CN115329265A

CN115329265A - Method, device and equipment for determining graph code track association degree and storage medium

Info

Publication number: CN115329265A
Application number: CN202211050967.1A
Authority: CN
Inventors: 王东锋; 梁杨智; 杨德武
Original assignee: Shenzhen Qianhai Zhongdian Huian Technology Co ltd
Current assignee: Shenzhen Qianhai Zhongdian Huian Technology Co ltd
Priority date: 2022-08-30
Filing date: 2022-08-30
Publication date: 2022-11-11

Abstract

The invention discloses a method, a device, equipment and a storage medium for determining graph code track association degree. The method comprises the following steps: acquiring face image data and International Mobile Subscriber Identity (IMSI) data within a preset area range; clustering the facial image data and the IMSI data to respectively obtain the same facial image motion track and the same IMSI motion track; constructing a graph code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track; carrying out singular value decomposition and dimension reduction processing on the graph code matrix to obtain a singular value graph code matrix; and determining the track association degree of the graph code according to the singular value graph code matrix for generating motion track information. The embodiment of the invention reduces the complexity of the graph code association and improves the accuracy of the graph code association.

Description

Method, device and equipment for determining graph code track association degree and storage medium

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for determining a graph code track association degree.

Background

With the advance of the urbanization process in China, the mobility of personnel and materials is increased day by day, and the potential risk of cities is increased. Therefore, the improvement of the construction of the urban security system is very important. In an urban security system, monitoring and early warning of important personnel and important places become a very important ring for urban security. In modern society, people usually do not leave their mobile phones. Therefore, in the detection and prevention and control scene, the graph code association can greatly enrich the running track information of people.

The existing graph code correlation techniques are mainly divided into two categories: the first type is graph code association based on an algorithm model under a big data condition; the second category is artificial modeling based on a priori rules. However, the first type of graph code association mode has higher complexity and lower real-time performance; the second type of graph code association mode depends on human experience to carry out calculation and judgment, and is low in complexity, but the accuracy of graph code association is low, and complex scenes under the condition of big data cannot be dealt with.

Disclosure of Invention

The invention provides a method, a device and equipment for determining the track association degree of a graph code and a storage medium, which are used for reducing the association complexity of the graph code and improving the association accuracy of the graph code.

According to an aspect of the present invention, a method for determining a track association degree of a graph code is provided, where the method includes:

acquiring face image data and International Mobile Subscriber Identity (IMSI) data within a preset area range;

clustering the facial image data and the IMSI data to respectively obtain the same facial image motion track and the same IMSI motion track;

constructing an image code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track;

carrying out singular value decomposition and dimension reduction processing on the graph code matrix to obtain a singular value graph code matrix;

and determining the track association degree of the graph code according to the singular value graph code matrix for generating motion track information.

According to another aspect of the present invention, there is provided an apparatus for determining a track association degree of a graph code, the apparatus including:

the data acquisition module is used for acquiring the face image data and the International Mobile Subscriber Identity (IMSI) data in a preset area range;

a motion track determining module, configured to cluster the face image data and the IMSI data to obtain a same face image motion track and a same IMSI motion track, respectively;

the image code matrix construction module is used for constructing an image code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track;

the singular value image code matrix determining module is used for carrying out singular value decomposition and dimension reduction processing on the image code matrix to obtain a singular value image code matrix;

and the track association degree determining module is used for determining the track association degree of the graph code according to the singular value graph code matrix and generating motion track information.

According to another aspect of the present invention, there is provided an electronic apparatus including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the method for determining the association degree of a graph code track according to any embodiment of the present invention.

According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the method for determining a correlation degree of a graph code track according to any embodiment of the present invention when the computer instructions are executed.

The embodiment of the invention obtains the face image data and the international mobile subscriber identity IMSI data in the preset area range; clustering the facial image data and the IMSI data to respectively obtain the same facial image motion track and the same IMSI motion track; constructing an image code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track; singular value decomposition and dimension reduction processing are carried out on the graph code matrix to obtain a singular value graph code matrix; and determining the track association degree of the graph code according to the singular value graph code matrix for generating motion track information. According to the scheme, the real-time determination of the association degree of the image code track is realized by processing the acquired face image data and IMSI data in real time and generating and updating the corresponding image code matrix in real time. And a large-scale network model is not required to be trained based on a large number of face images and IMSI data, and the complexity of determining the association degree of the image code track is reduced. Compared with a mode of determining the graph code track association degree by considering that modeling completely depends on human experience based on prior rules, the method and the device for determining the graph code track association degree reduce the graph code association complexity and improve the accuracy of graph code association.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present invention, nor do they necessarily limit the scope of the invention. Other features of the present invention will become apparent from the following description.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a flowchart of a method for determining a track association degree of a graph code according to an embodiment of the present invention;

fig. 2 is a flowchart of a method for determining a track association degree of a graph code according to a second embodiment of the present invention;

fig. 3 is a schematic structural diagram of a device for determining association degree of a graph code track according to a third embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device implementing the method for determining association degree of image code track according to the embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Example one

Fig. 1 is a flowchart of a method for determining a graph track association degree according to an embodiment of the present invention, where the embodiment is applicable to a case where a face image and an IMSI (International Mobile Subscriber Identity) belonging to the same user are associated with a graph, the method may be executed by a graph track association degree determining device, the graph track association degree determining device may be implemented in a hardware and/or software manner, and the graph track association degree determining device may be configured in an electronic device. As shown in fig. 1, the method includes:

s110, acquiring the face image data and the international mobile subscriber identity IMSI data in the preset area range.

