CN112199530A - Multi-dimensional face library picture automatic updating method, system, equipment and medium - Google Patents

Multi-dimensional face library picture automatic updating method, system, equipment and medium Download PDF

Info

Publication number
CN112199530A
CN112199530A CN202011141640.6A CN202011141640A CN112199530A CN 112199530 A CN112199530 A CN 112199530A CN 202011141640 A CN202011141640 A CN 202011141640A CN 112199530 A CN112199530 A CN 112199530A
Authority
CN
China
Prior art keywords
face
picture
value
index
fpm
Prior art date
Legal status (The legal status 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 status listed.)
Granted
Application number
CN202011141640.6A
Other languages
Chinese (zh)
Other versions
CN112199530B (en
Inventor
李莹
张凌飞
朱晓莉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin Zhongyi Science And Technology Co ltd
Original Assignee
Tianjin Zhongyi Science And Technology Co ltd
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 Tianjin Zhongyi Science And Technology Co ltd filed Critical Tianjin Zhongyi Science And Technology Co ltd
Priority to CN202011141640.6A priority Critical patent/CN112199530B/en
Publication of CN112199530A publication Critical patent/CN112199530A/en
Application granted granted Critical
Publication of CN112199530B publication Critical patent/CN112199530B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/51—Indexing; Data structures therefor; Storage structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/53—Querying
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/55—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • General Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Collating Specific Patterns (AREA)
  • Processing Or Creating Images (AREA)

Abstract

The invention provides a multi-dimensional face library picture automatic updating method suitable for face recognition, which comprises the following steps: dividing a face recognition scene into three types of static recognition, dynamic recognition and authentication comparison, collecting the three types of face pictures of the same user through different equipment, respectively extracting a picture quality index value and a face attribute index value of a face for each picture by utilizing picture quality calculation and face attribute calculation in a face recognition technology, and classifying and storing the pictures and the index values according to scene values; when the face of the same user is recognized again, extracting the recognition scene value, the picture quality index and the face attribute index of the current face sample picture, acquiring the corresponding face library picture and multiple index values thereof in the face database according to the recognition scene value, calculating the comprehensive quality evaluation score of the face sample picture from multiple dimensions, and if the score is greater than a system threshold value, replacing the original face library picture by the sample face picture, realizing the automatic update of the face library picture and improving the accuracy of face recognition.

