CN107920070A - Identity identifying method, server and system - Google Patents
Identity identifying method, server and system Download PDFInfo
- Publication number
- CN107920070A CN107920070A CN201711132494.9A CN201711132494A CN107920070A CN 107920070 A CN107920070 A CN 107920070A CN 201711132494 A CN201711132494 A CN 201711132494A CN 107920070 A CN107920070 A CN 107920070A
- Authority
- CN
- China
- Prior art keywords
- head portrait
- portrait image
- identification card
- convolution kernels
- targeted customer
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/08—Network architectures or network communication protocols for network security for authentication of entities
- H04L63/0861—Network architectures or network communication protocols for network security for authentication of entities using biometrical features, e.g. fingerprint, retina-scan
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention provides a kind of identity identifying method, server and system, wherein, this method includes:In response to the operation of targeted customer's brush identification card, the head portrait image of photographic subjects user;Obtain the corresponding head portrait image of the identification card (can be identity card or the access card comprising identity information etc.);The head portrait image head portrait image corresponding with the identification card that shooting is obtained carries out similarity and compares, and determines whether similarity exceeds predetermined threshold value;In the case where determining to exceed predetermined threshold value, determine whether the corresponding user of the identification card has operating right;In the case where having determined the operating right, it is allowed to which the targeted customer performs the operation of request.Solve the problems, such as that existing authentication application is more single through the above way, reached and authentication and control of authority have been subjected to effective combination, improved the technique effect of the application range of authentication.
Description
Technical field
The present invention relates to Internet technical field, more particularly to a kind of identity identifying method, server and system.
Background technology
With the continuous development of artificial intelligence technology, more and more manually intellectual technologies are widely used.For example, intelligence
Dialogue, intelligent recognition etc..At present, intelligent identification technology is used in more and more fields, for example, the license plate number in parking lot
Intelligent recognition, fingerprint recognition, recognition of face etc..
However, for recognition of face, general existing major part recognition of face operation is also all single identification people
Whether the corresponding identity of face is stored face, for how to be combined recognition of face and identification, at present still
Effective solution is not proposed, so as to limit the application of recognition of face.
The content of the invention
An embodiment of the present invention provides a kind of identity identifying method, the mesh that is combined is determined with authority to reach identification
, this method includes:
In response to the operation of targeted customer's brush identification card, the head portrait image of photographic subjects user;
Obtain the corresponding head portrait image of the identification card;
The head portrait image head portrait image corresponding with the identification card that shooting is obtained carries out similarity and compares, and determines
Whether similarity exceeds predetermined threshold value;
In the case where determining to exceed predetermined threshold value, determine whether the corresponding user of the identification card has operating rights
Limit;
In the case where having determined the operating right, it is allowed to which the targeted customer performs the operation of request.
In one embodiment, in response to the operation of targeted customer's brush identification card, the head portrait of photographic subjects user
Image, including:
Sense to obtain the operation of targeted customer's brush identification card by NFC reading devices;
In response to the operation of targeted customer's brush identification card, triggering camera obtains the head portrait of the targeted customer
Image.
In one embodiment, will shoot obtained head portrait image head portrait image corresponding with the identification card into
Row similarity compares, and determines whether similarity exceeds predetermined threshold value, including:
Extract the feature vector of the head portrait image for shooting and obtaining;
By the spy of the feature vector head portrait image corresponding with the identification card of the head portrait image for shooting and obtaining
Sign vector is compared, to determine whether similarity exceeds predetermined threshold value.
In one embodiment, the feature vector of the head portrait image for shooting and obtaining is extracted, including:
The feature vector for shooting obtained head portrait image is extracted by neural network model trained in advance.
In one embodiment, the neural network model trained in advance includes following neural net layer successively:
9*9 convolution kernels, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 1*1 convolution
Core, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*
2 convolution kernels, 2*2 convolution kernels, 1*1 convolution kernels, 3*3 ponds window, full articulamentum, full articulamentum.
In one embodiment, by the feature vector of the head portrait image for shooting and obtaining and the identification card pair
The feature vector for the head portrait image answered is compared, including:
Calculate the feature vector head portrait image corresponding with the identification card of the head portrait image for shooting and obtaining
COS distance between feature vector;
COS distance obtained by calculation characterizes similarity.
In one embodiment, COS distance is calculated according to the following formula:
Wherein, cos (θ) represents COS distance, and a represents the feature vector of the head portrait image for shooting and obtaining, and b represents institute
The feature vector of the corresponding head portrait image of identification card is stated, | | | | represent to calculate vector field homoemorphism, represent dot product, x1 is represented
The x directions value of vectorial a, y1 represent the y directions value of vector a, and x2 represents the x directions value of vector b, and y2 represents the y of vector b
Direction value.
In one embodiment, the identification card includes at least one of:Identity card, unit access card, society
Protect card, bank card.
The embodiment of the present invention additionally provides a kind of authentication server, determines to be combined with authority to reach identification
Purpose, which includes processor and the memory for storing processor-executable instruction, and the processor performs
Realized during described instruction:
In response to the operation of targeted customer's brush identification card, the head portrait image of photographic subjects user;
Obtain the corresponding head portrait image of the identification card;
The head portrait image head portrait image corresponding with the identification card that shooting is obtained carries out similarity and compares, and determines
Whether similarity exceeds predetermined threshold value;
In the case where determining to exceed predetermined threshold value, determine whether the corresponding user of the identification card has operating rights
Limit;
In the case where having determined the operating right, it is allowed to which the targeted customer performs the operation of request.