Wherein the preset area range can be preset by a related technician. For example, the preset area range may be a city or a certain area in a city, which is not limited in this embodiment and may be preset according to actual situations.

The face image data can be acquired by an image acquisition device. For example, at least one image acquisition device for acquiring a stable human face image data source can be deployed under each scene area within a preset area range. For example, the image capture device may be a smart face camera.

The IMSI data may be obtained by the spying equipment. The code detection equipment can be equipment capable of acquiring IMSI signals of mobile phones within a certain range, and specifically can acquire IMSI codes, acquisition time, geographical positions and other information. For example, at least one code detection device for stably acquiring IMSI data may be deployed in each scene area within a preset area range.

It should be noted that the code detecting devices and the image capturing devices within the preset area range are randomly distributed, and since the effective capturing range of the code detecting devices is 200-1500 meters, a plurality of image capturing devices may exist around one code detecting device.

For example, the face image data and the IMSI data in the preset area range may be acquired by the background server. Specifically, when all image acquisition devices deployed in a preset area range acquire a face image, face image data is uploaded to a background server in real time; and uploading the acquired IMSI data to a background server in real time when all code detecting devices deployed in the preset area range acquire the IMSI of the mobile phone.

And S120, clustering the face image data and the IMSI data to respectively obtain the same face image motion track and the same IMSI motion track.

The same face image motion trail can be a motion trail generated aiming at the same face image; the same IMSI movement trace may be a movement trace generated for the same IMSI.

Illustratively, for face image data, when the background server acquires the face image data (x, y, t, d), the background server judges whether motion trajectory data corresponding to the face image is stored in real time; if yes, real-time aggregation is carried out on the obtained face image data and the existing motion trail data, and a new motion trail corresponding to the face image data is obtained; if not, generating a motion track corresponding to the face image in real time. Where x and y may be represented as the geographic location of the face image at time t, and d may be represented as an identity Identification (ID) of an image capturing device that obtains the face image. Wherein, the motion track of the face image can be expressed as { (x) ₁ ,y ₁ ,t ₁ ,d ₁ ),(x ₂ ,y ₂ ,t ₂ ,d ₂ ),…,(x _n ,y _n ,t _n ,d _n ) }. Wherein n represents the number of elements in the motion trajectory, that is, the number of the face image data acquired by the same or different image acquisition devices at different time points in different geographic positions within a preset time period.

It should be noted that the background server may also perform clustering processing on the face image data acquired within a preset time period at preset time. Exemplarily, the background server may perform unified clustering on the face image data acquired on the same day at time 00.

Illustratively, if the motion trajectories of the face image a, the face image B and the face image C are aggregated in the background server, and if the background server acquires the face image data of the face image B from the image acquisition device in real time, aggregating the face image data acquired in real time with the motion trajectory of the face image B to obtain a new motion trajectory of the face image B.

Exemplary ofWhen the background server acquires IMSI data (x, y, t, d) according to the IMSI data, the background server judges whether motion trail data corresponding to the IMSI is stored in real time; if so, performing real-time aggregation on the obtained IMSI data and the existing motion trail data to obtain a new motion trail corresponding to the IMSI; if not, generating a motion track corresponding to the IMSI in real time. Where x and y may represent the geographical location of the IMSI at time t, and d may represent the spy ID for obtaining the IMSI data. Wherein, the IMSI motion trajectory can be expressed as { (x) ₁ ,y ₁ ,t ₁ ,d ₁ ),(x ₂ ,y ₂ ,t ₂ ,d ₂ ),…,(x _n ,y _n ,t _n ,d _n ) }. Wherein n represents the number of elements in the motion trajectory, that is, the number of IMSI data acquired by the same or different code detection devices at different time points in different geographic locations within a preset time period.

It should be noted that, the face image data and the IMSI data acquired by the background server in real time are large, so after the face image data and the IMSI data are acquired, the acquired face image data are subjected to the same face image clustering in real time, and the acquired IMSI data are subjected to the same IMSI data clustering in real time, so that the instantaneity of determining the same face image motion trajectory and the same IMSI motion trajectory is improved, real-time response can be made to the inflowing face image data and IMSI data, and therefore, the timely response to the subsequently generated image code trajectory association degree is facilitated to be improved, and the real-time generation of the image code trajectory association degree is realized.

It can be understood that, after clustering the face image data in the background server, at least one identical face image motion trajectory can be obtained, and after clustering the IMSI data, at least one identical IMSI motion trajectory can be obtained. The same facial image motion trajectory may be a result of aggregating facial image data of the same facial image acquired by the same or different image acquisition devices at different times. The same IMSI movement trajectory may be a result of aggregating IMSI data of the same IMSI acquired by the same or different code detecting devices at different times.