Description

Multi-dimensional face library picture automatic updating method, system, equipment and medium
Technical Field
The invention belongs to the technical field of face image library updating, and particularly relates to a multi-dimensional face library image automatic updating method suitable for face recognition.
Background
In recent years, with popularization and promotion of face recognition technology, more and more business scenes comprise face recognition, such as witness verification, face attendance checking, face unlocking, face payment and the like. These business systems including face recognition technology necessarily need to establish a face picture library of system users, generally face pictures provided during user registration, and the business systems store the face pictures in a face database. Therefore, when the face of the user needs to be identified, the current face picture of the user can be obtained through the face data acquisition equipment, and the current face picture and the face library picture are searched and compared. When the similarity of the two pictures is not smaller than the similarity threshold value, the face recognition is successful, otherwise, the face recognition is failed.
However, as time goes by, the user changes the physical or wearing appearance due to age increase, change of life style, requirement of working environment, etc., but the face library picture does not change in time, so that a plurality of recognition failures may occur when the user performs face recognition in a certain day, which may lead to embarrassment that the user cannot handle business or passes inspection, etc.
In order to solve the problem, a simpler method is to remind the user to update the face library picture frequently, or the system automatically replaces the original face library picture with the acquired face sample picture at intervals. However, both of these approaches have significant disadvantages: firstly, for the user individual, the user individual must actively upload the close-up photographs in time, which is a challenge to the memory and the autonomy, and the operation is complicated, and the service experience is not good. And secondly, the updating standard is unclear, for example, how often the updating is reasonable, how to ensure that the user can update in time, and the automatic replacement of pictures by the system can not cause the reduction of the subsequent recognition rate, and the like.
Disclosure of Invention
In view of this, the present invention is directed to a method for automatically updating a multi-dimensional face library picture, which is suitable for face recognition, so as to achieve correctness and scientificity of the automatic update of the face library picture in terms of period and quality, and meanwhile, ensure a high recognition rate.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
in a first aspect, the present invention provides a method for automatically updating a multi-dimensional face library picture suitable for face recognition, including the following steps:
s1, establishing a multi-dimensional index system of the face library pictures;
s2, acquiring a user face image and user registration information during user registration, obtaining each index value corresponding to the index system through calculation, and storing the face image and the index values;
and S3, when the face sample image of a certain user is obtained again, calculating the corresponding index value and comparing the index value with the previously stored index value, if the threshold range is met, updating the face image and the index value of the corresponding user, and if the threshold range is not met, performing no processing.
Further, the multi-dimensional index system in step S1 includes a face picture scene value FS, a face picture quality FQ, and a face attribute value FP;
the face picture scene value FS comprises a dynamic identification FSLStatic identification FSMFS for comparing with testimonyHCorresponding to the three human face pictures obtained by at least one data acquisition device and recording as the human face picture FSLFace picture FSMFace picture FSH；
The face picture quality FQ utilizes a picture quality detection algorithm to face picture FSLFace picture FSMFace picture FSHCalculating to obtain FQL、FQM、FQH；
The face attribute value FP comprises acquisition time T, age A, gender S and a living body value B and is recorded as a matrix FPM.
Further, the index FQ comprises light L, blur F, shielding O, angle R, expression E and person of the face in the face pictureFace proportion P and background C, the actual values of the characteristics are normalized to 0,1 through linear change]Interval, setting the weight of each characteristic value as wi，0<wi<1(1≤i≤7)。
