model to obtain the forward or reverse value to determine the recognition of the face.

[0112] A first generating module 402 is configured to generate a personnel link according to the face information set, wherein the personnel link is a sequence string arranged in sequence according to the face information;

[0113] For example, based on the time period, the people who appear in each camera can be arranged in order to form a personnel link<personid1,personid2、personid3...> .

[0114] The second generating module 403 is configured to generate a frequent pattern tree according to the personnel link;

[0115] Optionally, in the embodiment of the present disclosure, if Figure 5 As shown, the data mining device also includes a processing module 501, which is configured to delete the face information with less than the minimum support before generating the frequent pattern tree according to the personnel link, wherein the support is the number of times the person appears in the time relationship when being continuously captured by the same data acquisition device; or, merge the repeated face information. Because sometimes the image acquisition interval is short, a large amount of repeated data is generated, such as a person is continuously photographed with multiple pictures under the same camera, and directly using this part of the data will seriously affect the program processing time. Therefore, the adjacent and continuous data of the same person under the same camera are merged into one, and the time is recorded as the time of the first appearance in this period under this camera, thereby improving the mining efficiency of the target data.

[0116] Optionally, in the embodiment of the present disclosure, if Figure 6 As shown, the second generation module 403 includes:

[0117] The first acquisition unit 601 is configured to traverse the personnel link, obtain the frequency of occurrence of each face ID, and obtain a frequency table;

[0118] The processing unit 602 is configured to sort each face ID in the personnel link according to the frequency of occurrence according to the frequency table, and delete the face ID that does not appear in the frequency table to obtain a filter table;

[0119] The generating unit 603 is configured to read the face IDs in the filter table one by one to generate the frequent pattern tree.

[0120] For example, in the embodiment of the present disclosure, the personnel link is<personid1、personid2、personid3、personid4> According to statistics, personid1 appears 5 times, personid2 appears 3 times, personid3 appears 7 times, and personid4 appears 1 time. Sort personid1, personid2, personid3, and personid4 by their frequencies to get a frequency table. Then re-sort them by their frequencies, and delete the personnel information with a frequency less than 2 to get a filter table.<personid3、personid1、personid2> ,Finally, the data in the filter table is read into the frequent pattern tree in sequence to generate a frequent pattern tree.

[0121] For another example, in the embodiment of the present disclosure, the personnel link is<personid1、personid2、personid3、personid4> According to statistics, personid1 appears 5 times, personid2 appears 3 times, personid3 appears 7 times, and personid4 appears 4 times. Sort personid1, personid2, personid3, and personid4 by their frequencies to get a frequency table. Then re-sort them by their frequencies, and delete the personnel information with a frequency less than 2 to get a filter table.<personid3、personid1、personid4、personid2> ,Finally, the data in the filter table is read into the frequent pattern tree in sequence to generate a frequent pattern tree.

[0122] For another example, in the embodiment of the present disclosure, the personnel link is<personid1、personid2、personid3、personid4> According to statistics, personid1 appears 5 times, personid2 appears 3 times, personid3 appears 7 times, and personid4 appears 5 times (personid1 and personid4 have the same frequency). Sort personid1, personid2, personid3, and personid4 by their frequency of occurrence to get a frequency table. Then re-sort them by their frequency of occurrence, and delete the personnel information with a frequency of less than 2 to get a filter table.<personid3 / personid4、personid1、personid2> In this case, the embodiment of the present disclosure can add time sequence information. For example, the time point of personid1 is 9:40, and the time point of personid4 is 9:45. Then the new filter table is<personid4、personid3、personid1、personid2> Finally, the data in the filter table are read into the frequent pattern tree in sequence to generate a frequent pattern tree.

[0123] pass Figure 6 The device shown filters out useless and repeated data, thereby improving the generation efficiency of the frequent pattern tree.

[0124] Optionally, in the embodiment of the present disclosure, if Figure 7 As shown, the above-mentioned generating unit 603 includes:

[0125] The acquisition subunit 701 is configured to acquire all personnel relationship records according to the face ID in the filter table, wherein the personnel relationship record is a sequence string sorted by time by the data acquisition device;

[0126] The mapping subunit 702 is configured to construct nodes respectively with the personnel relationship records as tags until all the personnel relationship records are mapped into the frequent pattern tree, wherein the personnel relationship records are respectively mapped into a path of the frequent pattern tree.

