WO2022245240A1

WO2022245240A1 - Method, system and device for identifying people to be searched upon passing through a metal detector

Info

Publication number: WO2022245240A1
Application number: PCT/RU2021/000222
Authority: WO
Inventors: Александр Алексеевич КНЯЗЕВ
Original assignee: Публичное Акционерное Общество "Сбербанк России"
Priority date: 2021-05-20
Filing date: 2021-05-27
Publication date: 2022-11-24

Abstract

The present technical solution relates to the field of surveillance systems used at checkpoints where people are examined for the presence of prohibited items or with respect to possible unlawful activity. The technical result is that of increasing the accuracy with which people are identified for additional checking by using a personal profile generated on the basis of a mass distribution model formed using historical data from a person's passages through a metal detector. This result is achieved using a computer-implemented method for identifying people to be searched upon passing through a metal detector, which is carried out with the aid of at least one processor and comprises the steps of: generating a personal profile containing a unique ID and a mass distribution model (MDM) containing at least data about a type of metal and the mass and position of the given type of metal on the person, said MDM being formed on the basis of historical information from passages made by the person in question through a metal detector; obtaining the ID of the person and the data of the current MDM for said person, formed upon passage through a metal detector; comparing the current MDM with the reference MDM to identify any deviation of the mass, type and position of a given type of metal on the person from the reference MDM, wherein a signal to search the person is generated if a deviation is found in at least one of each of said parameters of the MDM.

Description

METHOD, SYSTEM AND DEVICE FOR IDENTIFYING PEOPLE FOR SCREENING WHEN PASSING MET ALLODETECTOR

FIELD OF TECHNOLOGY

[0001] This technical solution relates to the field of control systems that are used in the areas of inspection of people for the presence of prohibited items or the fact of committing illegal acts.

BACKGROUND OF THE INVENTION

[0002] The metal detectors used today, as a rule, are not guided by the individual habits of a person and are designed to inform the presence of metal objects on a person. Stationary metal detectors are capable of detecting metal objects, determining the mass, type and location of metal-containing objects. Detection occurs according to the specified settings - to the mass. As a result, this leads to the fact that security officers are forced to inspect each person individually, thereby significantly reducing the throughput of the checkpoint (inspection).

[0003] A solution is known from the prior art, which is a system for detecting the smuggling of prohibited items using a metal detector (application WO 2020025965 A1, Keeny et al., 06.02.2020). In the well-known solution, retrospective information is used, which forms magnetic signatures, upon several passes by a person of the metal detector frame. Based on the signatures, a person's profile is formed, which is associated with an external identifier obtained from the ACS and/or the biometric recognition system.

[0004] The disadvantages of the known solution and existing approaches are the limited functionality and accuracy of determining possible anomalies during the passage of a metal detector, as well as in the case of using metal detectors of different manufacturers and / or types in the system, which is due to the lack of application of the metal distribution model (MRM) on the human body in the use of historical data of his passage through the metal detector, and the use of this information to form a profile of a person for subsequent analysis. SUMMARY OF THE INVENTION

[0005] The technical problem to be solved is the creation of a new approach in the application of metal detector readings, which makes it possible to form a person's profile for subsequent analysis for the need for additional control. [0006] The technical result is to increase the accuracy of identifying people for additional control, through the use of a person's profile, formed on the basis of the MRM, formed according to retrospective data on the passage of a person.

[0007] The claimed result is achieved by a computer-implemented method for determining people for inspection when passing through a metal detector, performed using at least one processor and comprising the steps of:

- forming a profile of a person, containing a unique ID and a mass distribution model (MDM), containing at least data on the type of metal, mass and location of a certain type of metal located on the person, and the MRM is formed based on the historical information of the passes of the metal detector by this person;

- receive the ID of the person and the data of the current MRM for him, formed when passing through the metal detector;

- comparison of the current MRM with the base MRM is carried out, in which the deviation of the mass, type and location of a certain type of metal on a person from the base MRM is analyzed, and if a deviation is detected in at least one of each of the specified MRM parameters, a signal is generated for screening the person.

[0008] In one particular embodiment of the method, the type of metal is a non-ferrous and/or ferrous metal.

[0009] In another particular embodiment of the method, the location of the metal on a person is characterized by spatial coordinates.

[0010] In another particular example of the implementation of the method, the relative position of the non-ferrous metal to black is characterized by spatial coordinates or vectors.

[0011] In another particular example of the implementation of the MPM method is represented as a graph reflecting the totality of the set of connections of all masses located on a person. [0012] In another particular embodiment of the method, the location of the metal is further characterized by the ratio of the mass of non-ferrous metal to black.