It should be noted that, in order to ensure the accuracy of the clustering result obtained after the background server clusters the face image data and the IMSI data, the obtained face image data and IMSI data may be preprocessed.

For example, the face image data and/or IMSI data with abnormal geographic location coordinates may be removed, and the face image data of the non-face image may be removed, with respect to the obtained face image data and/or IMSI data. And the face image data and/or IMSI data without missing key fields such as equipment place ID or acquisition time and the like.

S130, constructing a graph code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track.

The image code matrix is used for representing the corresponding relation between the face image and the IMSI, namely the mapping relation of the face image and the IMSI in time and space.

Illustratively, a human face spatiotemporal relationship matrix of the same human face image motion track in time and space and an IMSI spatiotemporal relationship matrix of the same IMSI motion track in time and space can be respectively established. And constructing an image code matrix between the face image and the IMSI based on the face spatiotemporal relationship matrix and the IMSI spatiotemporal relationship matrix through the incidence relation between each code detection device and the image acquisition device in the preset area range.

And S140, carrying out singular value decomposition and dimension reduction processing on the graph code matrix to obtain a singular value graph code matrix.

Illustratively, singular value decomposition may be performed on the graph code matrix to obtain an expansion term corresponding to the graph code matrix. Because the graph code matrix is quite sparse, in order to improve the subsequent data processing efficiency, the graph code matrix can be subjected to dimension reduction according to the expansion items, only a part of singular values, corresponding left singular column vectors and corresponding right singular column vectors are reserved, and the singular value graph code matrix after dimension reduction is obtained.

And S150, determining the graph code track association degree according to the singular value graph code matrix, and generating motion track information.

For example, the graph code track association degree may be determined according to an association degree network model obtained through pre-training. The relevance network model can be obtained by training based on a singular value image code matrix obtained in a historical period. Specifically, the singular value graph code matrix may be input into the relevance network model, and the model output result is used as the graph code track relevance.

The graph code track association degree is used for representing the association degree between the face image and the IMSI, namely determining the possibility degree of the face image and the IMSI belonging to the same user. The method can be used for generating the motion trail information of the user after the graph code trail association degree is determined, and can be applied to various fields such as intelligent security construction.

The embodiment of the invention obtains the face image data and the international mobile subscriber identity IMSI data in the preset area range; clustering the face image data and the IMSI data to respectively obtain the same face image motion track and the same IMSI motion track; constructing an image code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track; singular value decomposition and dimension reduction processing are carried out on the graph code matrix to obtain a singular value graph code matrix; and determining the graph code track association degree according to the singular value graph code matrix for generating motion track information. According to the scheme, the real-time determination of the association degree of the image code track is realized by processing the acquired face image data and IMSI data in real time and generating and updating the corresponding image code matrix in real time. And a large-scale network model is not required to be trained based on a large number of face images and IMSI data, and the complexity of determining the association degree of the image code track is reduced. Compared with a mode of determining the graph code track association degree by considering that modeling completely depends on human experience based on prior rules, the method and the device for determining the graph code track association degree reduce the graph code association complexity and improve the accuracy of the graph code association.

Example two

Fig. 2 is a flowchart of a method for determining a track association degree of an image code according to a second embodiment of the present invention, and this embodiment performs optimization and improvement on the basis of the foregoing technical solutions.

Further, the step of ' constructing a picture code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track ' is refined into the step of ' constructing a face space-time relation matrix according to the same face image motion track; constructing an IMSI space-time relation matrix according to the same IMSI motion trajectory; and constructing an image code matrix between the face image and the IMSI according to the face space-time relation matrix and the IMSI space-time relation matrix. "to perfect the construction mode of the graph code matrix between the face image and the IMSI.

Further, the singular value decomposition and dimension reduction processing are carried out on the graph code matrix in the step to obtain a singular value graph code matrix, the singular value graph code matrix is refined into the singular value decomposition is carried out on the graph code matrix, and the singular value corresponding to the graph code matrix is obtained; determining the number of column vectors of the image code matrix; determining the number of dimensionality reduction singular values according to the number of the column vectors and a preset element threshold value; and performing dimension reduction processing on the graph code matrix according to the dimension reduction singular value quantity and the singular value to obtain a singular value graph code matrix. "to refine the determination of the singular value map code matrix.

Further, the step of determining the track association degree of the graph code according to the singular value graph code matrix is refined into the step of determining a reference number sequence according to the number of dimension-reduced singular values; and determining the track association degree of the graph code according to the singular value graph code matrix and the reference number sequence. "to refine the determination method of the correlation degree of the graph code track.

As shown in fig. 2, the method includes the following specific steps:

s210, acquiring the face image data and the IMSI data in the preset area range.

And S220, clustering the face image data and the IMSI data to respectively obtain the same face image motion track and the same IMSI motion track.

And S230, constructing a human face space-time relation matrix according to the same human face image motion trail.

Illustratively, for any same face image motion track, a corresponding face spatiotemporal relationship matrix can be determined. The time discretization and the space unification can be carried out on the same face image motion trail, so that a time-space relation matrix on a time dimension and a space dimension is constructed.