In one embodiment, the processor response is in the operation of targeted customer's brush identification card, photographic subjects
The head portrait image of user, including:
Sense to obtain the operation of targeted customer's brush identification card by NFC reading devices;
In response to the operation of targeted customer's brush identification card, triggering camera obtains the head portrait of the targeted customer
Image.
In one embodiment, the head portrait image that the processor obtains shooting is corresponding with the identification card
Head portrait image carries out similarity comparison, determines whether similarity exceeds predetermined threshold value, including:
Extract the feature vector of the head portrait image for shooting and obtaining;
By the spy of the feature vector head portrait image corresponding with the identification card of the head portrait image for shooting and obtaining
Sign vector is compared, to determine whether similarity exceeds predetermined threshold value.
In one embodiment, the processor extraction feature vector for shooting obtained head portrait image, including:
The feature vector for shooting obtained head portrait image is extracted by neural network model trained in advance.
In one embodiment, the neural network model trained in advance includes following neural net layer successively:
9*9 convolution kernels, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 1*1 convolution
Core, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*
2 convolution kernels, 2*2 convolution kernels, 1*1 convolution kernels, 3*3 ponds window, full articulamentum, full articulamentum.
In one embodiment, the feature vector for the head portrait image that the processor obtains the shooting and the body
The feature vector of the corresponding head portrait image of part identification card is compared, including:
Calculate the feature vector head portrait image corresponding with the identification card of the head portrait image for shooting and obtaining
COS distance between feature vector;
COS distance obtained by calculation characterizes similarity.
In one embodiment, the processor calculates COS distance according to the following formula:
Wherein, cos (θ) represents COS distance, and a represents the feature vector of the head portrait image for shooting and obtaining, and b represents institute
The feature vector of the corresponding head portrait image of identification card is stated, | | | | represent to calculate vector field homoemorphism, represent dot product, x1 is represented
The x directions value of vectorial a, y1 represent the y directions value of vector a, and x2 represents the x directions value of vector b, and y2 represents the y of vector b
Direction value.
In one embodiment, the identification card includes at least one of:Identity card, unit access card, society
Protect card, bank card.
The embodiment of the present invention additionally provides a kind of identity authorization system, including:High in the clouds management platform, certificate identifier, take the photograph
As equipment, database, video computing unit, controlled device, wherein:
The high in the clouds management platform calculates the certificate identifier, the picture pick-up device, the database, the video
Data communication between unit, the controlled device is managed;
Certificate identifier, for identifying the operation of targeted customer's brush identification card;
The picture pick-up device, for the operation in response to targeted customer's brush identification card, the head portrait of photographic subjects user
Image;
The video computing unit, for obtained head portrait image head portrait figure corresponding with the identification card will to be shot
As carrying out similarity comparison, determine whether similarity exceeds predetermined threshold value;In the case where determining to exceed predetermined threshold value, institute is determined
State whether the corresponding user of identification card has operating right;In the case where having determined the operating right, it is allowed to the mesh
Mark the operation that user performs request.
The embodiment of the present invention additionally provides a kind of computer-readable recording medium, is stored thereon with computer instruction, described
Instruction is performed the step of realizing the above method.
In embodiments of the present invention, when user's brush identification card trigger action, the head portrait image of user is captured,
And the corresponding head portrait image of identification card is obtained, both are subjected to similarity comparison, to realize authentication, in authentication
Afterwards, it may further determine that whether user has operating right.Solves existing authentication application through the above way more
The problem of single, reached authentication and control of authority having carried out effective combination, improved the application range of authentication
Technique effect.
Brief description of the drawings
Attached drawing described herein is used for providing a further understanding of the present invention, forms the part of the application, not
Form limitation of the invention.In the accompanying drawings:
Fig. 1 is the method flow diagram of identity identifying method according to embodiments of the present invention;
Fig. 2 is a kind of configuration diagram of identity authorization system according to embodiments of the present invention;
Fig. 3 is the another method flow chart of identity identifying method according to embodiments of the present invention;
Fig. 4 is the model schematic of neural network model according to embodiments of the present invention;
Fig. 5 is the configuration diagram of authentication server according to embodiments of the present invention;
Fig. 6 is the structure diagram of identification authentication system according to embodiments of the present invention;
Fig. 7 is another configuration diagram of identity authorization system according to embodiments of the present invention;
Fig. 8 shows the portable front end recognition terminal of the embodiment of the present invention.
Embodiment
It is right with reference to embodiment and attached drawing for the object, technical solutions and advantages of the present invention are more clearly understood
The present invention is described in further details.Here, the exemplary embodiment and its explanation of the present invention are used to explain the present invention, but simultaneously
It is not as a limitation of the invention.
It is exactly mainly simple recognition of face in view of existing face recognition technology, if by its power with definite personage
Limit is combined, and not yet proposes effective solution at present.And the identification way of contrast of recognition of face in itself is diversity,
It is considered that can trigger the head portrait image for capturing the user after user's brush identification card and obtain the identity in this example
Both are carried out similarity comparison by the corresponding head portrait image of identification card, to realize authentication, in authentication by afterwards,
Determine whether the user has operating right, so that authentication and control of authority have been carried out effective combination, improve identity and recognize
The application range of card.