Further, in step S2, when the user registers, three clear face pictures of the user are obtained by three ways of video shooting, photographing and card reader of the identification card and are respectively recorded as FS in the FS indexL、FSM、FSH；
Date of birth conversion by user fill-in is an age note A0Sex S0And automatically recording the time of acquisition as T0And the living body is B0(ii) a The obtained FQL、FQM、FQHAre respectively connected with FQ0Comparing, and if the condition is not met, acquiring the corresponding face image again by using a corresponding mode; the obtained FPML、FPMM、FPMHAnd FPM0And comparing, and if the condition is not met, re-acquiring the face image of the user by using a corresponding mode or checking whether the registration information is wrong and is corrected by the user.
Further, the specific steps of calculating the corresponding index value and comparing the index value with the previously stored index value in step S3 are as follows:
after a face sample image of a certain user is obtained, recording the acquisition time T in the FS index and the FP index;
calculating the picture quality of the face sample image and recording as FQ2Calculating the face attribute value, and recording as FPM2；
Obtaining two previously stored corresponding index values of FQ and FPM of the same user according to the FS index value, and marking as FQ1、FPM1(ii) a Acquiring the weight w occupied by the preset minimum interval duration T of the face updating interval and the face acquisition time TtAnd face age A weight wa，wt+wa1 is ═ 1; with FPM1Subtract FPM2Is calculated as
Figure BDA0002738450080000031
Calculating face attribute value according to formulaFP, obtaining the evaluation weight w of the face picture quality FQ set by the systemqAnd the evaluation weight w of the face property FPp，wq+wp1 is ═ 1; calculating according to a formula to obtain a multi-dimensional comprehensive quality index MDQ of the face sample image;
judging whether the obtained MDQ is larger than a preset quality threshold MDQ or nottIf the MDQ is greater than the quality threshold MDQtReplacing the face image corresponding to the FS value with the face sample image, and simultaneously using FQ2、FPM2Replacement of FQ1、FPM1。
In a second aspect, the present invention provides an automatic updating system for a multi-dimensional face library picture suitable for face recognition, including:
the face image acquisition module is used for acquiring a face image;
the face image quality calculation module is used for calculating an image quality index according to the face image;
the face image attribute calculation module is used for acquiring attribute indexes of the face image;
and the face library picture updating module judges whether to update the face picture corresponding to the face library or not through comparison according to a preset threshold value.
In a third aspect, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for automatically updating a multi-dimensional face library picture suitable for face recognition when executing the program.
In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the above-mentioned method for automatically updating a multi-dimensional face library picture suitable for face recognition.
Compared with the prior art, the method has the following advantages:
precision: the method establishes a multi-dimensional index system of the face library pictures, mainly comprises three indexes of a face picture scene value FS, face picture quality FQ and a face attribute value FP, specifically comprises 3 types and 10 characteristic values, realizes the correctness and scientificity of the automatic updating of the face library pictures in the aspects of period and quality, and simultaneously ensures high recognition rate.
Universality: the method does not take special equipment as a precondition, covers common service scenes such as photographing, shooting and testimony comparison, and is suitable for the butt joint of face recognition modules in most service systems in the market at present.
Controllability: the method can set the minimum face updating period value t in the face library picture updating module, or select to close the face library picture updating module without influencing the acquisition and recognition functions of the face, thereby reducing the performance consumption of a service system.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic structural diagram of an automatic multi-dimensional face picture updating system according to an embodiment of the present invention.