In an alternative embodiment, the constructing of the face spatiotemporal relationship matrix according to the motion trajectories of the same face images comprises: performing track segmentation on the same face image motion track according to a preset time slice length to obtain at least one same face image motion sub-track; constructing a human face space-time relationship matrix according to the association relationship between each same human face image motion sub-track and at least one image acquisition device in a preset area range; the image acquisition equipment is used for acquiring face image data.

Wherein the time slice length may be preset by a person skilled in the art. For example, the time slice length may be 10 minutes.

Illustratively, if the motion trail A of the same face image is { (x) ₁ ,y ₁ ,t ₁ ,d ₁ ),(x ₂ ,y ₂ ,t ₂ ,d ₂ ),…,(x _n ,y _n ,t _n ,d _n ) And slicing the same face image motion trajectory A according to a preset time slice length alpha to obtain at least one same face image motion sub-trajectory which is respectively the same face image motion sub-trajectory A1{ (x) ₁ ,y ₁ ,t ₁ ,d ₁ )…,(x _k ,y _k ,t _k ,d _k ) And (x) moving sub-track A2 of the same face image _k+1 ,y _k+1 ,t _k+1 ,d _k+1 )…,(x _m ,y _m ,t _m ,d _m ) ' 8230 _m+1 ,y _m+1 ,t _m+1 ,d _m+1 )…,(x _n ,y _n ,t _n ,d _n ) }. The sub-tracks A1, A2, \ 8230of the same face image motion, the number of elements in An may be the same or different, but the time lengths are the same and are all alpha. For example, if α is 10min, the elements in the motion sub-trajectory A1 of the same face image may be the facial image data of the face image acquired within a time period of 00 to 00; the elements in the same facial image motion sub-trajectory A2 may be facial image data of the facial image acquired within a time period of 00. This embodiment will not be described in detail。

Determining at least one image acquisition device in a preset area range, determining the incidence relation between each image acquisition device and the same face image motion sub-track, and constructing a face spatiotemporal relation matrix based on the incidence relation.

Illustratively, if n image acquisition devices exist within the preset area range. The number of sub-tracks corresponding to the same face image motion track is m, that is, the sub-tracks correspond to m groups of time slices, and then the face spatiotemporal relationship matrix a of the face image motion track a can be represented by the following matrix.

Wherein, a _m,n Whether the nth image acquisition equipment acquires the face image or not can be represented in a time period corresponding to the sub-track m of the same face image motion track; if so, then a _m,n Has a value of 1; if not, then a _m,n The value of (d) is 0.

For a _0,0 ,……,a _0,n It may represent whether the 0 th to n th image capturing devices capture the face image at the 0 th time slice, which may be, for example, 00-00. For a _0,0 ,……,a _m,0 It can indicate whether the 0 th image acquisition device respectively acquires the face image under the 0 th-nth time slices; if yes, the corresponding value is 1; if not, the corresponding value is 0.

Optionally, in order to maintain the time continuity, the face spatiotemporal relationship matrix corresponding to the obtained same face image motion trajectory is expanded according to the spatial dimension (expanded according to the column), and a one-dimensional face spatiotemporal relationship trajectory X = (a) can be obtained _0,0 …a _m,0 )…(a _0,n …a _m,n )。

According to the optional embodiment, the face space-time relationship matrix is constructed according to the incidence relation between each identical face image motion sub-track and at least one image acquisition device in the preset area range, so that the accurate construction of the space-time relationship matrix of the face image motion tracks in time and space dimensions is realized, and the construction accuracy of the image code matrix between the face image and the IMSI in the follow-up process is improved.

S240, constructing an IMSI space-time relation matrix according to the same IMSI motion trajectory.

For example, for any same IMSI movement trajectory, the corresponding IMSI spatio-temporal relationship matrix may be determined. The time discretization and the space unification can be carried out on the same IMSI motion track, so that a space-time relation matrix on a time dimension and a space dimension is constructed.

In an optional embodiment, constructing an IMSI spatio-temporal relationship matrix according to the same IMSI movement trajectory includes: according to the preset time slice length, performing track segmentation on the same IMSI movement track to obtain at least one same IMSI movement sub-track; establishing an equipment mapping relation between each code detecting equipment and each image acquisition equipment in a preset area range; the code detection equipment is used for collecting IMSI data; and constructing an IMSI space-time relation matrix according to the sub-tracks of the same IMSI motion and the equipment mapping relation.

Illustratively, if the same IMSI movement locus B is { (x) ₁ ,y ₁ ,t ₁ ,d ₁ ),(x ₂ ,y ₂ ,t ₂ ,d ₂ ),…,(x _n ,y _n ,t _n ,d _n ) Slicing the same IMSI motion trajectory B according to a preset time slice length alpha to obtain at least one same IMSI motion sub-trajectory, which is respectively the same IMSI motion sub-trajectory B1{ (x) ₁ ,y ₁ ,t ₁ ,d ₁ )…,(x _k ,y _k ,t _k ,d _k ) And the same IMSI kinematic sub-track B2{ (x) _k+1 ,y _k+1 ,t _k+1 ,d _k+1 )…,(x _m ,y _m ,t _m ,d _m ) }, \8230 { \\ 8230; }, the same IMSI motron trajectory Bn { (x) _m+1 ,y _m+1 ,t _m+1 ,d _m+1 )…,(x _n ,y _n ,t _n ,d _n ) }. Wherein the same IMSI movesThe sub-tracks B1, B2, \ 8230, bn may be identical or different in number of elements, but of the same temporal length, all being alpha. For example, if α is 10min, the IMSI data of the IMSI obtained within 10 time periods of 00; the elements in the same IMSI movement sub-trajectory B2 may be IMSI data of the IMSI obtained within a time period of 00.

