CN105894262A - Customer authentication system based on image identification - Google Patents
Customer authentication system based on image identification Download PDFInfo
- Publication number
- CN105894262A CN105894262A CN201610204738.9A CN201610204738A CN105894262A CN 105894262 A CN105894262 A CN 105894262A CN 201610204738 A CN201610204738 A CN 201610204738A CN 105894262 A CN105894262 A CN 105894262A
- Authority
- CN
- China
- Prior art keywords
- image
- face
- equipment
- supercilium
- identity
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/08—Payment architectures
- G06Q20/20—Point-of-sale [POS] network systems
- G06Q20/206—Point-of-sale [POS] network systems comprising security or operator identification provisions, e.g. password entry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
- G06Q20/40145—Biometric identity checks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- Computer Security & Cryptography (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention relates to a customer authentication system based on image identification, comprising an image acquisition device, an image identification device, and a master control device. The image acquisition device is used for acquiring an image of a customer. The image identification device is connected with the image acquisition device, and is used for detecting a face sub image from the image of the customer and authenticating the identity of the customer based on the face sub image. The master control device is connected with the image identification device, and is used for determining a corresponding payment strategy based on the result of customer identity authentication. With the system, a customer can be identified accurately, and consumption and payment operation of a customer can be completed quickly.
Description
Technical field
The present invention relates to field of image recognition, particularly relate to a kind of client of based on image recognition certification system
System.
Background technology
From the point of view of technically, face identification system mainly includes four ingredients, is respectively as follows: face
Image acquisition and detection, facial image pretreatment, facial image feature extraction and coupling with identify.
Man face image acquiring: different facial images can be transferred through pick-up lens and collects, the most quiet
The aspects such as state image, dynamic image, different positions, different expressions can well be gathered.
When user is in the coverage of collecting device, collecting device can search for and shoot the people of user automatically
Face image.
Face datection: Face datection is mainly used in the pretreatment of recognition of face in practice, i.e. at image
Middle accurate calibration goes out position and the size of face.The pattern feature comprised in facial image is the abundantest,
Such as histogram feature, color characteristic, template characteristic, architectural feature and Haar feature etc..Face datection
It is exactly that the most useful information is picked out, and utilizes these features to realize Face datection.
The method for detecting human face of main flow uses Adaboost learning algorithm, Adaboost based on features above
Algorithm is a kind of method for classifying, and it is combined the sorting technique that some are more weak, combination
The strongest sorting technique made new advances.
Adaboost algorithm is used to pick out some rectangles that can represent face during Face datection
Feature (Weak Classifier), is configured to a strong classifier according to the mode of Nearest Neighbor with Weighted Voting by Weak Classifier,
Some strong classifiers training obtained again are composed in series the cascade filtering of a cascade structure, effectively
Ground improves the detection speed of grader.
Facial image pretreatment: the Image semantic classification for face is based on Face datection result, to figure
As carrying out processing and finally serving the process of feature extraction.The original image that system obtains is owing to being subject to
The restriction of various conditions and random disturbances, tend not to directly use, it is necessary to image procossing early stage
Stage carries out the Image semantic classification such as gray correction, noise filtering to it.For facial image, its
Preprocessing process mainly includes the light compensation of facial image, greyscale transformation, histogram equalization, returns
One change, geometric correction, filter and sharpening etc..
Facial image feature extraction: the spendable feature of face identification system be generally divided into visual signature,
Pixels statistics feature, facial image conversion coefficient feature, facial image algebraic characteristic etc..Face characteristic
Extract and be aiming at what some feature of face was carried out.Face characteristic extracts, and also referred to as face characterizes, it
It it is the process that face is carried out feature modeling.The method that face characteristic extracts be summed up be divided into two big
Class: one is Knowledge based engineering characterizing method；Another is based on algebraic characteristic or statistical learning
Characterizing method.
Knowledge based engineering characterizing method is mainly according between the shape description of human face and they
Range performance obtain the characteristic contributing to face classification, its characteristic component generally includes feature
Euclidean distance, curvature and angle etc. between point.Face is by local structures such as eyes, nose, mouth, chins
Become, to these local and the geometric description of structural relation between them, can be as identifying the important of face
Feature, these features are referred to as geometric properties.Knowledge based engineering face characterizes and mainly includes based on geometry
The method of feature and template matching method.
Facial image coupling and identification: the characteristic of the facial image of extraction and storage in data base
Feature templates scans for coupling, by setting a threshold value, when similarity exceedes this threshold value, then
The result output that coupling is obtained.Recognition of face is exactly by face characteristic to be identified and the people obtained
Face feature templates compares, and judges the identity information of face according to similarity degree.This mistake
Journey is divided into again two classes: a class is to confirm, is the process that image compares that carries out one to one, and another kind of is to distinguish
Recognize, be the one-to-many process that carries out images match contrast.
