WO2023188332A1 - Dispositif d'identification de personne, procédé d'identification de personne et programme d'identification de personne - Google Patents

Dispositif d'identification de personne, procédé d'identification de personne et programme d'identification de personne Download PDF

Info

Publication number
WO2023188332A1
WO2023188332A1 PCT/JP2022/016703 JP2022016703W WO2023188332A1 WO 2023188332 A1 WO2023188332 A1 WO 2023188332A1 JP 2022016703 W JP2022016703 W JP 2022016703W WO 2023188332 A1 WO2023188332 A1 WO 2023188332A1
Authority
WO
WIPO (PCT)
Prior art keywords
registrant
person
facial
unit
image
Prior art date
Application number
PCT/JP2022/016703
Other languages
English (en)
Japanese (ja)
Inventor
松濤智明
Original Assignee
富士通株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 富士通株式会社 filed Critical 富士通株式会社
Priority to PCT/JP2022/016703 priority Critical patent/WO2023188332A1/fr
Publication of WO2023188332A1 publication Critical patent/WO2023188332A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication
    • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis

Definitions

  • This case relates to a person identification device, a person identification method, and a person identification program.
  • Patent Documents 1 and 2 Technology related to face authentication for identity verification has been developed (for example, see Patent Documents 1 and 2).
  • the present invention aims to provide a person identification device, a person identification method, and a person identification program that can improve the accuracy of person identification.
  • the person identification device acquires a facial image from a captured image, and when at least two facial images are acquired from the captured image, each facial feature information extracted from each of the at least two facial images; Identification for identifying the registrant corresponding to each of the at least two facial images based on whether each degree of similarity with each facial feature information of the plurality of registrants stored in the storage unit is equal to or higher than a threshold value; A person identification device that performs processing, in the identification processing, using a first value as the threshold value to register a first registration corresponding to one face image among the at least two face images from the plurality of registrants.
  • a first identifying unit that identifies a person
  • a determining unit that determines whether a second registrant who is another registrant is registered as an accompanying person of the first registrant identified by the first identifying unit; , when the second registrant is registered, a second value smaller than the first value is used as the threshold to identify the first registrant among the at least two facial images.
  • a second identifying unit that identifies the second registrant by determining whether a facial image other than the facial image corresponds to the second registrant.
  • (a) is a block diagram illustrating the overall configuration of the person identification device
  • (b) is a block diagram illustrating the hardware configuration of the person identification device. It is a flowchart showing an example of the registration process which a person identification device performs. It is a figure which illustrates the registration data which a registration data storage part memorize
  • (a) to (c) are diagrams illustrating accompanying person determination.
  • (a) and (b) are diagrams illustrating accompanying person determination.
  • (a) and (b) are diagrams illustrating registration information addition processing.
  • Biometric authentication is a technology that uses biometric features such as fingerprints, faces, and veins to verify a person's identity.
  • biometric authentication when verification is required, biometric data for verification obtained by a sensor is compared (verified) with registered biometric data that has been registered in advance, and it is determined whether the degree of similarity exceeds the identity determination threshold. The person's identity is verified by making a judgment.
  • facial recognition technology is attracting attention as a means of contactless identification.
  • Face recognition technology is used in a variety of applications because facial images, unlike fingerprints and veins, can be easily acquired using a general camera.
  • facial recognition technology is used not only for access control of personal devices such as personal computers and smartphones, but also for a variety of applications such as entering and exiting rooms and verifying identity at boarding gates at airports.
  • facial recognition is broadly classified depending on whether or not the registrant is aware that his or her face is being photographed. When registrants are conscious of having their photos taken, facial recognition is used for logging into personal computers and smartphones, cashless payments, etc. If the registrant is not conscious of being photographed, facial recognition is used to identify people on surveillance cameras.
  • a person identification device a person identification method, and a person identification program that can improve the accuracy of face recognition (person identification accuracy) even when a registrant is not conscious of photographing will be described.
  • FIG. 1(a) is a block diagram illustrating the overall configuration of the person identification device 100.
  • the person identification device 100 includes an acquisition unit 11, an extraction unit 12, a biological information registration unit 13, an accompanying person registration unit 14, a registered data storage unit 15, a first identification unit 16, an accompanying person It functions as a user determination unit 17, a second identification unit 18, an update unit 19, an authentication information registration unit 20, and the like.
  • FIG. 1(b) is a block diagram illustrating the hardware configuration of the person identification device 100.
  • the person identification device 100 includes a CPU 101, a RAM 102, a storage device 103, an interface 104, a display device 105, an input device 106, a camera 107, and the like.
  • a CPU (Central Processing Unit) 101 is a central processing unit.
  • CPU 101 includes one or more cores.
  • a RAM (Random Access Memory) 102 is a volatile memory that temporarily stores programs executed by the CPU 101, data processed by the CPU 101, and the like.
  • the storage device 103 is a nonvolatile storage device.
  • a ROM Read Only Memory
  • SSD solid state drive
  • the storage device 103 stores a person identification program.