In the present specification, such as first and second adjective can be only used for by an element or action with it is another
One element or action distinguish, without requiring or implying any actual this relation or order.In the feelings that environment allows
Under condition, one in only element, component or step is should not be interpreted as limited to reference to element or component or step (s), and can
To be one or more of element, component or step etc..
As shown in Figure 1, providing a kind of identity identifying method in this example, may include steps of:
Step 101:In response to the operation of targeted customer's brush identification card, the head portrait image of photographic subjects user;
For example, application scenarios are company, towards user be company personnel, then enter company in user card punching request
When, the head portrait image of the employee can be obtained with the camera that some positions on trigger gate are set.
It should be noted, however, that above-mentioned cited application scenarios are only a kind of schematic descriptions, what is actually realized
When, it can be applied in others and realize in scene, for example, warehouse, hospital etc. need the scene that operating right controls, phase
The operation that should be triggered can also have access right, take authority etc., the application is not construed as limiting this nor only opening the door.
Step 102:Obtain the corresponding head portrait image of the identification card;
Step 103:Obtained head portrait image head portrait image progress similarity corresponding with the identification card will be shot
Compare, determine whether similarity exceeds predetermined threshold value;
In this example, using 1:1 similarity comparison mode is contrasted, i.e. can be stored with identity in the database
Identification card corresponds to the head portrait image that user prestores.Can be the head portrait image that will be captured with being deposited in system when realizing
Whether the corresponding head portrait image of the identification card of storage carries out similarity comparison, determine similarity between the two beyond default
Threshold value, if exceeding predetermined threshold value, it is determined that the user is unit personnel.
Step 104:In the case where determining to exceed predetermined threshold value, determine whether the corresponding user of the identification card has
Operating right;
In an exemplary embodiment, in the case where determining that user is company personnel, it is possible to determine the body
The corresponding operating right of part information, i.e. whether the user has permission, and has which authority, these corresponding authority informations can be
Prestore in the database, can be by the way of list, or mode of array etc. is stored, and is specifically deposited
Storage mode, the application are not especially limited.
Step 105:In the case where having determined the operating right, it is allowed to which the targeted customer performs the operation of request.
For example, targeted customer's request is to enter company gate, then after the matching, determines the targeted customer for public affairs
Employee is taken charge of, and possesses the authority that current point in time enters company gate, then can open gate.
In response to the operation of targeted customer's brush identification card, the head portrait figure of photographic subjects user in above-mentioned steps 101
Picture, can include:
S1:Sense to obtain the operation of targeted customer's brush identification card by NFC reading devices;
S2:In response to the operation of targeted customer's brush identification card, triggering camera obtains the targeted customer's
Head portrait image.
In order to realize image similarity compare, can be by way of extracting image feature vector, by feature to
The comparison of amount, realizes the comparison to similarity between image.For example, obtained head portrait image and the identification card will be shot
Corresponding head portrait image carries out similarity comparison, determines whether the similarity between image exceeds predetermined threshold value, can include:
S1:Extract the feature vector of the head portrait image for shooting and obtaining;
S2:By the feature vector head portrait image corresponding with the identification card of the head portrait image for shooting and obtaining
Feature vector is compared, to determine whether similarity exceeds predetermined threshold value.
In view of that in order to realize the extraction to the feature vector of image, can be carried by neural network model trained in advance
The feature vector of the head portrait image is taken, which can include following neural net layer successively:
9*9 convolution kernels, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 1*1 convolution kernels, 3*3 ponds
Change window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels,
2*2 convolution kernels, 1*1 convolution kernels, 3*3 ponds window, full articulamentum, full articulamentum.
When the feature vector for carrying out image compares, it can be by the COS distance between feature vector, come true
Determine the similarity between image, i.e. the feature vector of the head portrait image for shooting and obtaining and the identification can be calculated
Block the COS distance between the feature vector of corresponding head portrait image;COS distance obtained by calculation characterizes similarity.
Specifically, COS distance can be calculated by the following formula:
Wherein, cos (θ) represents COS distance, and a represents the feature vector of the head portrait image for shooting and obtaining, and b represents institute
The feature vector of the corresponding head portrait image of identification card is stated, | | | | represent to calculate vector field homoemorphism, represent dot product, x1 is represented
The x directions value of vectorial a, y1 represent the y directions value of vector a, and x2 represents the x directions value of vector b, and y2 represents the y of vector b
Direction value.
In one embodiment, according to the difference of application scenarios, identification card can include but is not limited to down toward
It is one of few:Identity card, unit access card, social security card, bank card.
Above-mentioned identity identifying method is illustrated with reference to a specific implementation, it should be noted, however, that the tool
Body embodiment does not form the improper restriction to the application merely to the application is better described.
In this example, there is provided a kind of identity authorization system, as shown in Fig. 2, can include:High in the clouds management platform, front end
Registration terminal, front end recognition hardware (such as identification terminal), database storage unit, video computing unit.
In one embodiment, high in the clouds management platform includes cloud service platform, undertake with the communication of each submodule of system and
Management function.
In one embodiment, front end registration terminal can be PC, mobile phone, tablet etc., can be logged in by account system
Management platform, which looks into the information progress additions and deletions in database, to be changed, other each submodules is managed for configuration.