And establishing an equipment mapping relation between each code detecting equipment and each image acquisition equipment in a preset area range. It should be noted that the code detection device is acquired through electronic signals, and has the characteristics of large coverage and low missing rate. The intelligent face camera equipment for acquiring the face image data captures images through optical information and has the characteristics of accurate geographic position and high missing mining rate. The method comprises the steps of establishing a device mapping relation between each code detecting device and each image acquisition device in a preset area range in a mode of setting a spatial distance threshold value beta, mapping the code detecting devices to intelligent face camera devices, establishing a spatial mapping table, and achieving spatial unification. Where beta may be set to 400 meters.

Illustratively, if the code detection device A, the code detection device B and the code detection device C exist in the preset area range. There are image capturing device a, image capturing device B, image capturing device C, image capturing device D, and image capturing device E. And establishing a mapping relation between each code detecting device and the image acquisition device according to a preset spatial distance threshold. Specifically, if the image capturing apparatus B and the image capturing apparatus C are located within the range centered on the image capturing apparatus a and having a distance radius of 400 meters. Then a mapping relationship is established between the code detection device A and the image acquisition device B and the image acquisition device C. Similarly, the mapping relationship between the image acquisition device and the code detection device B and the image acquisition device C is established for the code detection device B and the code detection device C, which is not described in detail in this embodiment.

At least one code detecting device in the range of the preset area is determined. And constructing an IMSI space-time relation matrix according to the sub-track and the equipment mapping relation of each identical IMSI motion. Illustratively, if n image capturing devices exist within the preset area. The number of sub-tracks corresponding to the same IMSI movement track is m, that is, the sub-tracks correspond to m groups of time slices, and the IMSI space-time relationship matrix B of the IMSI movement track B can be represented by the following matrix.

Wherein, b _m,n It can be shown whether the code detecting device corresponding to the nth image capturing device captures the face image in the time period corresponding to the sub-track m of the same IMSI movement track. In order to establish the association relationship between the code detecting equipment and the image acquisition equipment. The values in the IMSI spatio-temporal relationship matrix are determined in the following manner.

If 4 image acquisition devices A, B, C and D exist and 3 code detection devices A, B and C aim at B _0,0 ,……,b _0,3 Denotes the values in different spatial dimensions in the case of the same time dimension, for example in the time range 00-00. For the same IMSI movement trace B, the subscripts 0, 1, 2, 3 denote 4 image capturing devices. For example, group 0:00-00, time period 10, code detection device a, group 1: 00-00, under a 10 time period, a code detecting device B in group 2, 00-00. Assume that the code detecting device a corresponds to the image collecting device a, the code detecting device B corresponds to the image collecting device B and the image collecting device C, and the code detecting device C corresponds to the image collecting device D. Then, in a time period of 00 _0,0 To 1, if the spying code device B collects IMSI data of the IMSI in time period 00 _0,1 And b _0,2 Is 1. Because, the code detecting device B corresponds to the image capturing device B and the image capturing device C. If the code detecting equipment C does not acquire the IMSI data of the IMSI, b _0,3 Is 0. For b _0,0 ,……,b _3,0 Indicating the values in different time dimensions under the same spatial dimension, for example, in the following time periods of 00-00IMSI data. If yes, the value is 1, if no, the value is 0.

Optionally, in order to keep the time continuity, the IMSI spatiotemporal relationship matrix corresponding to the obtained same IMSI movement trajectory is expanded according to the spatial dimension (expanded according to the column), and a one-dimensional IMSI spatiotemporal relationship trajectory Y = (b) can be obtained _0,0 …b _m,0 )…(b _0,n …b _m,n )。

According to the optional embodiment, the IMSI space-time relationship matrix is constructed according to the sub-track of each identical IMSI motion and the equipment mapping relation, so that the accurate construction of the space-time relationship matrix of the IMSI motion track in the time and space dimensions is realized, and the subsequent construction accuracy of the image code matrix between the face image and the IMSI is improved.

And S250, constructing an image code matrix between the face image and the IMSI according to the face space-time relation matrix and the IMSI space-time relation matrix.

Illustratively, the obtained at least one face spatiotemporal relationship matrix and IMSI spatiotemporal relationship matrix may be combined to obtain an image code matrix between the face image and the IMSI. Specifically, at least one human face space-time relationship track X = (a) _0,0 …a _m,0 )…(a _0,n …a _m,n ) And at least one IMS spatio-temporal relationship trajectory Y = (b) _0,0 …b _m,0 )…(b _0,n …b _m,n ) Combining to obtain a graph code matrix B = (X) between the face image and the IMSI ₁ ,…,X _g ,Y ₁ ,…,Y _n ). Wherein, g represents the number of the motion tracks of the face image, and n represents the number of the motion tracks of the IMSI.