  • the interface 104 is an interface device with external equipment.
  • the interface 104 is an interface device with a LAN (Local Area Network).
  • the display device 105 is a display device such as an LCD (Liquid Crystal Device).
  • the input device 106 is an input device such as a keyboard or a mouse.
  • the camera 107 is a MOS (Metal Oxide Semiconductor) sensor, a CCD (Charged Coupled Device) sensor, or the like.
  • the acquisition unit 11, the extraction unit 12, the biological information registration unit 13, the companion registration unit 14, the registered data storage unit 15, the first identification unit 16, the companion determination unit 17, and the second A specifying unit 18, an updating unit 19, and an authentication information registration unit 20 are implemented.
  • hardware such as a dedicated circuit may be used.
  • FIG. 2 is a flowchart illustrating an example of a registration process executed by the person identification device 100.
  • the acquisition unit 11 acquires the registrant's face image taken by the camera 107 and the registrant's ID (identification information) (step S1). Since the registration process is a process for the registrant to register his/her own face image, the registrant is aware of the photographing process. Therefore, it is possible to obtain sufficient information necessary for facial recognition, such as whether the user is facing forward.
  • the ID is input by, for example, a registrant using the input device 106.
  • the extraction unit 12 extracts facial features from the facial image acquired in step S1 as biological information (facial feature information).
  • the biometric information registration unit 13 associates the biometric information extracted by the extraction unit 12 with the ID of the registrant, and registers it in the registration data stored in the registration data storage unit 15 (step S2).
  • the acquisition unit 11 acquires the accompanying person information of the registrant (step S3).
  • the accompanying person information is input by, for example, a registrant using the input device 106.
  • the companion registration unit 14 registers the companion information in the registration data stored in the registration data storage unit 15 in association with the ID of the registrant (step S4).
  • information on each registrant is stored as registration data.
  • FIG. 3 is a diagram illustrating the registration data stored (or registered) in the registration data storage section 15.
  • biometric information and companion information are associated with each ID.
  • the accompanying person is a person who has many opportunities to act together with each registrant.
  • accompanying persons include relatives of the registrant such as wife, husband, and children.
  • accompanying persons include business associates of the registrant, such as colleagues, superiors, and subordinates.
  • accompanying persons include people with whom the registrant thinks there are many opportunities to act together. Therefore, for one registrant, one accompanying person may be registered, or a plurality of accompanying persons may be registered. In the example of FIG.
  • FIG. 4 is a flowchart illustrating an example of the authentication process executed by the person identification device 100.
  • the acquisition unit 11 acquires an image photographed by the camera 107 (photographed image) (step S11).
  • the camera 107 in this case may be installed separately from the camera for registration processing.
  • the authentication process is a process in which the registrant identifies an ID and the like when taking normal actions, so the registrant is not conscious of photographing. Therefore, the face may be turned sideways or downwards with respect to the camera 107, and it may not be possible to obtain sufficient information necessary for face authentication.
  • FIG. 5(a) is a diagram illustrating an image acquired in step S11.
  • the image includes face images of two people (the first person on the left and the second person on the right), but the image may contain only one person. Sometimes there are three or more people.
  • the acquisition unit 11 attaches time (time information) to the images so as to be able to identify the order in which images were taken at different times. Further, for example, the acquisition unit 11 may acquire images taken by a plurality of cameras 107 from different angles.
  • the extraction unit 12 extracts biometric information from the image acquired in step S11 (step S12). For example, as illustrated in FIG. 5B, the facial region of each person is extracted from the image, and biological information such as facial features is extracted from the facial region. If one person is included in the image, the face of one person is extracted. If the image includes multiple people, each person's face is extracted. In the example of FIG. 5(b), the faces of two people (a first person and a second person) are extracted.
  • the first identifying unit 16 performs the first authentication by comparing the biometric information of each face extracted in step S12 with each biometric information of the registered data stored in the registered data storage unit 15.
  • Step S13 the degree of similarity (first similarity) with the biometric information of each face extracted in step S12 is equal to or greater than the first threshold (first degree of similarity). value) or higher. If there is biometric information having a degree of similarity determined to be greater than or equal to the first threshold, the face is identified (determined) as the face of the person (first person) having the ID of the biometric information. For example, in the example of FIG.
  • the ID of the first person on the left is specified as 0001 (the first registrant).
  • the person on the right no biometric information with a degree of similarity equal to or higher than the first threshold was found, so the ID has not been specified.
  • the person whose face has the biometric information ID having the highest similarity among the plurality of first similarities may be specified.
  • the companion determining unit 17 determines whether there is a face whose ID has not been specified among the faces extracted in step S12 (whether there is a person) (step S14).
  • person A second person whose ID was not specified in the image acquired in step S11 is specified.
  • the person on the right is identified as person A.
  • the accompanying person determination unit 17 performs accompanying person determination for each person appearing in the image acquired in step S11, and identifies the accompanying person of person A (step S15).