In one embodiment, front end recognition hardware construction is close into that can perceive user in induction region, and/or touches
Send out camera and catch figure information, and/or read the information of authentication device.For example, front end recognition hardware may include
Authentication device read module, such as NFC read modules, reads user identity card information OR gate and prohibits card information, and handle is collected into
All information be sent to high in the clouds management platform.Video computing unit can complete portrait feature extraction, portrait alignment score etc.
Function.Database storage unit is responsible for storing userspersonal information, can include but is not limited to:Portrait data, identity information number
According to, permissions data etc..
Based on above-mentioned identity authorization system, authentication and rights management can be carried out according to step as shown in Figure 3:
S1:User's brush identity card triggering authentication compares flow;
S2:1:1 certification compares;
S3:Confirm the identity and authority information of identification employee;
S4:Comparison result is prompted;
S5:The processing of identification record.
Illustrate, may include steps of so that authority is opened at the gate of company as an example:
S1:User by the region brush ID card of swiping the card, the chip informations of sam module reading identity cards (such as:Chip shines
Piece, identity information), while the flow of testimony of a witness contrast is triggered, APP tune plays " camera ", opens camera, capture pictures, interaction
Show the identity information read.
S2:The portrait photo that " camera " is grabbed, the identity card chip photo read with " sam modules " sensing,
Algorithm is submitted to carry out characteristics extraction and compare flow, comparison score value is calculated, the comparison score value that will calculate, with
Given threshold is compared, and then thinks to compare successfully higher than threshold value.Wherein, algorithm can be server end algorithm service or
The algorithm service of client SDK.
S3:Confirm to compare successful personnel identity (that is, it is determined whether being our company personnel), confirm whether user has currently
The access right of equipment;
S4:APP showing interface comparison result contents, such as:Success prompting, stranger's prompting, capture pictures, personnel's name,
Welcome words etc..When contrasting successfully, voice prompt " compare and successfully, welcome to enter " simultaneously sends door open command to the control module of door.
S5:The record information of generation is returned to server end by APP ends, generates the identification record of standardization, and allows to lead to
Later platform such as is managed and checks at the operation.
Illustrated with reference to an instantiation to how to carry out image comparison, specifically, can be by the figure of input
As being cut to the 3 passage RGB pictures that resolution ratio is 227*227 as required, at a series of intermediate layer as shown in Figure 4
Reason (mainly convolution and pondization processing), finally obtains the feature vector of image, and any two portrait pictures pass through algorithm process
Obtain feature vector and can be used for similarity measure.Exemplified by inputting picture size and be 227*227*3, in god as shown in Figure 4
Each layer through network, can be performed according to following rule, to extract the feature vector of image:
Level 1 volume lamination:Core size is 9, port number 3;Output characteristic figure size is 55*55, port number 96;
1st pond layer:Core size is 3;Output characteristic figure size is 27*27, port number 96;
2_1 layers of convolutional layer:Core size is 2, port number 96;Output characteristic figure size is 28*28, port number 128
2_2 layers of convolutional layer:Core size is 2, port number 128;Output characteristic figure size is 27*27, port number 128
2_3 layers of convolutional layer:Core size is 2, port number 128;Output characteristic figure size is 28*28, port number 192
2_4 layers of convolutional layer:Core size is 2, port number 192;Output characteristic figure size is 27*27, passage numerical digit 192
2_5 layers of convolutional layer:Core size is 1, port number 192;Output characteristic figure size is 27*27, port number 192
2nd layer of pond layer:Core size is 3;Output characteristic figure size is 13*13, port number 192;
3_1 layers of convolutional layer:Core size is 2, port number 192;Output characteristic figure size is 14*14, and port number is
256;
3_2 layers of convolutional layer:Core size is 2, port number 256;Output characteristic figure size is 13*13, and port number is
256;
4_1 layers of convolutional layer:Core size is 2, port number 256;Output characteristic figure size is 14*14, and port number is
256;
4_2 layers of convolutional layer:Core size is 2, port number 256;Output characteristic figure size is 13*13, and port number is
256;
4_3 layers of convolutional layer:Core size is 2, port number 256;Output characteristic figure size is 14*14, and port number is
320;
4_4 layers of convolutional layer:Core size is 2, port number 320;Output characteristic figure size is 13*13, and port number is
320;
5_1 layers of convolutional layer:Core size is 2, port number 320;Output characteristic figure size is 14*14, and port number is
384;
5_2 layers of convolutional layer:Core size is 2, port number 384;Output characteristic figure size is 13*13, and port number is
384;
5_3 layers of convolutional layer:Core size is 1, port number 384;Output characteristic figure size is 13*13, and port number is
384;
3rd layer of pond layer:Core size is 3;Output characteristic figure size is 6*6, port number 384;
First full articulamentum:Output characteristic dimension is 4096;
Second full articulamentum:Output characteristic number of dimensions is 2048;
Wherein, convolutional layer can extract characteristic pattern according to the following formula:
Fij=relu (BN ((W*x) ij+b))
So, two photos (for example, one is head portrait image, one is source images) are inputted into above-mentioned neutral net,
Output is exactly two feature vectors, and COS distance can be calculated based on the two feature vectors, specifically, can be by following public
Formula calculates COS distance:
Wherein, cos (θ) represents COS distance, and a represents the feature vector of the head portrait image for shooting and obtaining, and b represents institute
The feature vector of the corresponding head portrait image of identification card is stated, | | | | represent to calculate vector field homoemorphism, represent dot product, x1 is represented
The x directions value of vectorial a, y1 represent the y directions value of vector a, and x2 represents the x directions value of vector b, and y2 represents the y of vector b
Direction value.