It is by face recognition application to payment platform that recognition of face pays, and is carried by customer face feature
Take, confirm customer identification, and the spending amount coordinating cashier to provide completes customer payment process.Whole
Individual payment process automaticity is high, and artificial interference link is few, for too much queuing checkout client
Time period, its effect is especially pronounced.
But, existing recognition of face payments mechanism there is also several drawback.First, identify and pay ring
Save the most simple and crude, lack the secondary authentication mechanism of necessity；Second, at queuing checkout, client is too much
Time, non-checkout client may be carried out facial recognition and payment；3rd, recognition of face itself still has and changes
The space of kind performance.
Accordingly, it would be desirable to a kind of new facial payment devices for client's certification, existing face is known
Other payments mechanism improves, and increases assistant authentification means and client selects link, and to face characteristic
The mode of detection is optimized, thus improves the safety and reliability of face payment platform comprehensively.
Summary of the invention
In order to solve the problems referred to above, the invention provides a kind of client of based on image recognition certification system
System, by increasing capacitance it is possible to increase secondary identities authenticating device, after facial recognition identity, uses secondary identities certification
Equipment carries out further identity validation, provides checkout client to select pattern for cashier simultaneously, it is to avoid
The situation that other clients are settled accounts by mistake occurs, and pays it addition, also optimize existing facial recognition, protects
Demonstrate,prove the accuracy of facial recognition.
According to an aspect of the present invention, it is provided that a kind of client's Verification System based on image recognition, institute
The system of stating includes image capture device, image recognition apparatus and main control device, and image capture device is used for
Gathering the image of client, image recognition apparatus is connected with image capture device, for from the image of client
In detect face's subimage and carry out customer identification certification, main control device and figure based on face's subimage
As identifying that equipment connects, pay strategy accordingly for determining based on customer identification authentication result.
More specifically, in described client's Verification System based on image recognition, including: double track is raised
Sound device, is connected with AVR32 chip, for playing and identity validation failure signal or identity validation success
The voice prompted file that signal is corresponding；Supercilium splitting equipment, puts down with static storage device and Gauss respectively
Sliding filter apparatus connects to receive preset reference supercilium profile and image to be identified, based on preset reference eyebrow
Contouring is partitioned into supercilium subimage in image to be identified；Mouth splitting equipment, deposits with static state respectively
Storage equipment and Gaussian smoothing filter equipment connect to receive preset reference mouth profile and image to be identified,
In image to be identified, mouth subimage it is partitioned into based on preset reference mouth profile；Eye segmentation sets
Standby, it is connected with static storage device and Gaussian smoothing filter equipment respectively to receive preset reference eye wheel
Image wide and to be identified, is partitioned into eye subgraph based on preset reference eye profile in image to be identified
Picture；Feature analysis equipment, respectively with supercilium splitting equipment, mouth splitting equipment and eye splitting equipment
Connect, determine supercilium thickness and supercilium flexibility based on supercilium subimage, determine based on mouth subimage
Mouth thickness, determines eye opening degree based on eye subimage；Characteristic matching equipment, respectively with static state
The face recognition data storehouse of storage device, feature analysis equipment and far-end connects, face recognition data storehouse
The supercilium thickness of everyone face-image, supercilium flexibility, mouth thickness and eye are prestored
Portion's opening degree, the supercilium thickness exported by feature analysis equipment and face recognition data storehouse prestore
The supercilium thickness of everyone face-image carries out mating to obtain supercilium thickness matching degree, by feature
Everyone face that the supercilium flexibility of analytical equipment output prestores with face recognition data storehouse
The supercilium flexibility of portion's image carries out mating to obtain supercilium flexibility matching degree, by feature analysis equipment
The mouth of everyone face-image that the mouth thickness of output prestores with face recognition data storehouse
Portion's thickness carries out mating to obtain mouth thickness matching degree, the eye folding exported by feature analysis equipment
The eye opening degree spending everyone face-image prestored with face recognition data storehouse is carried out
Coupling, to obtain eye opening degree matching degree, based on supercilium thickness matching degree, presets supercilium thickness weight
Value, supercilium flexibility matching degree, preset supercilium flexibility weighted value, mouth thickness matching degree, preset
Mouth thickness weighted value, eye opening degree matching degree and default eye opening degree weighted value determine to be identified
The matching degree of everyone face-image that image and face recognition data storehouse prestore, general
Join piece identity corresponding to the highest face-image of degree as identifying identity output；High-definition image gathers
Equipment, is arranged on above POS, for shooting to obtain high definition people to the crowd of queuing checkout
Group's image；Face detection equipment, is connected with static storage device and high-definition image collecting device respectively,
For receiving high definition crowd's image and preset reference face contour, based on preset reference face contour at height
Clear crowd's image matches multiple face subimage；Display device, with face detection equipment be connected with
Receiving and show multiple face subimage, display device is also with touch screen, with based on cashier defeated