  • Accompanying person determination is to sequentially focus on each person in the image and determine whether or not the person is a person A's accompanying person.
  • the accompanying person determination unit 17 calculates the probability C that each person is an accompanying person of person A, and identifies a person whose certainty C is equal to or greater than a threshold value as an accompanying person of person A.
  • the likelihood C is the number of consecutive frames in which the person of interest and person A are photographed together, as illustrated in FIG. 5(c).
  • the predetermined person can be identified as a companion of person A.
  • the likelihood C is a variable depending on the reciprocal of the distance.
  • the variable according to the reciprocal of the distance is, for example, ⁇ N (1/d(i)).
  • N is the number of consecutive frames of images in which a predetermined person is photographed together with person A
  • d(i) is the distance between the center of the face of person A in the i-th frame and the center of the face of the predetermined person. It is. This variable takes a large value when a predetermined person is close to person A for a predetermined period of time.
  • the second specifying unit 18 determines whether or not there is a person whose ID was specified in step S13 among the accompanying persons specified in step S15 (step S16).
  • the determination is "Yes". In this case, the person on the left will be identified as the accompanying person.
  • the second identification unit 18 performs second authentication by comparing the biometric information of each acquired ID with the biometric information of the person A extracted in step S12 (step S17). Specifically, the second identifying unit 18 determines that each degree of similarity (second degree of similarity) between the acquired biometric information of each ID and the biological information of person A extracted in step S12 is equal to or greater than the second threshold. Determine whether or not.
  • biometric information with a similarity determined to be equal to or higher than a second threshold (second value)
  • the biometric information ID for which the face of person A has the highest similarity among the plurality of second similarities is determined. It may be specified that it is a person's face.
  • the false acceptance rate is the rate at which a person is mistakenly recognized as the real person when the person is not the real person.
  • the second threshold is set to a lower value than the first threshold. It is preferable to set it to .
  • the identity acceptance rate is the rate at which a person can be recognized as the person in question.
  • the threshold value may be set to a value smaller than the first threshold value.
  • the second threshold value may be changed depending on the relationship with the person A in the image acquired by the acquisition unit 11.
  • the second threshold value may be changed depending on the probability C.
  • the second threshold value may be changed depending on the number of IDs included in the companion group. For example, the second threshold value may be lowered as the number of IDs included in the companion group is smaller, and the second threshold value may be lowered as the number of IDs included in the companion group is larger.
  • the second identification unit 18 determines whether or not the ID of person A was identified in step S18 (step S18).
  • step S18 If the determination in step S18 is "Yes”, IDs will be identified for all faces included in the image. Therefore, the execution of the flowchart ends. Even if the determination in step S14 is "No", the IDs will be identified for all faces included in the image, so the execution of the flowchart ends.
  • step S11 in this case, an image of the next frame after a predetermined time interval and at a different time is acquired.
  • the first identifying unit 16 can identify person A by repeating the process from step S11. may be possible. In this case, the second authentication would be omitted.
  • the authenticity of the person can be increased by using information on companions who are likely to act together. can be improved. Thereby, the accuracy of identifying a person can be improved. Furthermore, even when multiple people are photographed without the registrant being aware of the identity verification of surveillance camera images, the accuracy of person identification, authentication, and identification can be improved.
  • FIG. 6 is a flowchart illustrating an example of registration information addition processing executed by the person identification device 100.
  • the registration information addition process is a process that is executed when step S13 in FIG. 4 is executed.
  • the updating unit 19 determines whether the likelihood C acquired in step S22 is greater than or equal to the threshold (step S23).
  • step S23 the updating unit 19 updates the authentication information stored in the authentication information registration unit 20 with the result of accompanying person determination for each person whose likelihood C is determined to be equal to or higher than the threshold value.
  • the updating unit 19 determines whether there is an ID for which accompanying determination has been made a predetermined number of times (for example, three times) or more in the past in a predetermined period (for example, one week) (step S25).
  • step S25 the updating unit 19 adds the companion information to the authentication information stored in the authentication information registration unit 20 (step S26).
  • a person who is not registered in advance as an accompanying person and who is presumed to be an accompanying person is registered as an accompanying person.
  • the number of matching targets for the person who was not identified in step S12 increases, and the accuracy of identifying the person improves.
  • FIG. 9 is a flowchart illustrating an example of registered information deletion processing executed by the person identification device 100.
  • the registration information deletion process is periodically executed at a predetermined cycle. Further, the process is executed for each ID registered in the registration data stored in the registration data storage unit 15.
  • the updating unit 19 refers to the authentication information stored in the authentication information registration unit 20 (step S31).
  • the update unit 19 updates the IDs included in the accompanying person information of the registration data for which the number of times the accompanying person was determined to be less than a threshold value (for example, 1 time) during a predetermined period in the past (for example, one month) It is determined whether there is anyone accompanying the user (step S32).