[0127] For example, read the first record<v1、v2、v3> , and build nodes with v1, v2, and v3 as labels, to generate the path null->v1->v2->v3, and add 1 to the frequency between nodes on the path, which are initially 0. During the construction process, if there are the same prefix nodes between the paths (for example,<a,b,c,d> and<a,b,k,f> The prefix is<a,b> ), the nodes need to be merged; but if there are common nodes between paths, no additional processing is done. Until all records are built into the frequent tree, and all relationships are mapped to a path of the FP-tree. (such as Figure 3 As shown, by constructing a tree, multiple path strings can be formed, including the paths ABDC, ABDH, ABDI, ACEI, ACEJ).

[0128] For example, read the first record<v1、v2、v3、v4、v5> , and build nodes with v1, v2, v3, v4, and v5 as labels, to generate the path null->v1->v2->v3->v4->v5, and add 1 to the frequency between nodes on the path, which are all initially 0. During the construction process, if there are the same prefix nodes between the paths (for example,<a,b,c,d,e> and<a,b,c,k,f> The prefix is<a,b,c> ), the nodes need to be merged; but if there are common nodes between paths, no additional processing is done. This process continues until all records are built into the frequent tree and all relationships are mapped to a path in the FP-tree.

[0129] pass Figure 7 The device shown further improves the generation efficiency of the frequent pattern tree.

[0130] The mining module 404 is configured to mine target data through the frequent pattern tree.

[0131] Alternatively, if Figure 8 As shown, the mining module 404 includes:

[0132] The second acquisition unit 801 is configured to mine the target data through the frequent pattern tree, including repeatedly performing the following steps until all the nodes of the filter table are processed, and finally the target data is obtained: according to the order of the nodes in the filter table, traverse the first node in the frequent pattern tree from back to front in reverse order to obtain the condition base corresponding to the first node;

[0133] The above conditional base can be expressed as:

[0134] E p ={[(A,count A )],[(B,countB )],...,[(X,count X )]}

[0135] The deletion unit 802 is configured to calculate the support of a second node other than the first node based on the conditional base, and delete the personnel whose support is less than the minimum support, wherein the second node is a node other than the first node, and the second node includes one or more nodes.

[0136] In other words, the correlation between person t and other nodes is directly obtained through the number of occurrences of the prefix path ending with node p on the FP-tree, thereby mining and returning the frequency information of each person appearing with other people at the same time. The target data is identified by mining through the generated FP tree, and finally the frequent pattern relationship that meets the minimum support condition is obtained.

[0137] In some embodiments, Fig. 9 As shown, the above device also includes:

[0138] A setting module 901 is configured to set a time decreasing function after mining the target data through the frequent pattern tree;

[0139] The calculation module 902 is configured to calculate the correlation degree of the target data that decays with time according to the time decreasing function.

[0140] Optionally, the above time decreasing function is determined by the following formula:

[0141] T i =1 / (1+a(T0-t i ))

[0142] Among them, t i is the average appearance time of the companions of person i, T0 is the current time, and a is the decay parameter over time.

[0143] pass Fig. 9 The device shown can calculate the correlation degree of target data that decays with time.

[0144] In summary, the disclosed embodiment captures multiple pictures from the personnel movement data stream, and obtains the facial information of the personnel included in each of the multiple pictures, thereby obtaining a facial information set, wherein the facial information includes the facial ID, the ID of the data acquisition device that captured the picture, and the capture time of the picture; based on the facial information set, a personnel link is generated, wherein the personnel link is a sequence string arranged in order according to the facial information; based on the personnel link, a frequent pattern tree is generated; and through the frequent pattern tree, the target data is mined. That is to say, in the disclosed embodiment, the traditional frequent pattern mining frequent pattern tree method is introduced into the peer mining scenario, rather than simply calculating the time overlap degree to search for faces or querying peers by time intervals, thereby solving the problem of low efficiency in the method of mining peers in the related technology, and achieving the technical effect of improving the efficiency of mining peers.

[0145] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.

[0146] Fig.10 is a schematic diagram of an electronic device 10 provided in an embodiment of the present disclosure. Fig.10 As shown, the electronic device 10 of this embodiment includes: a processor 1001, a memory 1002, and a computer program 1003 stored in the memory 1002 and executable on the processor 1001. When the processor 1001 executes the computer program 1003, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 1001 executes the computer program 1003, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0147] Exemplarily, the computer program 1003 may be divided into one or more modules / units, which are stored in the memory 1002 and executed by the processor 1001 to complete the present disclosure. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 1003 in the electronic device 10.