It should be noted, however, that upper example is only a kind of schematic description, the Rotating fields in neutral net can be according to reality
Border accuracy requirement is adjusted, and the corresponding head portrait image of above-mentioned identification card can be that number is stored in a manner of feature vector
According in storehouse, when in use, directly reading feature vector can or store in the database in the form of images,
The feature vector of extract real-time image when in use, can specifically be chosen, this Shen using which kind of mode according to being actually needed
Please this is not construed as limiting.
Further, the similarity of two portraits can be obtained based on COS distance, when similarity threshold is set,
The threshold value at one thousandth misclassification rate can be used.That is, a threshold value is taken at interval of a step-length (0 to 1), when mistake is matched
(such as:1000 couples of people, then the number of mistake pairing is exactly 1000*1000-1000) in, it is considered to be same person (that is, is sentenced
Dislocation misses) similarity of paired data when reaching 0.1% seek to the threshold value of selection, then think that two people are same more than threshold value
One people, less than threshold value, then it is assumed that be not a people.Specifically, the selection of this threshold value can be by largely testing to obtain
's.
Fig. 5 shows the schematic configuration diagram based on server side of the exemplary embodiment according to the application.It refer to
Fig. 5, in hardware view, which includes processor, internal bus, network interface, memory and non-volatile
Memory, is also possible that the required hardware of other business certainly.Processor reads corresponding from nonvolatile memory
Computer program is into memory and then runs, and identification authentication system is formed on logic level.Certainly, except software realization mode
Outside, the application is not precluded from other implementations, such as mode of logical device or software and hardware combining etc., that is to say, that
The executive agent of following process flow is not limited to each logic unit or hardware or logical device.
Fig. 6 is refer to, in Software Implementation, which is applied to cloud service platform either terminal
In, it can include:Taking module, acquisition module, comparing module, determining module and control module.Wherein:
Taking module, for the operation in response to targeted customer's brush identification card, the head portrait image of photographic subjects user;
Acquisition module, for obtaining the corresponding head portrait image of the identification card;
Comparing module, phase is carried out for will shoot obtained head portrait image head portrait image corresponding with the identification card
Compared like degree, determine whether similarity exceeds predetermined threshold value;
Determining module, in the case where determining to exceed predetermined threshold value, determining the corresponding user of the identification card
Whether operating right is had;
Control module, in the case where having determined the operating right, it is allowed to which the targeted customer performs request
Operation.
In one embodiment, taking module can specifically sense to obtain the targeted customer by NFC reading devices
The operation of brush identification card;In response to the operation of targeted customer's brush identification card, triggering camera obtains the mesh
Mark the head portrait image of user.
In one embodiment, comparing module can specifically extract it is described shoot the obtained feature of head portrait image to
Amount;By the feature vector of the feature vector head portrait image corresponding with the identification card of the head portrait image for shooting and obtaining
It is compared, to determine whether similarity exceeds predetermined threshold value.
In one embodiment, the feature of the head portrait image can be extracted by neural network model trained in advance
Vector.
In one embodiment, neural network model trained in advance can include following neural net layer successively:9*9
Convolution kernel, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 1*1 convolution kernels, 3*3 ponds
Window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*
2 convolution kernels, 1*1 convolution kernels, 3*3 ponds window, full articulamentum, full articulamentum.
In one embodiment, by the feature vector of the head portrait image for shooting and obtaining and the identification card pair
The feature vector for the head portrait image answered is compared, and can be included:Calculate the feature vector of the head portrait image for shooting and obtaining
COS distance between the feature vector of head portrait image corresponding with the identification card;COS distance obtained by calculation
Characterize similarity.
Specifically, COS distance can be calculated according to the following formula:
Wherein, cos (θ) represents COS distance, and a represents the feature vector of the head portrait image for shooting and obtaining, and b represents institute
The feature vector of the corresponding head portrait image of identification card is stated, | | | | represent to calculate vector field homoemorphism, represent dot product, x1 is represented
The x directions value of vectorial a, y1 represent the y directions value of vector a, and x2 represents the x directions value of vector b, and y2 represents the y of vector b
Direction value.
In one embodiment, identification card can include but is not limited at least one of:Identity card, unit door
Prohibit card, social security card, bank card.
In this example, a kind of identity authorization system is additionally provided, as shown in fig. 7, can be identified with high in the clouds management platform, certificate
Device, picture pick-up device, database, video computing unit, controlled device, wherein:
The high in the clouds management platform calculates the certificate identifier, the picture pick-up device, the database, the video
Data communication between unit, the controlled device is managed;
Certificate identifier, for identifying the operation of targeted customer's brush identification card;
The picture pick-up device, for the operation in response to targeted customer's brush identification card, the head portrait of photographic subjects user
Image;
The video computing unit, for obtained head portrait image head portrait figure corresponding with the identification card will to be shot
As carrying out similarity comparison, determine whether similarity exceeds predetermined threshold value;In the case where determining to exceed predetermined threshold value, institute is determined
State whether the corresponding user of identification card has operating right;In the case where having determined the operating right, it is allowed to the mesh
Mark the operation that user performs request.
In one embodiment, there is provided for the identification device of authentication, such as portable front end recognition terminal.Should
Identification equipment can include one or more of above-mentioned identity authorization system component, as picture pick-up device, identifier, video calculate
Unit.The identification device or identification terminal can also include other devices, such as sensor, such as position sensor.
Such as Fig. 8 shows one embodiment of identification terminal.The identification terminal 800 may include in front side camera 801,
Sensor, such as position sensor 802, display or viewing area 803 and identifier or cog region, such as NFC identifiers or identification
Area 804 or Quick Response Code or RF tag identifier.It is envisioned that the identification terminal can include other components or mould
Block, such as wired or wireless interface, for the high in the clouds management platform with above-mentioned identity authorization system, database and/or regard
Frequency meter calculates unit communication.Such as the identification terminal may include image transmission interface and/or the information of identity or operating right information
Receiving interface.In one embodiment, which may include computing unit, which may include to be integrated with advance instruction
The neural network chip that experienced neural network model is contrasted with the feature vector to acquired head portrait image and source images.
The chip is for example integrated with the neural network model of this disclosure.
In another embodiment, a kind of software is additionally provided, which is used to perform above-described embodiment and preferred reality
Apply the technical solution described in mode.
In another embodiment, a kind of storage medium is additionally provided, above-mentioned software is stored with the storage medium, should
Storage medium includes but not limited to:CD, floppy disk, hard disk, scratch pad memory etc..
It can be seen from the above description that the embodiment of the present invention realizes following technique effect:Know in user's brush identity
Not card trigger action when, capture the head portrait image of user, and obtain the corresponding head portrait image of identification card, will both into
Row similarity compares, and to realize authentication, after authentication, may further determine that whether user has operating right.
Solve the problems, such as that existing authentication application is more single through the above way, reached by authentication and control of authority into
Go effective combination, improve the technique effect of the application range of authentication.
Obviously, those skilled in the art should be understood that each module of the above-mentioned embodiment of the present invention or each step can be with
Realized with general computing device, they can be concentrated on single computing device, or are distributed in multiple computing devices
On the network formed, alternatively, they can be realized with the program code that computing device can perform, it is thus possible to by it
Store and performed in the storage device by computing device, and in some cases, can be to be held different from order herein
They, are either fabricated to each integrated circuit modules or will be multiple in them by the shown or described step of row respectively
Module or step are fabricated to single integrated circuit module to realize.In this way, the embodiment of the present invention be not restricted to it is any specific hard
Part and software combine.
The foregoing is only a preferred embodiment of the present invention, is not intended to limit the invention, for the skill of this area
For art personnel, the embodiment of the present invention can have various modifications and variations.Within the spirit and principles of the invention, made
Any modification, equivalent substitution, improvement and etc., should all be included in the protection scope of the present invention.
Claims (10)
- A kind of 1. identity identifying method, it is characterised in that including:In response to the operation of targeted customer's brush identification card, the head portrait image of photographic subjects user;Obtain the corresponding head portrait image of the identification card;The head portrait image head portrait image corresponding with the identification card that shooting is obtained carries out similarity and compares, and determines similar Whether degree exceeds predetermined threshold value;In the case where determining to exceed predetermined threshold value, determine whether the corresponding user of the identification card has operating right;In the case where having determined the operating right, it is allowed to which the targeted customer performs the operation of request.
- 2. according to the method described in claim 1, it is characterized in that, operation in response to targeted customer's brush identification card, is clapped The head portrait image of targeted customer is taken the photograph, including:Sense to obtain the operation of targeted customer's brush identification card by NFC reading devices;In response to the operation of targeted customer's brush identification card, triggering camera obtains the head portrait figure of the targeted customer Picture.
- 3. according to the method described in claim 1, it is characterized in that, obtained head portrait image and the identification card will be shot Corresponding head portrait image carries out similarity comparison, determines whether similarity exceeds predetermined threshold value, including:Extract the feature vector of the head portrait image for shooting and obtaining;By the feature of the feature vector head portrait image corresponding with the identification card of head portrait image for shooting and obtaining to Amount is compared, to determine whether similarity exceeds predetermined threshold value.
- 4. according to the method described in claim 3, it is characterized in that, extraction it is described shoot the obtained feature of head portrait image to Amount, including:The feature vector for shooting obtained head portrait image is extracted by neural network model trained in advance.
- 5. according to the method described in claim 4, it is characterized in that, the neural network model trained in advance is included such as successively Lower neural net layer:9*9 convolution kernels, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 1*1 convolution kernels, 3*3 ponds window, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 convolution kernels, 2*2 volumes Product core, 2*2 convolution kernels, 1*1 convolution kernels, 3*3 ponds window, full articulamentum, full articulamentum.
- 6. according to the method described in claim 3, it is characterized in that, by it is described shoot the obtained feature vector of head portrait image with The feature vector of the corresponding head portrait image of the identification card is compared, including:Calculate the feature of the feature vector head portrait image corresponding with the identification card of the head portrait image for shooting and obtaining COS distance between vector;COS distance obtained by calculation characterizes similarity.