  • a threshold value for example, 1 time
  • a person who has been registered in advance as an accompanying person and who has not been accompanied for a while is deleted.
  • the person identification device 100 may include a calculation unit that calculates each degree of similarity between the biometric information of each acquired ID and the biometric information of the person extracted in step S12.
  • the first identifying unit 16 may calculate each similarity.
  • the second specifying unit 18 may calculate each similarity.
  • the first identifying unit 16 may perform the first authentication on the first person and the second person whose face image is included in the photographed image. Further, if the determination in step S18 is "No" regarding the second person, the first identification unit 16 may perform the first authentication on the second person in step S13. By doing so, for example, the processing load related to the first authentication for the second person can be reduced.
  • the first specifying unit 16 selects the first registrant based on some registrants narrowed down by other means among the plurality of registrants whose registration data is stored in the registration data storage unit 15. Authentication may also be performed. For example, among the plurality of registrants, the search may be narrowed down to registrants identified by the first authentication or the second authentication in a predetermined period. For example, the number of times specified by the first authentication or the second authentication may be stored, and among the plurality of registrants, the registrants may be narrowed down to registrants for whom the number of times specified is a certain number or more. By doing so, for example, the processing load related to the first authentication or the second authentication can be reduced.
  • the acquisition unit 11 is an example of an acquisition unit that acquires photographed images that include a plurality of first persons and second persons photographed at different times.
  • the extraction unit 12 is an example of an extraction unit that extracts biometric information from each of the first person and the second person.
  • the first identification unit 16 acquires a photographed image including a face image of a first person and a face image of a second person, facial feature information extracted from the face image of the first person included in the acquired photographed image.
  • This is an example of a first specifying unit that specifies a registrant corresponding to a first person among a plurality of registrants based on the facial feature information of each of the plurality of registrants registered in the registration data storage unit.
  • the second specifying unit 18 registers the facial feature information extracted from the facial image of the second person included in the acquired captured image in the registered data storage unit in association with the registrant specified by the first specifying unit.
  • This is an example of a second identifying unit that identifies a related registrant corresponding to a second person among one or more related registrants based on facial feature information of each of the one or more related registrants.
  • the authentication information registration unit 20 acquires a plurality of images taken at different times, the first identifying unit identifies the first person and the second person included in the image taken at the first time, and the first person and the second person included in the image taken at the first time are identified.
  • the identification information of the first registrant and the identification information of the second registrant are associated with each other in the registration data storage unit.
  • This is an example of a registration unit that registers.
  • the updating unit 19 is an example of an updating unit that updates the relationship between the identification information of each of a plurality of registrants and the identification information of other registrants according to the contents registered by the registration unit.
  • the first identification unit 16 acquires a facial image from the captured image and acquires at least two facial images from the captured image, each facial feature information extracted from each of the at least two facial images. and identifying the registrant corresponding to each of the at least two facial images based on whether each degree of similarity between the facial feature information of the plurality of registrants stored in the storage unit is equal to or higher than a threshold value.
  • a first identifying unit that uses a first value as a threshold to identify a first registrant corresponding to one face image among at least two face images from a plurality of registrants in the identifying process.
  • the accompanying person determining unit 17 is an example of a determining unit that determines whether a second registrant who is another registrant is registered as an accompanying person of the first registrant identified by the first identifying unit.
  • the second identifying unit 18 uses a second value smaller than the first value as the threshold to identify the first registrant from at least two facial images.
  • This is an example of a second specifying unit that specifies a second registrant by determining whether a face image other than the first registrant corresponds to the second registrant.
  • the acquisition unit 11 is an example of an acquisition unit that acquires a plurality of captured images taken at different times.
  • the authentication information registration unit 20 acquires a plurality of captured images taken at different times
  • the first identification unit and the second identification unit identify the first registrant and the second registration included in the captured image at the first time.
  • the identification information of the first registrant is stored in the storage section.
  • This is an example of a registration unit that registers the second registrant in association with the identification information of the second registrant.
  • the updating unit 19 is an example of an updating unit that updates the relationship regarding companions between the first registrant and the second registrant according to the content registered by the registration unit.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Collating Specific Patterns (AREA)