Fig. 2 is a flowchart of a method for automatically updating a multi-dimensional face picture according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the system for automatically updating a multi-dimensional face picture provided in the embodiment of the present invention may at least include: the system comprises a face image acquisition module 1, a face image quality calculation module 1, a face image attribute calculation module 1 and a face library image updating module 1. It should be noted that the system functional blocks and structures shown in fig. 1 are exemplary only, and not limiting, and the system may have other functional blocks and structures as desired.
The face library picture updating module can set a minimum interval duration t (such as milliseconds of 1 month and 1 year), whether to start automatic updating, whether to perform asynchronous processing and other parameter configurations.
Taking a service system supporting multiple parameter configurations such as mobile phone photographing recognition, camera monitoring recognition, whether to start automatic updating, whether to perform asynchronous processing or not for identity card verification and the like of the human face as an example, a multi-dimensional human face picture automatic updating method according to an embodiment of the invention will be described with reference to fig. 2.
In one embodiment, firstly, a multi-dimensional index system and a data structure of a face library picture need to be established in a system, wherein the multi-dimensional index system is mainly divided into three indexes of a face picture scene value FS, a face picture quality FQ and a face attribute value FP.
1) Establishing a data structure of an index FS, and setting a dynamic identification FS in the systemLStatic identification FSMWitness comparison FSHAnd waiting for three major classifications.
Dynamic identification of FSL(e.g., video, quality requirement level low), static identification FSM(for example, the quality requirement grade is midle) and the testimony contrast FSH(for example, certificate photo, quality requirement grade high)
2) Establishing an index FQ data structure, specifically comprising seven characteristics of light L, blur F, occlusion O, angle R, expression E, face proportion P, background C and the like of the face in the picture, and standardizing the actual values of the characteristics to [0,1 ] through linear change]Interval, in the system, setting the weight of each characteristic value as wi，0<wi<1 (i is more than or equal to 1 and less than or equal to 7), and setting the standard minimum value as FQ0Then, the formula for calculating FQ is:
FQ＝w1L+w2F+w3O+w4R+w5E+w6P+w7C
the standard minimum value is FQ0The standard values and the weights of the seven characteristics set by the system are calculated;
3) establishing an index FP numberAccording to the structure, the method specifically comprises the acquisition time T (milliseconds), the age A (milliseconds), and the sex
Figure BDA0002738450080000061
Value of living body
Figure BDA0002738450080000062
Equal 4 characteristics, recorded as a matrix
Figure BDA0002738450080000063
4) And establishing a face database and a structure thereof based on the index data structure.
The system registers a human face picture in an initial state and a scene value FS thereof through a human face picture acquisition module; for example, in an embodiment, when a user registers a face for the first time in the system, the user uploads a "face picture FS" with a static scene value to the system by taking a picture with a smart phoneL"; then the user normally walks under the intelligent monitoring camera, and the system automatically acquires a human face picture FS with a dynamically identified scene valueM"; then the user uses the ID card and the card reader to check the ID card, and after the check is successful, the system automatically obtains the read electronic photo (containing the face) of the ID card as a' face picture FSH"; finally, the user inputs other registration information including birth date, gender and the like in the system.
The system records the information as a' face picture FSL"," face picture FSM"," face picture FSH", which in turn correspond to FS in the FS index respectivelyL、FSM、FSHStoring three face pictures according to FS classification;
the system converts the birth date of the user into an age record A0Sex as S0And simultaneously, the system records the time of the acquisition as T0And the living body is recorded as B0According to
Figure BDA0002738450080000071
Calculate matrix data FPM0And storing.
In one embodiment, the system obtains an index FQ through a face recognition engine and a face picture quality calculation module; using picture quality detection algorithm and technique, for' face picture FSL"," face picture FSM"," face picture FSH", respectively calculate FQL、FQM、FQH(ii) a Will FQL、FQM、FQHAre respectively connected with FQ0Subtracting, and if the calculation result is less than 0, requiring to use a corresponding mode to obtain the user photo again; if the calculation result is greater than or equal to 0, storing the FQ according to the FS classificationL、FQM、FQH；