[0148] The electronic device 10 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 10 may include, but is not limited to, a processor 1001 and a memory 1002. Those skilled in the art will appreciate that Fig.10 It is only an example of the electronic device 10 and does not constitute a limitation of the electronic device 10. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0149] The processor 1001 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0150] The memory 1002 may be an internal storage unit of the electronic device 10, for example, a hard disk or memory of the electronic device 10. The memory 1002 may also be an external storage device of the electronic device 10, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the electronic device 10. Further, the memory 1002 may also include both an internal storage unit of the electronic device 10 and an external storage device. The memory 1002 is used to store computer programs and other programs and data required by the electronic device. The memory 1002 may also be used to temporarily store data that has been output or is to be output.

[0151] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0152] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0153] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0154] In the embodiments provided in the present disclosure, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, which may be electrical, mechanical or other forms.

[0155] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0156] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0157] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present disclosure implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0158] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. A data mining method, characterized in that: include: Capture multiple pictures from a personnel movement data stream, and obtain face information of the person included in each of the multiple pictures to obtain a face information set, wherein the face information includes a face ID, an ID of a data acquisition device that captures the picture, and a capture time of the picture; the face ID is obtained by first detecting and identifying the face in the picture, and then numbering the face; Generate a personnel link according to the face information set, wherein the personnel link is a sequence string arranged in order according to the face information; Generate a frequent pattern tree according to the personnel link; Mining target data through the frequent pattern tree; The generating a frequent pattern tree according to the personnel link comprises: Traverse the personnel link, obtain the frequency of each face ID, and obtain a frequency table; According to the frequency table, sort each face ID in the personnel link according to the frequency of occurrence, and delete the face ID that does not appear in the frequency table to obtain a filter table; Read the face IDs in the filter table one by one to generate the frequent pattern tree; The step of reading the face IDs in the filter table one by one to generate the frequent pattern tree comprises: According to the face ID in the filter table, all personnel relationship records are obtained, wherein the personnel relationship records are sequence strings sorted by time by the data acquisition device; Nodes are constructed respectively with the personnel relationship records as tags until all the personnel relationship records are mapped into the frequent pattern tree, wherein the personnel relationship records are respectively mapped into a path of the frequent pattern tree.

2. The method according to claim 1, characterized in that Before generating a frequent pattern tree according to the personnel link, the method further includes: Deleting face information with a support value less than a minimum support value, wherein the support value is the number of times the person is continuously captured by the same data acquisition device; or Merge duplicate face information.

3. The method according to claim 1, characterized in that Mining target data through the frequent pattern tree includes repeatedly performing the following steps until all nodes of the filter table are processed, and finally obtaining the target data: According to the order of the nodes in the filter table, traverse the first node in the frequent pattern tree from back to front in reverse order to obtain a conditional base corresponding to the first node; According to the conditional base, the support of second nodes other than the first node is calculated, and the persons with support less than the minimum support are deleted, wherein the second node is a node other than the first node, and the second node includes one or more nodes.

4. The method according to claim 1, characterized in that After mining the target data through the frequent pattern tree, the method further includes: Set up a time-decreasing function; According to the time decreasing function, the correlation degree of the target data which decays with time is calculated.

5. The method according to claim 4, characterized in that The time decreasing function is determined by the following formula: T i =1 / (1+a(T0-t i )) Among them, t i is the average appearance time of the companions of person i, T1 is the current time, and a is the decay parameter over time.

6. A data mining device, characterized in that: include: The acquisition module is used to capture multiple pictures from the personnel movement data stream, and obtain the face information of the person included in each of the multiple pictures to obtain a face information set, wherein the face information includes a face ID, an ID of a data acquisition device that captures the picture, and a capture time point of the picture; the face ID is obtained by first detecting and identifying the face in the picture, and then numbering the face; A first generating module, configured to generate a personnel link according to the face information set, wherein the personnel link is a sequence string arranged in sequence according to the face information; A second generating module, used for generating a frequent pattern tree according to the personnel link; A mining module, used for mining target data through the frequent pattern tree; The second generation module comprises: A first acquisition unit is configured to traverse the personnel link, obtain the frequency of occurrence of each face ID, and obtain a frequency table; A processing unit is configured to sort each face ID in the personnel link according to the frequency table according to the frequency of occurrence, and delete the face ID that does not appear in the frequency table to obtain a filter table; A generating unit configured to read the face IDs in the filtering table one by one and generate the frequent pattern tree; The generating unit comprises: an acquisition subunit, configured to acquire all personnel relationship records according to the face ID in the filtering table, wherein the personnel relationship records are sequence strings sorted by time by the data acquisition device; The mapping subunit is configured to construct nodes with the personnel relationship records as tags respectively until all the personnel relationship records are mapped into the frequent pattern tree, wherein the personnel relationship records are respectively mapped into a path of the frequent pattern tree.

7. An electronic 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 steps of the method according to any one of claims 1 to 5 are implemented.

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