Abstract

L'invention concerne un dispositif d'identification de personne qui, lorsqu'au moins deux images de visage ont été acquises à partir d'une image capturée, effectue un traitement d'identification pour identifier des personnes enregistrées correspondant aux au moins deux images de visage sur la base du fait que les similarités entre des informations de caractéristique faciale extraites de chacune des deux images de visage ou plus et des informations de caractéristique de visage d'une pluralité de personnes enregistrées stockées dans une unité de stockage sont supérieures ou égales à une valeur seuil. Le dispositif d'identification de personne comprend : une première unité d'identification qui utilise une première valeur en tant que valeur seuil pour identifier, à partir de la pluralité de personnes enregistrées, une première personne enregistrée qui correspond à une image de visage parmi les deux images de visage ou plus ; une unité de détermination qui détermine si une seconde personne enregistrée, qui est une autre personne enregistrée, a été enregistrée en tant que personne accompagnant la première personne enregistrée ; et une seconde unité d'identification qui, lorsqu'une seconde personne enregistrée a été enregistrée, utilise comme valeur seuil une seconde valeur qui est inférieure à la première valeur, et identifie la seconde personne enregistrée en déterminant si l'image de visage parmi les deux images de visage ou plus qui n'est pas l'image de visage dans laquelle la première personne enregistrée a été identifiée correspond ou non à la seconde personne enregistrée.　
PCT/JP2022/016703 2022-03-31 2022-03-31 Dispositif d'identification de personne, procédé d'identification de personne et programme d'identification de personne WO2023188332A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/016703 WO2023188332A1 (fr) 2022-03-31 2022-03-31 Dispositif d'identification de personne, procédé d'identification de personne et programme d'identification de personne