Thirdly, acquiring an index FPM through a face recognition engine and a face attribute calculation module; the specific method is that the face image FS is processed by face recognition algorithm and techniqueL"," face picture FSM"," face picture FSH", calculate FPML、FPMM、FPMH(ii) a Mixing FPML、FPMM、FPMHRespectively with FPM0Subtract and record the calculation result as
Figure BDA0002738450080000072
According to the formula
Figure BDA0002738450080000073
If FP is equal to 0, requiring to use the corresponding mode to obtain the user's photo again or requiring the user to check whether the registration information is wrong and correct, namely, re-performing step 2; if FP equals 1, then store FPM by FS classificationL、FPMM、FPMH。
For example, to "face picture FSL"carry out calculation to obtain FQLJudgment of FQLIf the value is qualified, if FQLLess than FQ0，Prompting the user to reuse the smart phone to take a picture and upload a 'face picture FS' with a static scene value to the systemL"; otherwise, the system proceeds toThe 'face picture FS' is processed by a face recognition engine and a face attribute calculation moduleL"carry out calculation to obtain FPMLFrom FPMLSubtract FPM0Calculating to obtain FPL，Judging FPLIf the value is qualified, if FPLIs equal to 0，Prompting the user to reuse the smart phone to take a picture and upload a 'face picture FS' with a static scene value to the systemL"; if FPLIf the number is equal to 1, saving the 'face picture FS' of the user in a face databaseL"and FQ thereofL、FPML。
In one embodiment, the system converts the 'face picture FS' through a face recognition engine and a face picture quality calculation moduleM"carry out calculation to obtain FQMJudgment of FQMIf the value is qualified, if FQMLess than FQ0，Prompting the user to reuse the smart phone to take a picture and upload a 'face picture FS' with a static scene value to the systemM"; otherwise, the system further uses the face recognition engine and the face attribute calculation module to convert the' face picture FSM"carry out calculation to obtain FPMMFrom FPMMSubtract FPM0Calculating to obtain FPM，Judging FPMIf the value is qualified, if FQMIs equal to 0，Prompting the user to reuse the smart phone to take a picture and upload a 'face picture FS' with a static scene value to the systemM"; if FQMIf the number is equal to 1, saving the 'face picture FS' of the user in a face databaseM"and FQ thereofM、FPMM。
In one embodiment, the system converts the 'face picture FS' through a face recognition engine and a face picture quality calculation moduleH"carry out calculation to obtain FQHJudgment of FQHIf the value is qualified, if FQHLess than FQ0，Prompting the user to reuse the smart phone to take a picture and upload a 'face picture FS' with a static scene value to the systemH"; otherwise, the system further uses the face recognition engine and the face attribute calculation module to convert the' face picture FSH"carry out calculation to obtain FPMHFrom FPMHSubtract FPM0Calculating to obtain FPH，Judging FPHIf the value is qualified, if FQHIs equal to 0，Prompting the user to reuse the smart phone to take a picture and upload a 'face picture FS' with a static scene value to the systemH"; if FQHIf the number is equal to 1, saving the 'face picture FS' of the user in a face databaseH"and FQ thereofH、FPMH。
In one embodiment, in the following operation process of the system, whenever a face image acquisition module of the system acquires a face image (hereinafter referred to as "sample image") of a certain user through video recording or photographing or witness comparison, a face image scene value FS (video scene is denoted as FS ═ FS) is recorded (video scene is denoted as FS ═ FSLOr the shooting scene is recorded as FS ═ FSMOr FS is recorded as the witness contrast sceneH) And recording the acquisition time as T, and transmitting the parameters to other functional modules (namely a face library picture updating module) downwards.
The face library picture updating module of the system acquires a corresponding value according to the FS (FS) through the FS valueL、FSMOr FSH) Acquiring a face library picture set corresponding to the FS grade of the same user from a system face database, carrying out 1: N searching and comparison on sample pictures, and if the comparison and identification are successful, further acquiring two index values, namely FQ and FPM, of the identified face library picture by the system, and recording the two index values as FQ1、FPM1；