And S260, carrying out singular value decomposition on the graph code matrix to obtain a singular value corresponding to the graph code matrix.

It should be noted that, because only the values of 0 and 1 are found in the obtained graph code matrix, 1 represents that the face image or IMSI appears sometime, and 0 represents unknown or not appears. The problem of the stepwise nature of the image code matrix is prominent, and the value of 0 is most dominant, i.e. there is a problem of matrix sparsity. Therefore, the influence of the step property and the matrix sparsity on the correlation accuracy of the graph code track needs to be reduced.

Exemplary ofIf the graph code matrix is B, B = (a) _i,j ) _n*m By transposing the matrix, E = BB ^T 、H＝B ^T B _n×n . Wherein, B ^T Is the transpose of the graph code matrix B. Performing feature decomposition processing on the matrix E and the matrix H to obtain E = PΛ ₁ Ρ ^T 、H＝ZΛ ₂ Z ^T . Wherein, pp, Z are the eigenvectors corresponding to matrix E, H, Λ ₁ 、Λ ₂ Is a diagonal matrix and the non-zero elements on the diagonal are all the same. Let the eigenvalue be { σ ₁ ,σ ₂ ,…σ _k And k is less than or equal to m and k is less than or equal to n, the singular value of a graph code matrix BETA can be obtained:

and k is the number of eigenvalues and/or singular values of the graph code matrix B. Singular value decomposition is carried out on the graph code matrix BETA to obtain: beta = Λ Z ^T Where p is the left singular matrix, Z is the right singular matrix, Λ is the diagonal matrix, the diagonal elements are represented by singular values

Is prepared from the following components. Developing to obtain BETA = lambda ₁ μ ₁ ν ₁ +λ ₂ μ ₂ ν ₂ +…+λ _κ μ _κ ν _κ μ is the left singular column vector and ν is the right singular column vector.

And S270, determining the number of column vectors of the image code matrix.

And S280, determining the number of the dimensionality reduction singular values according to the number of the column vectors and a preset element threshold value.

It will be appreciated that the map matrix beta is sufficiently sparse that the largest first singular values, and corresponding left and right singular column vectors, beta = lambda, can be retained ₁ μ ₁ ν ₁ +λ ₂ μ ₂ ν ₂ +…+λ _ζ μ _ζ ν _ζ . According to theory and a large number of experiments, it is shown that near ζ = K ÷ C, the combined effect is optimal. In the invention, we refer to the matrix at this time as a singular value map code matrix. Wherein, the value of C can be preset by related technicians according to actual requirements. For example, C =500. Wherein, K is the number of column vectors of the image code matrix. Zeta is the number of dimensionality reduction singular values.

S290, according to the number of the dimensionality reduction singular values and the singular values, dimensionality reduction processing is carried out on the graph code matrix to obtain a singular value graph code matrix.

And performing dimension reduction processing on the graph code matrix according to the determined dimension reduction singular value quantity and the graph code matrix expansion items obtained according to the singular values to obtain a singular value graph code matrix.

S2100, determining a reference number sequence according to the number of the dimensionality reduction singular values.

Illustratively, to map the trace information of a person to a value at a time, the trace has a non-zero value for each time period. Therefore, the reference number series χ is preset ₀ ，χ ₀ ＝{1,…,1}，χ ₀ Is equal to the size of the number of reduced-dimension singular values.

And S2110, determining the track association degree of the graph code according to the singular value graph code matrix and the reference number sequence.

For example, the graph code association confidence of the singular value graph code matrix may be calculated through gray association, and the graph code association confidence may be used as the graph code track association degree.

In an optional embodiment, determining the graph code track association degree according to the singular value graph code matrix and the reference number sequence comprises: determining a difference sequence between the singular value graph code matrix and the reference sequence; determining a graph code track association coefficient according to the difference sequence; and determining the association degree of the image code track according to the association coefficient of the image code track.

Wherein, the difference array between the singular value image code matrix and the reference array can be determined by adopting the following mode:

η _i (κ)＝|χ ₀ (κ)-χ _i (κ)|；

where i is the number of rows in the singular value map code matrix and κ is the position of the element in the column vector.

For example, the graph code track association coefficient may be determined as follows:

wherein the correlation coefficient ζ _i (κ)∈[0～1]The coefficient ρ is a parameter for controlling the degree of distinction, and the smaller ρ is, the larger the degree of distinction is, which may be preset by a relevant technician. Rho is an element of [0 to 1 ]]For example, ρ =0.3.

For example, the graph code track association degree may be determined as follows:

wherein the content of the first and second substances,

is the number of the motion trajectories of the face,

is the number of IMSI movement trajectories.