[0013] In another particular embodiment of the method, the basic MRM is formed based on a plurality of readings of electromagnetic fields obtained when a person passes a metal detector.

[0014] In another particular embodiment of the method, the correlations of each electromagnetic field are taken into account.

[0015] In another particular embodiment of the method, a person's ID is transmitted from an access control system.

[0016] In another particular embodiment of the method, the access control system recognizes a person by analyzing his image.

[0017] In another particular embodiment of the method, the access control system performs person recognition using an access tool.

[0018] The claimed technical result is also achieved by a device for determining people for inspection when passing through a metal detector, containing at least one processor and at least one memory means containing machine-readable instructions that, when executed by the processor, carry out the above method.

[0019] The claimed technical result is achieved by implementing a system for identifying people for inspection, containing a computing device connected to a metal detector, in which a person profile is formed using a computing device containing a unique ID and a mass distribution model (MRM) containing at least data on type of metal, mass and location of a certain type of metal carried by a person, and the MRM is formed on the basis of historical information about the passages of the metal detector by this person; determine the ID of the person; using a metal detector, data is obtained from the current MRM of a person; using a computing device, the current MRM is compared with the base MRM, in which the deviation of the mass, type and location of a certain type of metal on a person from the base MRM is analyzed, and in case of detection of a deviation for each of the specified parameters, the MRM generates a signal for screening a person.

[0019] In one particular embodiment of the system, the metal detector is multizone.

[0020] In another particular embodiment of the system, the metal type is non-ferrous and/or ferrous metal.

[0021] In another particular embodiment of the system, the location of metal on a person is characterized by spatial coordinates.

[0022] In another particular embodiment of the system, the relative position of the non-ferrous metal to the ferrous metal is characterized by spatial coordinates and/or vectors.

[0023] In another particular embodiment of the system, the location of the metal is further characterized by the ratio of the mass of non-ferrous metal to black.

In another particular example of the implementation of the MRM system is a graph that reflects the distances between the masses, the relative spatial arrangement (azimuth), masses, metal type, the ratio between the masses, the relationship between the types of metal masses on a person.

[0024] In another particular embodiment of the system, the base MRM is generated based on a plurality of readings of electromagnetic fields obtained when a person passes a metal detector.

[0025] In another particular embodiment of the system, the correlations of each electromagnetic field are taken into account.

[0026] In another particular embodiment of the system, a person's ID is transmitted from an access control system.

[0027] In another particular embodiment of the system, the access control system performs recognition of a person by analyzing his image.

[0028] In another particular embodiment of the system, the access control system performs person recognition using an access tool.

[0029] Other particular aspects of the present technical solution will be disclosed later in the detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

[0030] FIG. 1 illustrates the general view of the claimed solution.

[0031] FIG. 2A illustrates an example of a metal detector zone distribution.

[0032] FIG. 2B illustrates an example of metal detection on a human body. [0033] FIG. ZA - ZV illustrate an example of the formation of MRM in retrospective terms.

[0034] FIG. 4 illustrates an example of the formation of a single profile of a person with a general MRM.

[0035] FIG. 5 illustrates a block diagram of the claimed method.

[0036] FIG. 6 illustrates a general view of the computing device.

IMPLEMENTATION OF THE INVENTION

[0037] As shown in FIG. 1, the claimed solution is a distributed system (100) united by a data transmission network, which includes an access control system (110), a metal detector (120) and a computing device (130). Additionally, the system (100) may include one or more video fixation means (111), for example, surveillance cameras.

[0038] The access control system or ACS (software) can be any known means that provides primary control of people (10), for example, a turnstile activated using a key card, biometric information, RFID tag can be used as a system (110). etc. Cameras (111) associated with the access control system (110) provide images of people (10) in the control area for their subsequent recognition using a computing device (130). Cameras (111) may be IP, PTZ cameras, webcams, or any other similar device that provides data capture.

[0039] The computing device (130) is a computing tool, such as a server, personal computer, laptop, tablet, or smartphone. The device (130) provides the computational operations required within the framework of the claimed solution, in particular, data processing for recognizing people (10), forming and analyzing their profile (20), obtaining data from a metal detector (120), forming an MRM (210) and other operations that may be performed to implement the proper functionality of the system (100).

[0040] A local area network (LAN) or wireless communication means, such as a WLAN network, can be used as a data network for communication of system elements (100). The data communication network may also be an Intranet or the Internet deployed in the area of control, allowing data to be transferred between the elements of the system (100).

[0041] A unique identifier (ID) of a person (10) is generated and stored on a computing device (130), it can also be duplicated in any associated with the device (130) storage, for example, a corporate cloud server, etc. Profile (20) contains data identifying a person, for example, full name, position, photograph, as well as MPM (210), which reflects a retrospective picture of metal detector (120) triggers. The MRM (210) is executed as a mass matrix and updates its values with each new passage by the person (10) of the control area. Mass can be expressed not only in absolute values, but also in conditional ones, for example, in terms of the electromagnetic reflection coefficient, etc. Also, the profile (20) may contain any other additional information that allows additional identification of people (10).

[0042] The image received from the cameras (111) is compared with the photo profiles (20) to determine the desired profile and its ID. After that, the current MPM is compared with the MPM (210) stored in the profile (20). This process will be described in more detail in these application materials.

[0043] In FIG. 2A shows an example of the distribution of zones (121) for fixing a metal detector (120), in particular, a multi-zone metal detector, however, the present solution can also be implemented using metal detectors (120) of the usual type. The presented fixation zones (121) allow fixing radiation from metal objects, taking into account the coordinates of their spatial distribution, thereby accurately determining the necessary data on the type of object, determining its mass and type of metal, at one point or another on the human body (10).

[0044] An example of analysis of various items is shown in FIG. 2B. When metal detection is performed by the detector (120), various objects (122-124) are detected on the human body (10). When determining each type of object (122-124), recognition of its mass, type of metal (color or black), coordinates of location on the human body (10) is carried out. The determination occurs for each of the objects (122-124) and is reduced to a single MRM (210), which is transferred to the person's profile (20) stored on the device (130).

[0045] In FIG. 3A-3B presents a retrospective analysis of the passage of a person (10) metal detector (120) and obtaining for each pass information reflecting the MRM at a given point in time. As a result of each fixation and recognition of the type, mass, coordinates and ratio of the type of metal in each pass by the same person (10), a corresponding MPM is created (201, 202, 203), which is tied at the time of its creation to the formed profile (20). Each MPM (201-203) is used at the moment when a person (10) passes through a metal detector (120) to analyze possible anomalies, indicating the carrying of prohibited items and the need for additional control, for example, search of a person (10).

[0046] As a result of the formation of the MPM (201-203) with each passage of the control zone, a general MPM (210) is created, shown in FIG. 4, which is fixed in the profile (20) and used for subsequent analysis of the person (10) when passing through the metal detector (120).

[0047] In FIG. 5 shows a flowchart of the claimed method (300) for identifying people for additional control. At step (301), when a person (10) appears in the control area, using the control system (110), information is obtained to identify the profile ID (20), for example, using surveillance cameras (111) or reading access devices (smart card, pass, etc.).

[0048] When identifying the profile (20) of a person using a computing device (130), further waiting for information from the metal detector (120) is carried out. At step (302), when passing through the metal detector (120), the current MRM is recorded, showing the type, mass of metal and its location on the person (10). The generated MPM is transmitted to the device (130) for analysis of the deviation from the base MPM (210) stored in the profile (20) performed at step (303).

[0049] MRM analysis can be performed using computational processing of incoming data, for example, using a machine learning model based on an artificial neural network (ANN), which is trained on various data that form MRM for different types of situations, in particular, telemetry from different types of situations. metal detector (120) and validation on a sample with confirmation of the fact that people were carrying prohibited items.

[0050] The analysis at step (303) is performed to compare the readings of the current MPM when the person (10) passes through the metal detector (120) with the base MPM (210) of the user profile (20). The comparison is carried out to identify deviations in the mass, type of metal and its location on the human body (10) at a given point in time. When analyzing the deviation between the current and baseline, MRM (210) takes into account a threshold value that captures correlations and possible discrepancies in MRM that do not indicate the presence of anomalies or the need to trigger the system for additional control.

[0051 ] MRM correlations at different time points take into account the data recorded by the metal detector (120) for the same person (10), taking into account the passages performed by him and different readings of the data forming the MRM, in particular, different weight, type of metal, its location. As a rule, a relatively single MRM pattern is formed for each person during his daily life, for example, when passing control at the place of work, or other places for which there is a formed MRM profile for the current location of this person (10). The generated profile (20) may contain average MRM data with reference to the corresponding places visited by a person (10), for example, an airport, government agencies, etc. As a rule, a person (10) has a standardized set of items and accommodation when visiting certain places.

[0052] As a result of the operation of step (303), at step (304), the computing device (130) decides, based on the performed analysis, whether additional screening of a person is necessary (step 305), or that the current MRM does not deviate from the average value and additional screening is not required (step 306).

[0053] Returning to the circuit shown in FIG. 1, it should be noted that a person (10) in the control zone can first be checked through a metal detector (120) and only then through an access control system (OS), for example, a turnstile. Primarily, the ID and profile (20 of the person in this case are also determined using a means installed in the control area, for example, a camera (111), which acquires an image of a person (10) and transmits its data for determination to the device (130), which then waiting to receive the current MPM from the metal detector (120).

[0054] Moreover, in this implementation, in case of successful passing of control, a person (10) can automatically be passed through the access control system (110), or vice versa, access can be blocked if the check is unsuccessful (step 305).

[0055] The claimed solution can also be used to detect theft or unauthorized removal of items, such as equipment. This fact is worked out when a person (10) passes through the metal detector (120) several times a day in one control area, for example, during the working day. When step (304) is performed, an MRM analysis is performed, in which, in the event of an increase in the mass of metal when a person (10) exits above a predetermined threshold value, the fact of an event subject to additional control is recorded.

[0056] When the system is triggered in the alarm event mode, i.e. if additional control is necessary, using the device (130), the type of object can be additionally recorded based on the data contained in the MRM, while, if a threat is detected in terms of the presence of a weapon or other means defeat, a signal is generated that arrives at the operational communications console of the police and / or special services. Additionally, a signal can be generated using warning tools, blocking the control zone, activating threat neutralization tools in the control area, and other actions that can be implemented by automated means that provide triggering based on an alarm signal generated by one or more devices of the system (100). MRM (210) can be represented by a graph, the vertices of which determine the spatial coordinates, mass and type of metal, and the edges (bonds) determine the distance between the masses, the ratio of the masses of the metal, the ratio of metal types, the mutual spatial orientation (azimuth).

[0057] MPM analysis can be performed using computational processing of incoming data, for example, using graph comparison performed by a machine learning model that is trained on various data that form MPM for different types of situations, in particular, telemetry from various types of metal detector (120 ) and validation on a sample with confirmation of the fact that people were carrying prohibited items. The graph representation is invariant with respect to the type of metal detector (20) and even to the location of a person (10) inside the same metal detector (20). [0058] When performing step (303), in the case of representing the MRM in the form of graphs, the current graph obtained at the time of the passage of a person (10) is compared with the base graph, in which the deviation of the connections is analyzed, namely: the distances between the masses, the relative spatial arrangement (azimuth), mass, type of metal, ratio between masses, relationship between types of masses on a person from the connections of the base graph, and if a deviation is detected for each of the indicated parameters of the graph, a signal is generated for the inspection of a person (10).

[0059] In FIG. 6 is a perspective view of a computing device (400) suitable for performing the method (300) described above. The device (400) may be, in particular, a computing device (130), a server (302) and other non-represented devices that may participate in the overall architecture of the claimed solution.

[0060] In general, the computing device (400) comprises one or more processors (401), memory means such as RAM (402) and ROM (403), input/output interfaces (404), devices input/output (405), and a device for networking (406).

[0061] The processor (401) (or multiple processors, multi-core processor, graphics processor) can be selected from a range of devices, widely currently in use, such as Intel™, AMD™, Apple™, Samsung Exynos™, MediaTEK™, Qualcomm Snapdragon™, Nvidia™, ARM™, etc.

[0062] RAM (402) is a random access memory and is designed to store machine-readable instructions executable by the processor (401) to perform the necessary data logical processing operations. The RAM (402) typically contains the executable instructions of the operating system and associated software components (applications, program modules, etc.).

[0063] A ROM (403) is one or more persistent storage devices such as a hard disk drive (HDD), a solid state drive (SSD), flash memory (EEPROM, NAND, etc.), optical storage media ( CD-R/RW, DVD-R/RW, BlueRay Disc, MD), etc.

[0064] Various types of I/O interfaces (404) are used to organize the operation of device components (400) and organize the operation of external connected devices. The choice of appropriate interfaces depends on the particular design of the computing device, which can be, but not limited to: PCI, AGP, PS/2, IrDa, FireWire, LPT, COM, SATA, IDE, Lightning, USB (2.0, 3.0, 3.1, micro, mini, type C), TRS/Audio jack (2.5, 3.5, 6.35), HDMI, DVI, VGA, Display Port, RJ45, RS232, etc. [0065] To ensure user interaction with the computing device (400), various means (405) of I/O information are used, for example, a keyboard, a display (monitor), a touch screen, a touchpad, a joystick, a mouse, a light pen, a stylus, touch panel, trackball, speakers, microphone, augmented reality, optical sensors, tablet, indicator lights, projector, camera, biometric identification tools (retinal scanner, fingerprint scanner, voice recognition module), etc.

[0066] The networking means (406) enables data to be transmitted by the device (400) via an internal or external computer network, such as an Intranet, Internet, LAN, and the like. As one or more means (406) can be used, but not limited to: Ethernet card, GSM modem, GPRS modem, LTE modem, 5G modem, satellite communication module, NFC module, Bluetooth and / or BLE module, Wi-Fi module and others

[0067] Additionally, satellite navigation tools in the device (400) can also be used, for example, GPS, GLONASS, BeiDou, Galileo.

[0068] The submitted application materials disclose preferred examples of the implementation of a technical solution and should not be interpreted as limiting other, particular examples of its implementation that do not go beyond the requested legal protection that are obvious to those skilled in the relevant art.

Claims

FORMULA

1. A computer-implemented method for determining people for inspection when passing through a metal detector, performed using at least one processor and comprising the steps of:

- forming a profile of a person, containing a unique ID and a mass distribution model (MRM), containing at least data on the type of metal, mass and location of a certain type of metal on the human body, and the MRM is formed based on the retrospective information of the passes of the metal detector by this person;

2. The method according to claim 1, characterized in that the type of metal is a non-ferrous and/or ferrous metal.

3. The method according to claim 2, characterized in that the location of the metal on the person is characterized by spatial coordinates.

4. The method according to claim 3, characterized in that the relative position of the non-ferrous metal to the ferrous metal is characterized by spatial coordinates or vectors.

5. The method according to claim 3, characterized in that the location of the metal is additionally characterized by the ratio of the mass of non-ferrous metal to black.

6. The method according to claim 1, characterized in that the basic MRM is formed on the basis of a plurality of readings of electromagnetic fields obtained when a person passes a metal detector.

7. The method according to claim 6, characterized in that the correlations of each electromagnetic field are taken into account.

8. The method according to claim 1, characterized in that the ID of the person is transmitted from the access control system.

9. The method according to claim 8, characterized in that the access control system performs recognition of a person by analyzing his image.

10. The method according to claim 9, characterized in that the access control system performs person recognition using the access means.

11. The method according to claim 1, characterized in that the MRM is presented in the form of a graph reflecting the distances between the masses, the relative spatial arrangement (azimuth), the masses, the type of metal, the ratio between the masses, the relationship between the types of metal masses on a person.

12. A device for identifying people for inspection when passing through a metal detector, containing at least one processor and at least one memory means containing machine-readable instructions that, when executed by the processor, carry out the method according to any one of paragraphs. 1-11.

13. A system for identifying people for screening, containing a computing device connected to a metal detector, in which, using a computing device, a person profile is formed containing a unique ID and a mass distribution model (MPM) containing at least data on the type of metal, mass, location of a certain the type of metal carried by the person, and all relationships between the specified data, and the MRM is formed on the basis of retrospective information about the passes of the metal about the detector by this person; determine the ID of the person; using a metal detector, data is obtained from the current MRM of a person; using a computing device, the current MRM is compared with the base MRM, in which the deviation of the mass, type and location of a certain type of metal on a person from the base MRM is analyzed, and if a deviation is detected for each of the specified MRM parameters, a signal is generated for the inspection of the person.

14. The system according to claim 13, characterized in that the metal detector is multi-zone.

15. The system according to claim 13, characterized in that the type of metal is non-ferrous and/or ferrous metal.

16. The system according to claim 14, characterized in that the location of the metal on the person is characterized by spatial coordinates.

17. The system according to claim 16, characterized in that the relative position of the non-ferrous metal to the ferrous metal is characterized by spatial coordinates and/or vectors.

18. The system according to claim 14, characterized in that the location of the metal is further characterized by the ratio of the mass of non-ferrous metal to black.

19. The system according to claim 13, characterized in that the basic MRM is formed on the basis of a plurality of readings of electromagnetic fields obtained when a person passes a metal detector.

20. The system according to claim 19, characterized in that the correlations of each electromagnetic field are taken into account.

21. The system according to claim 13, characterized in that the person's ID is transmitted from the access control system.

22. The system according to claim 21, characterized in that the access control system performs recognition of a person by analyzing his image.

23. The system according to claim 21, characterized in that the access control system performs recognition of a person using an access tool.

24. The system according to claim 13, characterized in that the MRM is presented in the form of a graph reflecting the distances between the masses, the relative spatial arrangement (azimuth), the masses, the type of metal, the ratio between the masses, the relationship between the types of metal masses on a person.