The face image quality calculation module and the face attribute calculation module of the system calculate the image quality of the sample image and record the image quality as FQ by using an image quality detection algorithm and technology after receiving the sample image and FS data2(ii) a Calculating the face attribute value of the 'sample picture' by using the face recognition technology, and recording as FPM2；
In one embodiment, the system obtains the set minimum interval duration T of the face update interval and the weight w occupied by the face acquisition time TtAnd face age A weight wa(ii) a With FPM1Subtract FPM2The result of the calculation is a face attribute value FP, i.e. FPM1Subtract FPM2Is calculated byThe result is that
Figure BDA0002738450080000091
The face attribute value
Figure BDA0002738450080000092
The system obtains the evaluation weight w of the quality FQ of the set human face pictureqAnd the evaluation weight w of the face property FPp(ii) a According to the formula
MDQ＝wq×(FQ2-FQ1)+wp×FP
And calculating to obtain the multi-dimensional comprehensive quality index MDQ of the face sample picture.
If the MDQ is larger than the quality threshold value MDQ set by the systemtIf yes, the face library picture updating module replaces the face library picture corresponding to the FS value with the sample picture, and simultaneously uses the FQ2、FPM2Replacement of FQ1、FPM1(ii) a Otherwise, no update process is performed.
Note that, when the face image acquisition module is registered by a user, if the acquisition of all three items of data, namely dynamic identification, static identification and testimony comparison, cannot be completed due to reasons of the user or insufficient equipment conditions, any one of the acquired values (including the image, the FQ value and the FP value) can be stored as the other two items.
The embodiment of the invention can realize the automatic updating method of the multi-dimensional face library pictures suitable for face recognition, and provides a basis for improving the accuracy of the face recognition result.
In specific implementation, the face library picture updating module may not be operated; the synchronous execution can be performed after the face attribute calculation module, or the asynchronous execution is formed by utilizing a distributed architecture or a timing task; thereby reducing performance consumption of the business system.
Another embodiment discloses a system of the above disclosed method for automatically updating a multi-dimensional face library picture suitable for face recognition, which is a virtual device structure corresponding to the method, and includes:
the face image acquisition module is used for acquiring a face image;
the face image quality calculation module is used for calculating an image quality index according to the face image;
the face image attribute calculation module is used for acquiring attribute indexes of the face image;
and the face library picture updating module judges whether to update the face picture corresponding to the face library or not through comparison according to a preset threshold value.
Another embodiment discloses a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the processor implements the above-mentioned method for automatically updating a multi-dimensional face library picture suitable for face recognition.
Another embodiment discloses a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the above-mentioned method for automatically updating a multi-dimensional face library picture suitable for face recognition.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. It should be understood that the technical solutions of the present invention may be embodied in the form of a software product, and the computer software product may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes instructions for enabling an electronic device (which may be a mobile phone, a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the system for automatically updating a multi-dimensional face library picture suitable for face recognition, each included unit and module are only divided according to functional logic, but are not limited to the above division, as long as corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
For the system disclosed by the embodiment, the description is simple because the system corresponds to the method disclosed by the embodiment, and the relevant points refer to the description of the method part. In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The above description is only for the specific embodiment of the present invention or the description thereof, and the protection scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the protection scope of the present invention. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (9)