According to the scheme of the embodiment, the image code matrix between the face image and the IMSI is constructed according to the face spatiotemporal relationship matrix and the IMSI spatiotemporal relationship matrix, so that the accurate construction of the image code matrix is realized. The dimension reduction processing is carried out on the graph code matrix according to the dimension reduction singular value quantity and the singular value to obtain the singular value graph code matrix, so that the dimension reduction processing of the graph code matrix is realized, the subsequent calculation amount is reduced, and the determination efficiency is improved. The method and the device realize accurate determination of the track association degree of the graph code by determining the reference number sequence according to the number of the dimensionality reduction singular values and according to the singular value graph code matrix and the reference number sequence. According to the scheme, the real-time determination of the association degree of the image code track is realized by processing the acquired face image data and IMSI data in real time and generating and updating the corresponding image code matrix in real time. And a large-scale network model is not required to be trained based on a large number of face images and IMSI data, and the complexity of determining the association degree of the image code track is reduced. Compared with a mode of determining the graph code track association degree by considering that modeling completely depends on human experience based on prior rules, the method and the device for determining the graph code track association degree reduce the graph code association complexity and improve the accuracy of graph code association.

EXAMPLE III

Fig. 3 is a schematic structural diagram of a device for determining association degree of a graph code track according to a third embodiment of the present invention. The device for determining the association degree of the image code track provided by the embodiment of the invention can be suitable for the situation of performing image code association on face images and IMSI (international mobile subscriber identity) belonging to the same user, and can be realized in a hardware and/or software mode, as shown in fig. 3, the device specifically comprises: the system comprises a data acquisition module 301, a motion track determination module 302, a graph code matrix construction module 303, a singular value graph code matrix determination module 304 and a track relevancy determination module 305. Wherein the content of the first and second substances,

the data acquisition module 301 is configured to acquire face image data and international mobile subscriber identity IMSI data within a preset area range;

a motion trajectory determining module 302, configured to cluster the face image data and the IMSI data to obtain a same face image motion trajectory and a same IMSI motion trajectory respectively;

the graph code matrix constructing module 303 is configured to construct a graph code matrix between the face image and the IMSI according to the same face image motion trajectory and the same IMSI motion trajectory;

a singular value map code matrix determination module 304, configured to perform singular value decomposition and dimension reduction on the map code matrix to obtain a singular value map code matrix;

a track association degree determining module 305, configured to determine a graph code track association degree according to the singular value graph code matrix, and configured to generate motion track information.

The embodiment of the invention obtains the face image data and the international mobile subscriber identity IMSI data in the preset area range; clustering the facial image data and the IMSI data to respectively obtain the same facial image motion track and the same IMSI motion track; constructing an image code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track; singular value decomposition and dimension reduction processing are carried out on the graph code matrix to obtain a singular value graph code matrix; and determining the track association degree of the graph code according to the singular value graph code matrix for generating motion track information. According to the scheme, the real-time determination of the association degree of the image code track is realized by processing the acquired face image data and IMSI data in real time and generating and updating the corresponding image code matrix in real time. And a large-scale network model does not need to be trained based on a large number of face images and IMSI data, so that the complexity of determining the association degree of the image code track is reduced. Compared with a mode of determining the graph code track association degree by considering that modeling completely depends on human experience based on prior rules, the method and the device for determining the graph code track association degree reduce the graph code association complexity and improve the accuracy of the graph code association.

Optionally, the graph code matrix constructing module 303 includes:

the human face space-time relationship matrix construction unit is used for constructing a human face space-time relationship matrix according to the same human face image motion trail;

the IMSI space-time relationship matrix construction unit is used for constructing an IMSI space-time relationship matrix according to the same IMSI motion trajectory;

and the image code matrix constructing unit is used for constructing an image code matrix between the face image and the IMSI according to the face spatiotemporal relationship matrix and the IMSI spatiotemporal relationship matrix.

Optionally, the face spatiotemporal relationship matrix constructing unit includes:

the image motion sub-track determining subunit is used for carrying out track segmentation on the same face image motion track according to a preset time slice length to obtain at least one same face image motion sub-track;

the human face space-time relationship matrix construction subunit is used for constructing a human face space-time relationship matrix according to the incidence relationship between each identical human face image motion sub-track and at least one image acquisition device in the preset region range; the image acquisition equipment is used for acquiring face image data.

Optionally, the IMSI space-time relationship matrix constructing unit includes:

the IMSI sub-track determining subunit is used for carrying out track segmentation on the same IMSI motion track according to a preset time slice length to obtain at least one same IMSI motion sub-track;

the device mapping relation establishing subunit is used for establishing a device mapping relation between each code detection device and each image acquisition device in the preset area range; the code detection equipment is used for collecting IMSI data;

and the IMSI space-time relation matrix constructing subunit is used for constructing an IMSI space-time relation matrix according to each identical IMSI movement sub-track and the equipment mapping relation.

Optionally, the singular value map code matrix determining module 304 includes:

the singular value determining unit is used for carrying out singular value decomposition on the graph code matrix to obtain a singular value corresponding to the graph code matrix;

a column vector quantity determining unit for determining the column vector quantity of the image code matrix;

the dimensionality reduction singular value quantity determining unit is used for determining the dimensionality reduction singular value quantity according to the column vector quantity and a preset element threshold value;

and the singular value image code matrix determining unit is used for performing dimension reduction processing on the image code matrix according to the dimension reduction singular value quantity and the singular value to obtain a singular value image code matrix.