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/016703 WO2023188332A1 (fr) 2022-03-31 2022-03-31 Dispositif d'identification de personne, procédé d'identification de personne et programme d'identification de personne

Publications (1)

Publication Number Publication Date
WO2023188332A1 true WO2023188332A1 (fr) 2023-10-05

Family

ID=88200355

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/016703 WO2023188332A1 (fr) 2022-03-31 2022-03-31 Dispositif d'identification de personne, procédé d'identification de personne et programme d'identification de personne

Country Status (1)

Country Link
WO (1) WO2023188332A1 (fr)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007249953A (ja) * 2006-02-15 2007-09-27 Toshiba Corp 人物認識装置および人物認識方法
JP2016046639A (ja) * 2014-08-21 2016-04-04 セコム株式会社 認証装置
JP2018200597A (ja) * 2017-05-29 2018-12-20 株式会社Ｎｔｔドコモ 画像推定装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007249953A (ja) * 2006-02-15 2007-09-27 Toshiba Corp 人物認識装置および人物認識方法
JP2016046639A (ja) * 2014-08-21 2016-04-04 セコム株式会社 認証装置
JP2018200597A (ja) * 2017-05-29 2018-12-20 株式会社Ｎｔｔドコモ 画像推定装置

Similar Documents

Publication Publication Date Title
JP6634127B2 (ja) カメラ付きデバイスに関連する生体認証のためのシステム及び方法
KR102455633B1 (ko) 라이브니스 검사 방법 및 장치
JP4862447B2 (ja) 顔認識システム
JP6483485B2 (ja) 人物認証方法
US6810480B1 (en) Verification of identity and continued presence of computer users
TWI727329B (zh) 用於基於深度學習方法提供對資源之選擇性存取之防欺騙系統及方法
JP7196932B2 (ja) 情報処理装置、情報処理方法、およびプログラム
US9292752B2 (en) Image processing device and image processing method
WO2020070821A1 (fr) Dispositif d'identification biométrique, procédé d'identification biométrique et programme d'identification biométrique
Fenu et al. Controlling user access to cloud-connected mobile applications by means of biometrics
CN114127801A (zh) 跨设备网络使用人物可识别性的系统和方法
JP2008257329A (ja) 顔照合装置
CN112183167A (zh) 考勤方法、认证方法、活体检测方法、装置及设备
JP2006085289A (ja) 顔認証システムおよび顔認証方法
WO2023188332A1 (fr) Dispositif d'identification de personne, procédé d'identification de personne et programme d'identification de personne
JP5873959B2 (ja) ホワイトリスト内外判定装置及び方法
Indrawal et al. Development of efficient and secured face recognition using biometrics
JP2021140622A (ja) 認証方法、情報処理装置、及び認証プログラム
Grabovskyi et al. Facial recognition with using of the microsoft face API Service
WO2021060256A1 (fr) Dispositif d'authentification faciale, procédé d'authentification faciale et support d'enregistrement lisible par ordinateur
US12019725B2 (en) Methods and systems for employing an edge device to provide multifactor authentication
Tamilselvi et al. Empirical Assessment of Artificial Intelligence Enabled Electronic Voting System Using Face Biometric Verification Strategy
Gopal et al. Score improvement using backpropagation in biometric recognition system
WO2022185486A1 (fr) Procédé d'authentification, programme d'authentification et dispositif de traitement d'informations
Singh et al. Adapted Facial Recognition And Spoofing Detection For Management Decision Making System: A Visually Impaired People Perspective

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22935474

Country of ref document: EP

Kind code of ref document: A1