- 7. according to the method described in claim 6, it is characterized in that, calculate COS distance according to the following formula:<mrow> <mi>c</mi> <mi>o</mi> <mi>s</mi> <mrow> <mo>(</mo> <mi>&theta;</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mi>a</mi> <mo>&CenterDot;</mo> <mi>b</mi> </mrow> <mrow> <mo>|</mo> <mo>|</mo> <mi>a</mi> <mo>|</mo> <mo>|</mo> <mo>&times;</mo> <mo>|</mo> <mo>|</mo> <mi>b</mi> <mo>|</mo> <mo>|</mo> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <mi>x</mi> <mn>1</mn> <mi>x</mi> <mn>2</mn> <mo>+</mo> <mi>y</mi> <mn>1</mn> <mi>y</mi> <mn>2</mn> </mrow> <mrow> <msqrt> <mrow> <mi>x</mi> <msup> <mn>1</mn> <mn>2</mn> </msup> <mo>+</mo> <mi>y</mi> <msup> <mn>1</mn> <mn>2</mn> </msup> </mrow> </msqrt> <mo>&times;</mo> <msqrt> <mrow> <mi>x</mi> <msup> <mn>2</mn> <mn>2</mn> </msup> <mo>+</mo> <mi>y</mi> <msup> <mn>2</mn> <mn>2</mn> </msup> </mrow> </msqrt> </mrow> </mfrac> </mrow>Wherein, cos (θ) represents COS distance, and a represents the feature vector of the head portrait image for shooting and obtaining, and b represents the body The feature vector of the corresponding head portrait image of part identification card, | | | | represent to calculate vector field homoemorphism, represent dot product, x1 represents vectorial The x directions value of a, y1 represent the y directions value of vector a, and x2 represents the x directions value of vector b, and y2 represents the y directions of vector b Value.
- 8. method according to any one of claim 1 to 7, it is characterised in that the identification card include with down toward It is one of few:Identity card, unit access card, social security card, bank card.
- 9. a kind of authentication server, including processor and the memory for storing processor-executable instruction, described Processor is realized when performing described instruction:In response to the operation of targeted customer's brush identification card, the head portrait image of photographic subjects user;Obtain the corresponding head portrait image of the identification card;The head portrait image head portrait image corresponding with the identification card that shooting is obtained carries out similarity and compares, and determines similar Whether degree exceeds predetermined threshold value;In the case where determining to exceed predetermined threshold value, determine whether the corresponding user of the identification card has operating right;In the case where having determined the operating right, it is allowed to which the targeted customer performs the operation of request.
- 10. server according to claim 9, it is characterised in that the processor response is known in targeted customer's brush identity The operation not blocked, the head portrait image of photographic subjects user, including:Sense to obtain the operation of targeted customer's brush identification card by NFC reading devices;In response to the operation of targeted customer's brush identification card, triggering camera obtains the head portrait figure of the targeted customer Picture.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711132494.9A CN107920070A (en) | 2017-11-15 | 2017-11-15 | Identity identifying method, server and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711132494.9A CN107920070A (en) | 2017-11-15 | 2017-11-15 | Identity identifying method, server and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107920070A true CN107920070A (en) | 2018-04-17 |
Family
ID=61895646
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711132494.9A Pending CN107920070A (en) | 2017-11-15 | 2017-11-15 | Identity identifying method, server and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107920070A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109462579A (en) * | 2018-10-22 | 2019-03-12 | 维沃移动通信有限公司 | A kind of auth method and terminal device |
CN109495727A (en) * | 2019-01-04 | 2019-03-19 | 京东方科技集团股份有限公司 | Intelligent control method and device, system, readable storage medium storing program for executing |
CN109544367A (en) * | 2018-10-27 | 2019-03-29 | 平安医疗健康管理股份有限公司 | A kind of medical insurance card based on data processing is swiped the card processing method and relevant device |
CN109598630A (en) * | 2018-12-04 | 2019-04-09 | 平安医疗健康管理股份有限公司 | The monitoring and managing method and relevant device of fixed point medical institution based on living things feature recognition |
CN110390257A (en) * | 2019-05-30 | 2019-10-29 | 郑州中软高科信息技术有限公司 | A kind of quick verifying system of identity information |
CN111241884A (en) * | 2018-11-29 | 2020-06-05 | 中科天网(广东)科技有限公司 | Cloud architecture shared face recognition system |
CN112102085A (en) * | 2020-08-31 | 2020-12-18 | 深圳市富途网络科技有限公司 | Option recommendation method, apparatus, device and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103559485A (en) * | 2013-11-08 | 2014-02-05 | 广州广电运通金融电子股份有限公司 | Identity authentication method and device based on remote bank VTM |
CN103942630A (en) * | 2013-01-21 | 2014-07-23 | 神讯电脑(昆山)有限公司 | Employee information managing system |
CN104734852A (en) * | 2013-12-24 | 2015-06-24 | 中国移动通信集团湖南有限公司 | Identity authentication method and device |
CN105809111A (en) * | 2016-02-25 | 2016-07-27 | 曹宇 | Identity authentication method, apparatus, device and system |
CN106022317A (en) * | 2016-06-27 | 2016-10-12 | 北京小米移动软件有限公司 | Face identification method and apparatus |
US20160364559A1 (en) * | 2015-06-09 | 2016-12-15 | Intel Corporation | Secure biometric data capture, processing and management |
CN106778607A (en) * | 2016-12-15 | 2017-05-31 | 国政通科技股份有限公司 | A kind of people based on recognition of face and identity card homogeneity authentication device and method |
CN106909821A (en) * | 2017-01-20 | 2017-06-30 | 奇酷互联网络科技(深圳)有限公司 | The control method of terminal, device and terminal device |
-
2017
- 2017-11-15 CN CN201711132494.9A patent/CN107920070A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942630A (en) * | 2013-01-21 | 2014-07-23 | 神讯电脑(昆山)有限公司 | Employee information managing system |
CN103559485A (en) * | 2013-11-08 | 2014-02-05 | 广州广电运通金融电子股份有限公司 | Identity authentication method and device based on remote bank VTM |
CN104734852A (en) * | 2013-12-24 | 2015-06-24 | 中国移动通信集团湖南有限公司 | Identity authentication method and device |
US20160364559A1 (en) * | 2015-06-09 | 2016-12-15 | Intel Corporation | Secure biometric data capture, processing and management |
CN105809111A (en) * | 2016-02-25 | 2016-07-27 | 曹宇 | Identity authentication method, apparatus, device and system |
CN106022317A (en) * | 2016-06-27 | 2016-10-12 | 北京小米移动软件有限公司 | Face identification method and apparatus |
CN106778607A (en) * | 2016-12-15 | 2017-05-31 | 国政通科技股份有限公司 | A kind of people based on recognition of face and identity card homogeneity authentication device and method |
CN106909821A (en) * | 2017-01-20 | 2017-06-30 | 奇酷互联网络科技(深圳)有限公司 | The control method of terminal, device and terminal device |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109462579A (en) * | 2018-10-22 | 2019-03-12 | 维沃移动通信有限公司 | A kind of auth method and terminal device |
CN109544367A (en) * | 2018-10-27 | 2019-03-29 | 平安医疗健康管理股份有限公司 | A kind of medical insurance card based on data processing is swiped the card processing method and relevant device |
CN111241884A (en) * | 2018-11-29 | 2020-06-05 | 中科天网(广东)科技有限公司 | Cloud architecture shared face recognition system |
CN109598630A (en) * | 2018-12-04 | 2019-04-09 | 平安医疗健康管理股份有限公司 | The monitoring and managing method and relevant device of fixed point medical institution based on living things feature recognition |
WO2020114221A1 (en) * | 2018-12-04 | 2020-06-11 | 平安医疗健康管理股份有限公司 | Biometric recognition-based designated medical institution supervision method and related device |
CN109495727A (en) * | 2019-01-04 | 2019-03-19 | 京东方科技集团股份有限公司 | Intelligent control method and device, system, readable storage medium storing program for executing |
CN110390257A (en) * | 2019-05-30 | 2019-10-29 | 郑州中软高科信息技术有限公司 | A kind of quick verifying system of identity information |
CN112102085A (en) * | 2020-08-31 | 2020-12-18 | 深圳市富途网络科技有限公司 | Option recommendation method, apparatus, device and storage medium |
CN112102085B (en) * | 2020-08-31 | 2023-11-03 | 深圳市富途网络科技有限公司 | Option recommendation method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107920070A (en) | Identity identifying method, server and system | |
US10438053B2 (en) | Biometric identification systems and methods | |
CN107944238A (en) | Identity identifying method, server and system | |
US9773151B2 (en) | System and methods for contactless biometrics-based identification | |
US20190124079A1 (en) | System and method for biometric authentication in connection with camera-equipped devices | |
WO2019114376A1 (en) | Document verification method, device, electronic device, and storage medium | |
CN102509053A (en) | Authentication and authorization method, processor, equipment and mobile terminal | |
US11716330B2 (en) | Mobile enrollment using a known biometric | |
US10963552B2 (en) | Method and electronic device for authenticating a user | |
EP2148295A1 (en) | Vein pattern management system, vein pattern registration device, vein pattern authentication device, vein pattern registration method, vein pattern authentication method, program, and vein data structure | |
KR20170056193A (en) | Access authentication system and method | |
KR20210004319A (en) | Method, apparatus and system for performing authentication using face recognition | |
CN111695099A (en) | Security system and corresponding access method, machine readable storage device | |
CN106780864A (en) | Gate control system based on skeleton identification technology | |
JP6311237B2 (en) | Collation device and collation method, collation system, and computer program | |
US20150120543A1 (en) | EyeWatch ATM and Wire Transfer Fraud Prevention System | |
CN107395369A (en) | Towards mobile Internet from the authentication method of carrying device, access method and system | |
CN208077174U (en) | Identification device for authentication | |
US20150100493A1 (en) | EyeWatch credit card fraud prevention system | |
JP2018169943A (en) | Face authentication processing device, face authentication processing method and face authentication processing system | |
KR102539533B1 (en) | Method and apparatus for preventing other people from photographing identification | |
JP6911999B2 (en) | Admission management system | |
Selvakumar et al. | Face Biometric Authentication System for ATM using Deep Learning | |
More et al. | Design and implementation of anti-theft module for ATM machine | |
JP7238617B2 (en) | Service providing device, service providing system, program, and service providing method |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180417 |