Optionally, the track association degree determining module 305 includes:

a reference number sequence determining unit, configured to determine a reference number sequence according to the number of the dimensionality reduction singular values;

and the graph code track association degree determining unit is used for determining the graph code track association degree according to the singular value graph code matrix and the reference number sequence.

Optionally, the graph code track association degree determining unit includes:

a difference sequence determining subunit, configured to determine a difference sequence between the singular value graph code matrix and the reference sequence;

the correlation coefficient determining subunit is used for determining the graph code track correlation coefficient according to the difference sequence;

and the graph code track association degree determining subunit is used for determining the graph code track association degree according to the graph code track association coefficient.

The device for determining the association degree of the graph code track, provided by the embodiment of the invention, can execute the method for determining the association degree of the graph code track, provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.

Example 4

FIG. 4 illustrates a schematic diagram of an electronic device 40 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 4, the electronic device 40 includes at least one processor 41, and a memory communicatively connected to the at least one processor 41, such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, and the like, wherein the memory stores a computer program executable by the at least one processor, and the processor 41 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM) 42 or the computer program loaded from a storage unit 48 into the Random Access Memory (RAM) 43. In the RAM 43, various programs and data necessary for the operation of the electronic apparatus 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to the bus 44.

A number of components in the electronic device 40 are connected to the I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, modem, wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.

Processor 41 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, or the like. Processor 41 performs the various methods and processes described above, such as the graph code trajectory relevance determination method.

In some embodiments, the graph code trajectory relevance determination method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 40 via the ROM 42 and/or the communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the graph code trajectory relevance determination method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the graph code trajectory relevance determination method in any other suitable manner (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be performed. A computer program can execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service are overcome.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present invention may be executed in parallel, sequentially, or in different orders, and are not limited herein as long as the desired results of the technical solution of the present invention can be achieved.

The above-described embodiments should not be construed as limiting the scope of the invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining the association degree of a graph code track is characterized by comprising the following steps:

constructing a graph code matrix between the face image and the IMSI according to the same face image motion track and the same IMSI motion track;

2. The method of claim 1, wherein constructing a graph code matrix between the face image and the IMSI according to the same face image motion trajectory and the same IMSI motion trajectory comprises:

constructing a human face space-time relationship matrix according to the same human face image motion track;

constructing an IMSI space-time relation matrix according to the same IMSI motion trajectory;

and constructing an image code matrix between the face image and the IMSI according to the face space-time relationship matrix and the IMSI space-time relationship matrix.

3. The method according to claim 2, wherein the constructing a human face spatiotemporal relationship matrix according to the same human face image motion trajectory comprises:

performing track segmentation on the same face image motion track according to a preset time slice length to obtain at least one same face image motion sub-track;

constructing a human face space-time relationship matrix according to the incidence relationship between each same human face image motion sub-track and at least one image acquisition device in the preset area range; the image acquisition equipment is used for acquiring face image data.

4. The method of claim 2, wherein constructing an IMSI spatiotemporal relationship matrix based on the same IMSI movement trajectory comprises:

according to a preset time slice length, performing track segmentation on the same IMSI movement track to obtain at least one same IMSI movement sub-track;

establishing an equipment mapping relation between each code detection equipment and each image acquisition equipment in the preset area range; the code detection equipment is used for collecting IMSI data;

and constructing an IMSI space-time relation matrix according to the same IMSI movement sub-track and the equipment mapping relation.

5. The method according to any one of claims 1 to 4, wherein the performing singular value decomposition and dimension reduction on the graph code matrix to obtain a singular value graph code matrix comprises:

singular value decomposition is carried out on the graph code matrix to obtain singular values corresponding to the graph code matrix;

determining the number of column vectors of the image code matrix;

determining the number of dimensionality reduction singular values according to the number of the column vectors and a preset element threshold;

and performing dimension reduction processing on the graph code matrix according to the dimension reduction singular value quantity and the singular value to obtain a singular value graph code matrix.

6. The method of claim 5, wherein determining a graph code trajectory correlation based on the singular value graph code matrix comprises:

determining a reference number sequence according to the number of the dimensionality reduction singular values;

and determining the track association degree of the graph code according to the singular value graph code matrix and the reference number sequence.

7. The method of claim 6, wherein determining the graph code track association according to the singular value graph code matrix and the reference number sequence comprises:

determining a difference sequence between the singular value graph code matrix and the reference sequence;

determining a track association coefficient of the image code according to the difference sequence;

and determining the association degree of the image code track according to the association coefficient of the image code track.

8. An apparatus for determining a track association degree of a graph code, comprising:

the motion track determining module is used for clustering the facial image data and the IMSI data to respectively obtain the same facial image motion track and the same IMSI motion track;

9. An electronic device, characterized in that the electronic device comprises:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,

the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to enable the at least one processor to perform the method of determining a correlation of a graph code trajectory according to any one of claims 1 to 7.

10. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions for causing a processor to implement the graph code track association degree determination method according to any one of claims 1-7 when executed.