1. The multi-dimensional face library picture automatic updating method suitable for face recognition is characterized by comprising the following steps of:
s1, establishing a multi-dimensional index system of the face library pictures;
s2, acquiring a user face image and user registration information during user registration, obtaining each index value corresponding to the index system through calculation, and storing the face image and the index values;
and S3, when the face sample image of a certain user is obtained again, calculating the corresponding index value and comparing the index value with the previously stored index value, if the threshold range is met, updating the face image and the index value of the corresponding user, and if the threshold range is not met, performing no processing.
2. The method of claim 1, wherein: the multi-dimensional index system in the step S1 includes a face picture scene value FS, a face picture quality FQ, and a face attribute value FP;
the face picture scene value FS comprises a dynamic identification FSLStatic identification FSMFS for comparing with testimonyHCorresponding to the three human face pictures obtained by at least one data acquisition device and recording as the human face picture FSLFace picture FSMFace picture FSH；
The face picture quality FQ comprises a plurality of index values, and a picture quality detection algorithm is utilized to carry out FS on the face pictureLFace picture FSMFace picture FSHCalculating to obtain FQL、FQM、FQH；
The face attribute value FP comprises acquisition time T, age A, gender S and living body value B, and is recorded as a matrix
Figure FDA0002738450070000011
3. The method of claim 2, wherein: the index FQ comprises seven characteristics of light L, blur F, shielding O, angle R, expression E, face proportion P and background C of the face in the face picture, and the actual values of the characteristics are normalized to [0,1 ] through linear change]Interval, setting the weight of each characteristic value as wi，0<wi<1 (i is more than or equal to 1 and less than or equal to 7), the formula for calculating FQ is
FQ＝w1L+w2F+w3O+w4R+w5E+w6P+w7B
Set the standard minimum value to FQ0And calculating the standard value and the weight of the seven characteristics set by the system.
4. The method of claim 2, wherein: in step S2, when the user registers, three clear face pictures of the user are obtained by three ways of video shooting, photographing and card readerIs recorded as FS in FS indexL、FSM、FSH；
Date of birth conversion by user fill-in is an age note A0Sex S0And automatically recording the time of acquisition as T0And the living body is B0Is denoted as FPM0；
Will FQL、FQM、FQHRespectively with the standard minimum value FQ0Subtracting, and if the calculation result is less than 0, re-acquiring the corresponding face image by using a corresponding mode; if the calculation result is greater than or equal to 0, storing the FQ according to the FS classificationL、FQM、FQH；
Mixing FPML、FPMM、FPMHRespectively with FPM0Subtract and record the calculation result as
Figure FDA0002738450070000021
According to the formula
Figure FDA0002738450070000022
If FP is equal to 0, the corresponding mode is required to be used to obtain the face image of the user again or the user checks whether the registration information is wrong and correct in modification; if FP equals 1, then store FPM by FS classificationL、FPMM、FPMH。
5. The method of claim 2, wherein: the specific steps of calculating the corresponding index value and comparing the index value with the previously stored index value in step S3 are as follows:
after a face sample image of a certain user is obtained, recording the acquisition time T in the FS index and the FP index;
calculating the picture quality of the face sample image and recording as FQ2Calculating the face attribute value, and recording as FPM2；
Obtaining two previously stored corresponding index values of FQ and FPM of the same user according to the FS index value, and marking as FQ1、FPM1(ii) a Acquiring the weight w occupied by the preset minimum interval duration T of the face updating interval and the face acquisition time TtAnd face age A weight wa，wt+wa1 is ═ 1; with FPM1Subtract FPM2Is calculated as
Figure FDA0002738450070000031
The face attribute value
Figure FDA0002738450070000032
Obtaining the evaluation weight w of the face picture quality FQ set by the systemqAnd the evaluation weight w of the face property FPp，wq+wp1 is ═ 1; according to the formula
MDQ＝wq×(FQ2-FQ1)+wp×FP
Calculating to obtain a multi-dimensional comprehensive quality index MDQ of the face sample image;
judging whether the obtained MDQ is larger than a preset quality threshold MDQ or nottIf the MDQ is greater than the quality threshold MDQtReplacing the face image corresponding to the FS value with the face sample image, and simultaneously using FQ2、FPM2Replacement of FQ1、FPM1。
6. The method of claim 2, wherein: if the collection of all three data of dynamic identification, static identification and testimony comparison can not be finished during the registration of the user, any one collection value can be simultaneously stored as the other two collection values.
7. Multi-dimensional face storehouse picture automatic update system suitable for face identification, its characterized in that:
the face image acquisition module is used for acquiring a face image;
the face image quality calculation module is used for calculating an image quality index according to the face image;
the face image attribute calculation module is used for acquiring attribute indexes of the face image;
and the face library picture updating module judges whether to update the face picture corresponding to the face library or not through comparison according to a preset threshold value.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1-6 are implemented when the program is executed by the processor.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6.
CN202011141640.6A 2020-10-22 2020-10-22 Multi-dimensional face library picture automatic updating method, system, equipment and medium Active CN112199530B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011141640.6A CN112199530B (en) 2020-10-22 2020-10-22 Multi-dimensional face library picture automatic updating method, system, equipment and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011141640.6A CN112199530B (en) 2020-10-22 2020-10-22 Multi-dimensional face library picture automatic updating method, system, equipment and medium