Enter and select target face subimage from multiple face subimages；Geometrical correction device, respectively with face
Detection equipment and display device connect, and receive target face subimage and carry out target face subimage
Geometric correction processes to obtain geometric correction image；Image orbiting facility, is connected with geometrical correction device
To receive geometric correction image, geometric correction image is carried out image rotation and processes to obtain rotation figure
Picture；Image translation device, is connected with image orbiting facility and rotates image to receive, and enters rotating image
Row image translation processes to obtain displacement images；Image segmentation apparatus, respectively with static storage device and
Image translation device connects, and is used for receiving pre-set image block size and displacement images, enters displacement images
Row segmentation is to obtain the segmentation image of pre-set image block size；Histogram equalization equipment, divides with image
The equipment that cuts connects, and is used for receiving segmentation image and segmentation image being carried out histogram equalization process, with
Obtain the grey level histogram of segmentation image；Gaussian smoothing filter equipment, with histogram equalization equipment even
Connect, be used for receiving grey level histogram and grey level histogram being carried out Gaussian smoothing filter process, to obtain
Image to be identified；Static storage device, is used for having prestored preset reference face contour and having preset figure
As block size, it is additionally operable to prestore default supercilium thickness weighted value, preset supercilium flexibility weight
Value, preset mouth thickness weighted value, preset eye opening degree weighted value, preset reference supercilium profile,
Preset reference mouth profile and preset reference eye profile；Wireless Telecom Equipment, is used for passing through channel radio
Connection between letter link establishment characteristic matching equipment and the face recognition data storehouse of far-end, is additionally operable to lead to
Cross the connection that wireless communication link is set up between fingerprint matching device and the fingerprint database of far-end；Fingerprint
Reception equipment, for gathering the finger print information of client；Fingerprint matching device, receives with fingerprint respectively and sets
Standby and far-end fingerprint database connects, and the fingerprint that fingerprint database has prestored everyone is special
Levy, receive the finger print information of equipment output based on fingerprint and find the fingerprint spy of coupling at fingerprint database
Levy, and using piece identity corresponding for the fingerprint characteristic of coupling as confirming identity output；AVR32 chip,
Receive equipment with display device, characteristic matching equipment, fingerprint respectively and fingerprint matching device is connected, when
When receiving identification identity at characteristic matching equipment, the client to current checkout starts fingerprint reception and sets
Standby and fingerprint matching device confirms identity to receive, and the client of current checkout is in the crowd of queuing checkout
In, when identifying identity with when confirming that identity is consistent, identification identity and cashier are passed through display device
The value data of touch screen input be sent collectively to the payment devices of far-end to complete to pay, simultaneously to
Identify the E-mail address transmission confirmation mail that identity is corresponding, confirm that mail includes customer payment video；
Wherein, preset reference face contour is that benchmark face-image is carried out the figure that contours extract obtains,
Pre-set image block size elects 45 pixel × 50 pixels as；High-definition image collecting device is additionally operable to record and turns round and look at
Visitor pays video.
More specifically, in described client's Verification System based on image recognition: preset reference supercilium is taken turns
The wide figure obtained for benchmark supercilium image is carried out contours extract.
More specifically, in described client's Verification System based on image recognition: preset reference mouth is taken turns
The wide figure obtained for benchmark mouth image is carried out contours extract.
More specifically, in described client's Verification System based on image recognition: preset reference eye is taken turns
The wide figure obtained for benchmark eyes image is carried out contours extract.
More specifically, in described client's Verification System based on image recognition: AVR32 chip is
When not receiving identification identity after one Preset Time, send identity validation failure signal, AVR32 chip
When not receiving confirmation identity after the second Preset Time, send identity validation failure signal, AVR32
Chip, when identifying that identity is inconsistent with confirmation identity, sends identity validation failure signal, AVR32
Chip, when identifying that identity is consistent with confirmation identity, sends identity validation pass signal.
More specifically, in described client's Verification System based on image recognition: display device is additionally operable to
Show the text prompt information corresponding with identity validation failure signal or identity validation pass signal.
Accompanying drawing explanation
Below with reference to accompanying drawing, embodiment of the present invention are described, wherein:
Fig. 1 is the knot according to the client's Verification System based on image recognition shown in embodiment of the present invention
Structure block diagram.
Reference: 1 image capture device；2 image recognition apparatus；3 main control devices
Detailed description of the invention
Embodiment party to client's Verification System based on image recognition of the present invention below with reference to accompanying drawings
Case is described in detail.
Traditional face recognition technology is mainly based upon the recognition of face of visible images, and this is also people
The recognition method being familiar with, the development history of existing more than 30 year.But this mode has and is difficult to overcome
Defect, especially when ambient lighting changes, recognition effect can drastically decline, it is impossible to meets reality
The needs of system.The scheme solving lighting issues has 3-D view recognition of face, and thermal imaging face is known
Not.But both technology are the most immature, and recognition effect is unsatisfactory.
The a solution developed rapidly is multiple light courcess face based on active near-infrared image
Identification technology.He can overcome the impact that light changes, and has been achieved for the recognition performance of brilliance,
Integral systematicness in terms of precision, stability and speed can exceed that 3-D view recognition of face.This skill
Art quickly grew at nearly 1 year, made face recognition technology gradually move towards practical.
Recognition of face payment is an important applied field of recognition of face.Under tradition charging mode,
Customer need provides bank card to cashier, and bank card is swiped the card behaviour by cashier according to toll amount
Making, and inputted corresponding bank card password by client, payment system is entered in can entering clients account
Capable operation of remitting money, client also needs to sign to confirm subsequently.Whole process needs lasting manual operation
Seamless cooperation with client cashier, it is evident that operate the most loaded down with trivial details, and take considerable time.
And recognition of face pay can Direct Recognition client's facial characteristics, confirm customer identification, it is not necessary to client
Operation, i.e. can complete the process of remitting money.
But, it is directly to remit money owing to recognition of face pays, therefore, its security performance is particularly closed
Key.Existing recognition of face pays and determines that face feature is i.e. remitted money after meeting, and does not accounts for
The situation of face feature identification mistake, meanwhile, the client of queuing checkout is a lot, it is easy to non-checkout
Client carries out facial characteristics identification, and, the existing recognition of face structural redundancy of mechanism own is high,
Also it is improved precision and optimizes the space of function.
In order to overcome above-mentioned deficiency, the present invention has built a kind of client of based on image recognition certification system
System, it is possible to solve above-mentioned technical problem, is replacing loaded down with trivial details manual operation, is improving shopper checkout efficiency
While, improve payments mechanism on the whole, it is ensured that the account number safety of each client.
Fig. 1 is the knot according to the client's Verification System based on image recognition shown in embodiment of the present invention
Structure block diagram, described system includes image capture device, image recognition apparatus and main control device, image
Collecting device is for gathering the image of client, and image recognition apparatus is connected with image capture device, is used for
From the image of client, detect face's subimage and carry out customer identification certification based on face's subimage,
Main control device is connected with image recognition apparatus, for determining corresponding based on customer identification authentication result
Pay strategy.
Then, continue the concrete structure of client's Verification System based on image recognition of the present invention is carried out
Further instruction.
Described system includes: two-way speaker, is connected with AVR32 chip, for playing and body
Part confirm failure signal or voice prompted file corresponding to identity validation pass signal；Supercilium segmentation sets
Standby, it is connected with static storage device and Gaussian smoothing filter equipment respectively to receive preset reference supercilium wheel
Image wide and to be identified, is partitioned into supercilium subgraph based on preset reference supercilium profile in image to be identified
Picture.
Described system includes: mouth splitting equipment, respectively with static storage device and Gaussian smoothing filter
Equipment connects to receive preset reference mouth profile and image to be identified, based on preset reference mouth profile
Mouth subimage it is partitioned in image to be identified；Eye splitting equipment, respectively with static storage device
Connect to receive preset reference eye profile and image to be identified with Gaussian smoothing filter equipment, based in advance
If benchmark eye profile is partitioned into eye subimage in image to be identified.
Described system includes: feature analysis equipment, respectively with supercilium splitting equipment, mouth splitting equipment
Connect with eye splitting equipment, determine supercilium thickness and supercilium flexibility based on supercilium subimage, based on
Mouth subimage determines mouth thickness, determines eye opening degree based on eye subimage.
Described system includes: characteristic matching equipment, respectively with static storage device, feature analysis equipment
Connecting with the face recognition data storehouse of far-end, face recognition data storehouse has prestored everyone face
The supercilium thickness of portion's image, supercilium flexibility, mouth thickness and eye opening degree, set feature analysis
Everyone face-image that the supercilium thickness of standby output prestores with face recognition data storehouse
Supercilium thickness carries out mating to obtain supercilium thickness matching degree, and the supercilium exported by feature analysis equipment is curved
The supercilium flexibility of everyone face-image that curvature and face recognition data storehouse prestore is entered
Row coupling is to obtain supercilium flexibility matching degree, and the mouth thickness exported by feature analysis equipment is with facial
The mouth thickness of everyone face-image that identification database prestores carries out mating to obtain
Mouth thickness matching degree, the eye opening degree exported by feature analysis equipment is pre-with face recognition data storehouse
The eye opening degree of everyone face-image first stored carries out mating to obtain eye opening degree
Matching degree, based on supercilium thickness matching degree, preset supercilium thickness weighted value, supercilium flexibility matching degree,
Default supercilium flexibility weighted value, mouth thickness matching degree, default mouth thickness weighted value, eye are opened
Right matching degree and default eye opening degree weighted value determine image to be identified and face recognition data storehouse
The matching degree of everyone face-image prestored, by face-image the highest for matching degree
Corresponding piece identity is as identifying identity output.
Described system includes: high-definition image collecting device, is arranged on above POS, for queuing
The crowd of checkout carries out shooting to obtain high definition crowd's image；Face detection equipment, deposits with static state respectively
Storage equipment and high-definition image collecting device connect, and are used for receiving high definition crowd's image and preset reference face
Profile, matches multiple face subimage based on preset reference face contour in high definition crowd's image；
Display device, is connected to receive with face detection equipment and shows multiple face subimage, display device
Also with touch screen, from multiple face subimages, select target face with input based on cashier
Image；Geometrical correction device, is connected with face detection equipment and display device respectively, receives target face
Portion's subimage also carries out geometric correction process to obtain geometric correction image to target face subimage.
Described system includes: image orbiting facility, is connected with geometrical correction device to receive geometric correction
Image, carries out image rotation and processes to obtain rotation image geometric correction image；Image translation device,
It is connected with image orbiting facility and rotates image to receive, carry out image translation process to obtain to rotating image
Obtain displacement images；Image segmentation apparatus, is connected with static storage device and image translation device respectively,
For receiving pre-set image block size and displacement images, displacement images is split and presets figure to obtain
Segmentation image as block size.
Described system includes: histogram equalization equipment, is connected with image segmentation apparatus, is used for receiving
Segmentation image also carries out histogram equalization process to segmentation image, straight to obtain the gray scale of segmentation image
Fang Tu；Gaussian smoothing filter equipment, is connected with histogram equalization equipment, is used for receiving intensity histogram
Figure also carries out Gaussian smoothing filter process to grey level histogram, to obtain image to be identified；Static storage
Equipment, is used for having prestored preset reference face contour and pre-set image block size, is additionally operable in advance
Store default supercilium thickness weighted value, preset supercilium flexibility weighted value, default mouth thickness weight
Value, preset eye opening degree weighted value, preset reference supercilium profile, preset reference mouth profile and pre-
If benchmark eye profile.
Described system includes: Wireless Telecom Equipment, for setting up characteristic matching by wireless communication link
Connection between the face recognition data storehouse of equipment and far-end, is additionally operable to be set up by wireless communication link
Connection between fingerprint matching device and the fingerprint database of far-end；Fingerprint receives equipment, is used for gathering
The finger print information of client.
Described system includes: fingerprint matching device, receives equipment and the fingerprint number of far-end with fingerprint respectively
Connecting according to storehouse, fingerprint database has prestored everyone fingerprint characteristic, receives based on fingerprint and sets
The finger print information of standby output finds the fingerprint characteristic of coupling at fingerprint database, and by special for the fingerprint of coupling
Levy the piece identity of correspondence as confirming identity output.
Described system includes: AVR32 chip, respectively with display device, characteristic matching equipment, fingerprint
Reception equipment and fingerprint matching device connect, when receiving identification identity at characteristic matching equipment,
Client to current checkout starts fingerprint reception equipment and fingerprint matching device and confirms identity to receive, when
The client of front checkout is in the crowd of queuing checkout, when identifying that identity is consistent with confirmation identity,
To identify that the value data that identity and cashier are inputted by the touch screen of display device is sent collectively to
The payment devices of far-end is to complete to pay, simultaneously to the E-mail address transmission confirmation postal identifying that identity is corresponding
Part, confirms that mail includes customer payment video.
Wherein, preset reference face contour is that benchmark face-image is carried out the figure that contours extract obtains
Shape, pre-set image block size elects 45 pixel × 50 pixels as；High-definition image collecting device is additionally operable to record
Customer payment video processed.
Alternatively, in the system: preset reference supercilium profile is for take turns benchmark supercilium image
The figure that exterior feature extracts and obtains；Preset reference mouth profile is for carry out contours extract to benchmark mouth image
And the figure obtained；Preset reference eye profile obtains for benchmark eyes image is carried out contours extract
Figure；When AVR32 chip does not receives identification identity after the first Preset Time, send identity true
Recognize failure signal, when AVR32 chip does not receives confirmation identity after the second Preset Time, send body
Part confirms failure signal, and AVR32 chip, when identifying that identity is inconsistent with confirmation identity, sends body
Part confirms failure signal, and AVR32 chip, when identifying that identity is consistent with confirmation identity, sends identity
Confirm pass signal；And display device is additionally operable to display and identity validation failure signal or identity validation
The text prompt information that pass signal is corresponding.
It addition, the advantage of recognition of face is its naturality and the feature do not discovered by tested individuality.
So-called naturality, refers to that the same mankind of this recognition method (even other biological) carry out individual identification
Time the biological characteristic that utilized identical.Such as recognition of face, the mankind are also to compare face district by observation
Dividing and confirm identity, the identification additionally with naturality also has iris identification speech recognition, the bodily form
Identify, and fingerprint recognition, iris identification etc. the most do not have naturality, because the mankind or other lifes
Thing is not by this type of biological characteristic difference individuality.
Not detectable feature is the most critically important for a kind of recognition methods, and this can make this recognition methods not make
People dislikes, and because is not easy to arouse people's attention and be not easy to be spoofed.Recognition of face has this
The feature of aspect, his fully utilized visible ray obtains human face image information, and be different from fingerprint recognition or
Person's iris identification, needs to utilize electronic pressure transmitter to gather fingerprint, or utilizes infrared collection rainbow
Film image, these special acquisition modes are easy to be therefore easily perceived by humans, thus more likely by impersonation.
Recognition of face is mainly used in identification.Owing to video monitoring is the most quickly popularized, numerous regards
Frequency monitoring application is in the urgent need to the quick identification skill under a kind of remote, non-mated condition of user
Art, in the hope of quickly confirming personnel identity at a distance, it is achieved intelligent early-warning.Face recognition technology is beyond doubt
Optimal selection, use fast face detection technique can from monitor video image real-time searching people
Face, and carry out real-time comparison with face database, thus realize quick identification.
Use client's Verification System based on image recognition of the present invention, for prior art client face
The technical problem that payments mechanism accuracy is the highest, is facial recognition by increasing secondary identities authentication mechanism
To identity further confirmed, for cashier provide near multiple custom image for cashier select
Select out the client of current checkout, meanwhile, also face feature identification pattern is carried out performance improvement.
Although it is understood that the present invention discloses as above with preferred embodiment, but above-mentioned enforcement
Example is not limited to the present invention.For any those of ordinary skill in the art, without departing from
Under technical solution of the present invention ambit, all may utilize the technology contents of the disclosure above to the technology of the present invention
Scheme makes many possible variations and modification, or is revised as the Equivalent embodiments of equivalent variations.Therefore,
Every content without departing from technical solution of the present invention, the technical spirit of the foundation present invention is to above example
Any simple modification, equivalent variations and the modification done, all still falls within technical solution of the present invention protection
In the range of.
Claims (7)
1. client's Verification System based on image recognition, described system include image capture device,
Image recognition apparatus and main control device, image capture device is for gathering the image of client, image recognition
Equipment is connected with image capture device, for detect from the image of client face's subimage and based on
Face's subimage carries out customer identification certification, and main control device is connected with image recognition apparatus, for based on
Customer identification authentication result determines and pays strategy accordingly.
2. client's Verification System based on image recognition as claimed in claim 1, it is characterised in that
Described system includes:
Two-way speaker, is connected with AVR32 chip, for play with identity validation failure signal or
The voice prompted file that identity validation pass signal is corresponding；
Supercilium splitting equipment, is connected with static storage device and Gaussian smoothing filter equipment to receive respectively
Preset reference supercilium profile and image to be identified, based on preset reference supercilium profile in image to be identified
It is partitioned into supercilium subimage；
Mouth splitting equipment, is connected with static storage device and Gaussian smoothing filter equipment to receive respectively
Preset reference mouth profile and image to be identified, based on preset reference mouth profile in image to be identified
It is partitioned into mouth subimage；
Eye splitting equipment, is connected with static storage device and Gaussian smoothing filter equipment to receive respectively
Preset reference eye profile and image to be identified, based on preset reference eye profile in image to be identified
It is partitioned into eye subimage；
Feature analysis equipment, respectively with supercilium splitting equipment, mouth splitting equipment and eye splitting equipment
Connect, determine supercilium thickness and supercilium flexibility based on supercilium subimage, determine based on mouth subimage
Mouth thickness, determines eye opening degree based on eye subimage；
Characteristic matching equipment, the face with static storage device, feature analysis equipment and far-end is known respectively
Other data base connects, and the supercilium that face recognition data storehouse has prestored everyone face-image is thick
Degree, supercilium flexibility, mouth thickness and eye opening degree, the supercilium exported by feature analysis equipment is thick
The supercilium thickness of everyone face-image that degree and face recognition data storehouse prestore is carried out
It is equipped with acquisition supercilium thickness matching degree, the supercilium flexibility exported by feature analysis equipment and facial recognition
The supercilium flexibility of everyone face-image that data base prestores carries out mating to obtain eyebrow
Portion's flexibility matching degree, the mouth thickness exported by feature analysis equipment is with face recognition data storehouse in advance
The mouth thickness of everyone face-image of storage carries out mating to obtain mouth thickness matching degree,
Each that the eye opening degree that feature analysis equipment exports is prestored with face recognition data storehouse
The eye opening degree of the face-image of people carries out mating to obtain eye opening degree matching degree, based on supercilium
Thickness matching degree, default supercilium thickness weighted value, supercilium flexibility matching degree, default supercilium flexibility
Weighted value, mouth thickness matching degree, preset mouth thickness weighted value, eye opening degree matching degree and pre-
If eye opening degree weighted value determines that image to be identified is each with what face recognition data storehouse prestored
The matching degree of the face-image of individual, by piece identity corresponding for face-image the highest for matching degree
As identifying identity output；
High-definition image collecting device, is arranged on above POS, for carrying out the crowd of queuing checkout
Shooting is to obtain high definition crowd's image；
Face detection equipment, is connected with static storage device and high-definition image collecting device respectively, is used for
Receive high definition crowd's image and preset reference face contour, based on preset reference face contour high definition people
Group's image matches multiple face subimage；
Display device, is connected to receive with face detection equipment and shows multiple face subimage, display
Equipment also with touch screen, selects target face with input based on cashier from multiple face subimages
Portion's subimage；
Geometrical correction device, is connected with face detection equipment and display device respectively, receives target face
Subimage also carries out geometric correction process to obtain geometric correction image to target face subimage；
Image orbiting facility, is connected to receive geometric correction image, to geometry school with geometrical correction device
Positive image carries out image rotation and processes to obtain rotation image；
Image translation device, is connected with image orbiting facility and rotates image to receive, and enters rotating image
Row image translation processes to obtain displacement images；
Image segmentation apparatus, is connected with static storage device and image translation device respectively, is used for receiving
Pre-set image block size and displacement images, split displacement images to obtain pre-set image block size
Segmentation image；
Histogram equalization equipment, is connected with image segmentation apparatus, is used for receiving segmentation image and to dividing
Cut image and carry out histogram equalization process, to obtain the grey level histogram of segmentation image；
Gaussian smoothing filter equipment, is connected with histogram equalization equipment, is used for receiving grey level histogram
And grey level histogram is carried out Gaussian smoothing filter process, to obtain image to be identified；
Static storage device, is used for having prestored preset reference face contour and pre-set image block is big
Little, it is additionally operable to have prestored default supercilium thickness weighted value, presets supercilium flexibility weighted value, pre-
If mouth thickness weighted value, default eye opening degree weighted value, preset reference supercilium profile, default base
Quasi-mouth profile and preset reference eye profile；
Wireless Telecom Equipment, for setting up the face of characteristic matching equipment and far-end by wireless communication link
Connection between portion's identification database, be additionally operable to by wireless communication link set up fingerprint matching device with
Connection between the fingerprint database of far-end；
Fingerprint receives equipment, for gathering the finger print information of client；
Fingerprint matching device, the fingerprint database receiving equipment and far-end respectively with fingerprint is connected, fingerprint
Data base has prestored everyone fingerprint characteristic, receives the fingerprint letter of equipment output based on fingerprint
Cease the fingerprint characteristic finding coupling at fingerprint database, and by personage's body corresponding for the fingerprint characteristic of coupling
Part is as confirming identity output；
AVR32 chip, receives equipment and fingerprint with display device, characteristic matching equipment, fingerprint respectively
Matching unit connects, when receiving identification identity at characteristic matching equipment, to the Gu of current checkout
Visitor starts fingerprint and receives equipment and fingerprint matching device to receive confirmation identity, at the client of current checkout
In the crowd of queuing checkout, when identifying identity with when confirming that identity is consistent, identity and receipts will be identified
Silver member is sent collectively to the payment devices of far-end by the value data that the touch screen of display device inputs
To complete to pay, simultaneously to the E-mail address transmission confirmation mail identifying that identity is corresponding, confirm in mail
Including customer payment video；
Wherein, preset reference face contour is that benchmark face-image is carried out the figure that contours extract obtains
Shape, pre-set image block size elects 45 pixel × 50 pixels as；
Wherein, high-definition image collecting device is additionally operable to record customer payment video.
3. client's Verification System based on image recognition as claimed in claim 2, it is characterised in that:
Preset reference supercilium profile is that benchmark supercilium image is carried out the figure that contours extract obtains.
4. client's Verification System based on image recognition as claimed in claim 2, it is characterised in that:
Preset reference mouth profile is that benchmark mouth image is carried out the figure that contours extract obtains.
5. client's Verification System based on image recognition as claimed in claim 2, it is characterised in that:
Preset reference eye profile is that benchmark eyes image is carried out the figure that contours extract obtains.
6. client's Verification System based on image recognition as claimed in claim 2, it is characterised in that:
Wherein, when AVR32 chip does not receives identification identity after the first Preset Time, identity is sent
Confirm failure signal, when AVR32 chip does not receives confirmation identity after the second Preset Time, send
Identity validation failure signal, AVR32 chip, when identifying that identity is inconsistent with confirmation identity, sends
Identity validation failure signal, AVR32 chip, when identifying that identity is consistent with confirmation identity, sends body
Part confirms pass signal.
7. client's Verification System based on image recognition as claimed in claim 2, it is characterised in that:
It is corresponding with identity validation failure signal or identity validation pass signal that display device is additionally operable to display
Text prompt information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610204738.9A CN105894262A (en) | 2016-04-01 | 2016-04-01 | Customer authentication system based on image identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610204738.9A CN105894262A (en) | 2016-04-01 | 2016-04-01 | Customer authentication system based on image identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105894262A true CN105894262A (en) | 2016-08-24 |
Family
ID=57012384
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610204738.9A Withdrawn CN105894262A (en) | 2016-04-01 | 2016-04-01 | Customer authentication system based on image identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105894262A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106372606A (en) * | 2016-08-31 | 2017-02-01 | 北京旷视科技有限公司 | Target object information generation method and unit identification method and unit and system |
| CN108171496A (en) * | 2017-12-01 | 2018-06-15 | 运鼎科技（北京）有限公司 | A kind of shopping correlating method, server and system |
| CN108846664A (en) * | 2018-06-26 | 2018-11-20 | 刘晓英 | Mobile phone scan code system based on image procossing |
| CN111445701A (en) * | 2019-11-02 | 2020-07-24 | 泰州市海陵区一马商务信息咨询有限公司 | Taxi violation information big data registration system and method |
| CN112862638A (en) * | 2019-11-26 | 2021-05-28 | 上海径舟教育科技有限公司 | Remote education system and method based on face recognition |
| CN113128334A (en) * | 2021-03-03 | 2021-07-16 | 广州朗国电子科技有限公司 | AI intelligent image recognition method, electronic equipment and storage medium |
-
2016
- 2016-04-01 CN CN201610204738.9A patent/CN105894262A/en not_active Withdrawn
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106372606A (en) * | 2016-08-31 | 2017-02-01 | 北京旷视科技有限公司 | Target object information generation method and unit identification method and unit and system |
| CN108171496A (en) * | 2017-12-01 | 2018-06-15 | 运鼎科技（北京）有限公司 | A kind of shopping correlating method, server and system |
| CN108846664A (en) * | 2018-06-26 | 2018-11-20 | 刘晓英 | Mobile phone scan code system based on image procossing |
| CN111445701A (en) * | 2019-11-02 | 2020-07-24 | 泰州市海陵区一马商务信息咨询有限公司 | Taxi violation information big data registration system and method |
| CN112862638A (en) * | 2019-11-26 | 2021-05-28 | 上海径舟教育科技有限公司 | Remote education system and method based on face recognition |
| CN113128334A (en) * | 2021-03-03 | 2021-07-16 | 广州朗国电子科技有限公司 | AI intelligent image recognition method, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105894262A (en) | Customer authentication system based on image identification | |
| CN105913241A (en) | Application method of customer authentication system based on image identification | |
| CN101558431B (en) | Face authentication device | |
| CN110516576B (en) | Near-infrared living body face recognition method based on deep neural network | |
| CN106845414B (en) | Method and system for quality metrics for biometric verification | |
| Bolle et al. | Guide to biometrics | |
| EP1335329B1 (en) | Personal authentication method, personal authentication apparatus and image capturing device | |
| CN102629319B (en) | Robust iris region segmentation method based on specific boundary detectors | |
| CN110008813B (en) | Face recognition method and system based on living body detection technology | |
| US9064145B2 (en) | Identity recognition based on multiple feature fusion for an eye image | |
| CN105787454B (en) | Method and system for biometric verification | |
| CN105518709A (en) | Method, system and computer program product for identifying human face | |
| CN107438854A (en) | The system and method that the image captured using mobile device performs the user authentication based on fingerprint | |
| US20170262472A1 (en) | Systems and methods for recognition of faces e.g. from mobile-device-generated images of faces | |
| US20060222212A1 (en) | One-dimensional iris signature generation system and method | |
| CN106529414A (en) | Method for realizing result authentication through image comparison | |
| CN101142584A (en) | Method for facial features detection | |
| CN108922074A (en) | Without card withdrawal method, apparatus, computer equipment and storage medium | |
| CN105678290A (en) | Face payment platform based on image detection | |
| US10922399B2 (en) | Authentication verification using soft biometric traits | |
| CN103383723A (en) | Method and system for spoof detection for biometric authentication | |
| CN108446687A (en) | A kind of adaptive face vision authentication method based on mobile terminal and backstage interconnection | |
| CN105913263A (en) | Application method of face payment platform based on image detection | |
| CN109255618A (en) | For the recognition of face payment information method for anti-counterfeit of dynamic video | |
| US20220277311A1 (en) | A transaction processing system and a transaction method based on facial recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20160824 |
|
| WW01 | Invention patent application withdrawn after publication |