Publications (2)

Publication Number Publication Date
CN112199530A true CN112199530A (en) 2021-01-08
CN112199530B CN112199530B (en) 2023-04-07

Family

ID=74010897

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011141640.6A Active CN112199530B (en) 2020-10-22 2020-10-22 Multi-dimensional face library picture automatic updating method, system, equipment and medium

Country Status (1)

Country Link
CN (1) CN112199530B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112650134A (en) * 2021-01-25 2021-04-13 成都优奈特科技发展有限公司 Intelligent kindergarten control management system based on Internet of things and biological recognition technology
CN112818784A (en) * 2021-01-22 2021-05-18 浙江大华技术股份有限公司 Control method and device of access control equipment and storage medium
CN112966638A (en) * 2021-03-22 2021-06-15 国网浙江省电力有限公司电力科学研究院 Transformer station operator identification and positioning method based on multiple characteristics

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104537336A (en) * 2014-12-17 2015-04-22 厦门立林科技有限公司 Face identification method and system with self-learning function
CN107403165A (en) * 2017-08-01 2017-11-28 杰创智能科技股份有限公司 The data management framework and application method of intelligent face recognition system
CN110197146A (en) * 2019-05-23 2019-09-03 招商局金融科技有限公司 Facial image analysis method, electronic device and storage medium based on deep learning
CN110619628A (en) * 2019-09-09 2019-12-27 博云视觉（北京）科技有限公司 Human face image quality evaluation method
CN111241925A (en) * 2019-12-30 2020-06-05 新大陆数字技术股份有限公司 Face quality evaluation method, system, electronic equipment and readable storage medium
CN111666443A (en) * 2020-06-03 2020-09-15 腾讯科技（深圳）有限公司 Service processing method and device, electronic equipment and computer readable storage medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104537336A (en) * 2014-12-17 2015-04-22 厦门立林科技有限公司 Face identification method and system with self-learning function
CN107403165A (en) * 2017-08-01 2017-11-28 杰创智能科技股份有限公司 The data management framework and application method of intelligent face recognition system
CN110197146A (en) * 2019-05-23 2019-09-03 招商局金融科技有限公司 Facial image analysis method, electronic device and storage medium based on deep learning
CN110619628A (en) * 2019-09-09 2019-12-27 博云视觉（北京）科技有限公司 Human face image quality evaluation method
CN111241925A (en) * 2019-12-30 2020-06-05 新大陆数字技术股份有限公司 Face quality evaluation method, system, electronic equipment and readable storage medium
CN111666443A (en) * 2020-06-03 2020-09-15 腾讯科技（深圳）有限公司 Service processing method and device, electronic equipment and computer readable storage medium

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112818784A (en) * 2021-01-22 2021-05-18 浙江大华技术股份有限公司 Control method and device of access control equipment and storage medium
CN112818784B (en) * 2021-01-22 2024-02-06 浙江大华技术股份有限公司 Control method and device of access control equipment and storage medium
CN112650134A (en) * 2021-01-25 2021-04-13 成都优奈特科技发展有限公司 Intelligent kindergarten control management system based on Internet of things and biological recognition technology
CN112966638A (en) * 2021-03-22 2021-06-15 国网浙江省电力有限公司电力科学研究院 Transformer station operator identification and positioning method based on multiple characteristics

Also Published As

Publication number Publication date
CN112199530B (en) 2023-04-07

Similar Documents

Publication Publication Date Title
CN112199530B (en) Multi-dimensional face library picture automatic updating method, system, equipment and medium
CN109284733B (en) Shopping guide negative behavior monitoring method based on yolo and multitask convolutional neural network
EP1260941A2 (en) Face image recording system
WO2019153504A1 (en) Group creation method and terminal thereof
WO2022174699A1 (en) Image updating method and apparatus, and electronic device and computer-readable medium
CN102945366A (en) Method and device for face recognition
CN113505824B (en) Judgment updating method and device and face card punching system
JP6969663B2 (en) Devices and methods for identifying the user&#39;s imaging device
JP7107598B2 (en) Authentication face image candidate determination device, authentication face image candidate determination method, program, and recording medium
CN110610125A (en) Ox face identification method, device, equipment and storage medium based on neural network
CN114387548A (en) Video and liveness detection method, system, device, storage medium and program product
CN113762106A (en) Face recognition method and device, electronic equipment and storage medium
CN112613480A (en) Face recognition method, face recognition system, electronic equipment and storage medium
CN110838102A (en) Intelligent image uploading method
CN103609098B (en) Method and apparatus for being registered in telepresence system
WO2023019927A1 (en) Facial recognition method and apparatus, storage medium, and electronic device
CN111563105A (en) Sleep management method and device, electronic equipment and storage medium
CN113409056B (en) Payment method and device, local identification equipment, face payment system and equipment
KR102594093B1 (en) Dermatologic treatment recommendation system using deep learning model and method thereof
US8879805B2 (en) Automated image identification method
CN115527241A (en) Fingerprint template updating method and device, embedded equipment and storage medium
CN112580462A (en) Feature point selection method, terminal and storage medium
CN111400526A (en) Photo processing method and device and computer readable storage medium
CN113536953B (en) Face recognition method and device, electronic equipment and storage medium
CN116259091B (en) Method and device for detecting silent living body

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant