CN106529414A - Method for realizing result authentication through image comparison - Google Patents
Method for realizing result authentication through image comparison Download PDFInfo
- Publication number
- CN106529414A CN106529414A CN201610898697.8A CN201610898697A CN106529414A CN 106529414 A CN106529414 A CN 106529414A CN 201610898697 A CN201610898697 A CN 201610898697A CN 106529414 A CN106529414 A CN 106529414A
- Authority
- CN
- China
- Prior art keywords
- identity card
- card
- user
- photo
- 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.)
- Pending
Links
Classifications
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/142—Image acquisition using hand-held instruments; Constructional details of the instruments
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
Abstract
The invention discloses a method for realizing result authentication through image comparison. The method includes the following steps that: S1, the photo of the identity card of a user is acquired; S2, living body identification is performed on the user; and S3, the archived and registered photo of the identity card is compared with each photo extracted from the living body identification, whether a person indicated by the identity card and the living user are the same person is judged, if one photo of the living user, which indicates a person different from the person indicated by the identity card, exists, the method returns to the S1, if persons indicated by all the photos are the same person as the person indicated by the identity card, it is confirmed that the person indicated by the identity card and the living user are the same person. According to the method, the accurate input of the identity information of the user can be realized through specific algorithms of optical recognition technologies; existing identity authentication business is used in combination, so that the content of identity authentication can be enriched and improved; the verification of a real person is realized through a living body detection system; automatic comparison of the person and the preserved photos of a population library is realized through person image comparison technologies, and therefore, identification efficiency is improved; and diverse terminals such as operating hall system terminal systems, PC sides, mobile phone clients, are adopted.
Description
Technical field
The present invention relates to image processing field, in particular it relates to the method for realizing result certification is compared by image.
Background technology
Bank account real name is the important financial foundation system of China, more than bank by the way of artificial comparison,
It is as manpower comparing is inconsistent to standard, wide tight condition is personalized, it is difficult to which the authenticity of client identity is effectively verified, illegal
Molecule is opened the phenomenons such as false account family, false loan using false (or defrauding of) resident identification card and is also happened occasionally.A traditional person
Part checking means such as mode such as password, certificate, IC-card, due to the separability with identity people, causes forgery, usurps, decodes
Phenomenon happens occasionally, and can not meet the needs that modern social economy activity and social safety are taken precautions against.Living things feature recognition bag
Include fingerprint, palmmprint, voice, face, iris, gait, vena metacarpea etc..Biometrics identification technology first puts into wide variety of referring to
Line, palmmprint scanning recognition technology, but but usually draw because being subject to the conditionality such as dermatoglyph and degree of drying erroneous judgement occur
The trouble for sending out unnecessary, far can not meet the demand of people.
With commonly using for wireless Internet and smart mobile phone, recognition of face is also faced with increasing challenge.It is difficult
Degree is embodied in:1st, image resolution ratio is low.Under high-definition picture, facial image can transmit more abundant user profile.
And in these the Internet, applications, the quality of camera is very different, this brings difficulty to effectively extraction facial characteristics.2nd, mould
State diversity.Due to the difference of difference and the environment light source of collecting device, facial image with the presence of various different modes,
Under different illumination wave bands, face reflectivity also result in the loss of image information.
The content of the invention
The present invention is to provide one kind by image comparison realization knot by solving the technical scheme that above-mentioned technical problem is adopted
The method of fruit certification, including:
S1, collection user identity license piece, including:
S1-1, by terminating layer login service layer service interface, service layer pushes chained address, and terminating layer downloads APP,
S1-2, the double-edged photo of identity card is shot in terminating layer, sent with cell-phone number, IMEI, positional information in the lump
To inspected layer,
S1-3, inspected layer parse ID card information using OCR, extract and return name, ID card No., live to terminating layer
Location, sex, Folk Information, if having Liang Chu and two to sentence with real name, ID card No., address, sex, Folk Information
It is upper wrong, then return S1-2, if wrong at only one, correct, and to inspected layer feed back correct name, ID card No.,
Address, sex, Folk Information,
S1-4, by the correct name for being returned, ID card No., address, sex, Folk Information via inspected layer send
To authentication layers, authentication layers are authenticated from public security database gathering information, checking name, ID card No., address, sex,
Whether Folk Information is corresponding and true, if certification does not pass through, returns failure information to terminating layer, if passing through, extracts and uses
Family achieves to register in the identity card of public security database to be shone, and contrasts and judge that the identity card achieves registration and uploads according to shooting with terminating layer
Identity card picture in face whether be same people, if not, return S1, if it is, carrying out S2;
S2, vivo identification is carried out to user, including:
S2-1,3 composition set sequences of the random extraction from default behavior aggregate,
S2-2, the continuous action specified is completed before camera lens by voice message instruction user,
S2-3, computer disposal user action video, judge whether to complete required movement, if it is, right from video flowing
1 photo is extracted in each action, if not, return performing S2-1;
S3, contrast the identity card achieve registration takes and vivo identification in per an extraction picture, judge identity card and work
Whether body user is same people, if one is no, returns S1, if being all, confirms that identity card and live body user are same
One people.
Preferably, in step S1-4 contrast and judge the identity card achieve registration according to and the identity card that uploads of terminating layer shine
In piece, whether face is that same people includes:1) characteristic point of face face is extracted using active shape model algorithm:Canthus, eye
Eyeball center, the corners of the mouth, nose, chin profile;2) position of face organ is determined using the characteristic point extracted, it is big according to each organ
The little sample window for selecting appropriate pixels size, extracts the sampling block of all subregion;3) identity card is achieved registration to shine and whole
The identity card picture that end layer is uploaded is divided into sub-block according to sampling block, obtains sampling block matching degree using the method for Block- matching, will adopt
Photo whole matching degree is tried to achieve after sample Block- matching degree weighted average, is shone and terminating layer upload for differentiating that identity card achieves registration
Whether identity card picture is same person.
Preferably, the identity card is contrasted in step S3 and achieves to register take and per an extraction picture, sentence in vivo identification
Whether disconnected identity card and live body user are that same people includes:Using local feature reflection method, i.e., identity card is achieved registration according to and
The pretreatment for carrying out human face photo per an extraction photo in vivo identification, the face edge by face from photo is plucked out, fixed
The distance between position pupil EA, pupil and face's lateral edges for being closed on apart from EB, the distance between nose and two pupils
NL、NR, the distance between nostril center NB, the distance between lip upper limb and nose LC, between lip lower edge and chin bottom
Apart from LD, identity card is achieved registers according to corresponding E in the identity card picture uploaded with terminating layerA/EA’、EB/EB’、NL/NL’、
NR/NR’、NB/NB' Difference and two-by-two among big number ratio all within 2%, then calculate LC/LC’、LD/LD', its result
Ratio with the two among big number ratio within 5%, then it is assumed that identity card achieve registration according to and vivo identification in per
It is same people to extract photo.
Preferably, the terminating layer in step S1-2 is taken pictures including with PC or high photographing instrument or mobile phone photograph.
The present invention is logged in using face authentication, can remove loaded down with trivial details Password Input from, and interaction is more efficient and convenient.Except this it
Outward, the identification custom of the mankind, the low cost of image capture device as a typical biometrics identification technology, are more met.
The terminal camera that face authentication needs nearly all is the peripheral hardware of standard on PC and mobile phone, avoiding problems fingerprint, iris etc.
Technology of identification needs the trouble of special collecting device.The image realized by the algorithm of the present invention is compared, and overcomes image resolution
Rate is low, the multifarious technical problem of mode, and image comparison speed is fast, the degree of accuracy is high.Additionally, the present invention is compared by image realizing
The method of as a result certification has further the advantage that:
1. the accurate typing of subscriber identity information, by the special algorithm of optical recognition, is realized;
2. combined with existing authentication business, enrich and improve authentication content;
3., by In vivo detection system, realize verifying " true man ";
4., by portrait comparison technology, realize that the automatic comparison of photo is retained in person to person's mouth storehouse, improve recognition efficiency;
5. plurality of application scenes is suitable for, and both convenient reliable service can be provided for clients such as operators;Can also be for most
Whole user brings and whenever and wherever possible, flexibly, quickly experiences.Using terminal is various, including:Business hall terminal system, PC ends, mobile phone
Client.
Description of the drawings
Fig. 1 is to compare mode business schematic flow sheet using the cloud of mobile phone and OCR identification modules;
Fig. 2 is the local comparison mode business schematic flow sheet using PC or high photographing instrument;
Fig. 3 is the local comparison mode business schematic flow sheet using mobile phone;
Reference:N- names;ID- ID card No.;R1- identity authentication results;R2- photo comparison's results;P1- identity
Storehouse watermark is shone;P2- scenes living photo;P3- identity card pictures;Characteristic value is shone in P1 '-watermark;P2 '-living photo characteristic value;S- photos
(characteristic value) compares score.
Specific embodiment
As shown in figure 1, a kind of compare the method for realizing result certification by image, pattern is compared including using cloud, at this
Under pattern, the method combined with OCR identifications using mobile phone, specifically flat in local disposition human face photo characteristics extraction
Platform, extracts identification feature to the photo of photo in consolidated storage and user's shooting, uploads feature to be compared by cloud mode
Data give people to compare as comparing service provider, are then back to comparison result.
Concrete grammar includes:
From the authentication module of service handling terminal to business operation system initiate collection user identity license piece and
Call OCR module to ask, identity card picture is shot by mobile phone;Including shooting the double-edged photo of identity card, with cell-phone number,
IMEI, positional information send together to portrait the control module for comparing platform;
The identity card picture of shooting is sent to mobile phone the control module that portrait compares platform, and mobile phone is clapped by the control module
The identity card picture taken the photograph is sent to the OCR module that portrait compares platform, by name and ID card No. after OCR module identification
Return to the control module, the name recognized through OCR module and ID card No. are returned to mobile phone by the control module, by with
Family is corrected, if having more than at Liang Chu and two with real name, ID card No., address, sex, Folk Information had
By mistake, then identity card picture is re-shoot, if wrong at only one, is corrected, and fed back correct name, ID card No., live
Location, sex, Folk Information;
After user carries out information correction, the name after correction and identification card number result are compared into the control of platform by portrait
Module is sent to business operation system via GBOSS modules, and then is sent to service handling terminal;
The name for receiving and identification card number are sent to by service handling terminal via business operation system, GBOSS modules
Identity authentication result corresponding with the name and identification card number that receive, the watermark of identity storehouse are shone Jing by central platform, central platform
The control module that business operation system and portrait compare platform is simultaneously sent to by GOBSS modules, business operation system is by identity
Authentication result, the watermark of identity storehouse are shone and return to service handling terminal;The authentication is carried out from public security database gathering information
Whether certification, checking name, ID card No., address, sex, Folk Information are corresponding and true.
The portrait comparing module of service handling terminal via business operation system to mobile phone promoter as obtain, mobile phone to
Family carries out vivo identification, shoots living photo;
Living photo is sent to mobile phone the control module that portrait compares platform, and the watermark of identity storehouse is shone, showed by the control module
Field living photo is sent to the portrait characteristic extracting module that portrait compares platform, and the portrait characteristic extracting module will be through feature extraction
Watermark afterwards is sent to the control module that portrait compares platform according to characteristic value and living photo characteristic value, and watermark is shone by the control module
Characteristic value and living photo characteristic value are sent to cloud platform and carry out portrait contrast, and photo eigen value comparison score is sent to by cloud platform
Portrait compare platform control module, the control module according to compare score draw photo comparison result, by the result via
GBOSS modules, business operation system are sent to service handling terminal, complete portrait contrast, realize result certification.
As shown in Fig. 2 a kind of compare the method for realizing result certification by image, including using local comparison pattern,
Under the pattern, the method collection photo of PC or high photographing instrument, photograph taking mode and user's correcting mode and aforementioned reality can be adopted
Apply example identical, specifically, platform is compared in user side local disposition portrait.
Concrete grammar includes:
Name and ID card No. are sent to into production platform by service handling terminal, production platform is by which via comprehensive industry
Business platform is sent to data source, and data source is by identity authentication result corresponding with the name and identification card number that receive, identity storehouse
Watermark is sent to the control module that portrait compares platform according to via synthesis business platform, while identity authentication result, identity storehouse water
Print is sent to service handling terminal according to via synthesis business platform, production platform, and service handling terminal passes through PC or high photographing instrument is clapped
Living photo is taken the photograph, captured living photo is sent by PC or high photographing instrument via service handling terminal, production platform, synthesis business platform
The control module of platform is compared to portrait, it is flat that the watermark of identity storehouse is shone by the control module, live living photo is sent to portrait comparison
The comparing module of platform is compared, and photo eigen value is compared score and is sent to the control mould that portrait compares platform by the comparing module
Block, the control module draw photo comparison result according to score is compared, and the result is sent out via synthesis business platform, production platform
Service handling terminal is sent to, portrait contrast is completed, is realized result certification.
As shown in figure 3, a kind of compare the method for realizing result certification by image, including using local comparison pattern,
Under the pattern, photo can be gathered using mobile phone, photograph taking mode and user's correcting mode are identical with previous embodiment, specifically
Ground, compares platform in user side local disposition portrait.
Concrete grammar includes:
Name and ID card No. are sent to into production platform by service handling window, production platform is by which via comprehensive industry
Business platform is sent to data source, and data source is by identity authentication result corresponding with the name and identification card number that receive, identity storehouse
Watermark is sent to the control module that portrait compares platform according to via synthesis business platform, while identity authentication result, identity storehouse water
Print is sent to service handling window according to via synthesis business platform, production platform, and service handling window causes mobile phone by under APP
Set address downloads APP, and mobile phone carries out vivo identification to user, shoots living photo, and captured living photo is uploaded to people by mobile phone
As comparing the control module of platform, the watermark of identity storehouse is shone by the control module, live living photo is sent to portrait and compares platform
Comparing module is compared, and photo eigen value is compared score and is sent to the control module that portrait compares platform by the comparing module,
The control module draws photo comparison result according to score is compared, and the result is sent via synthesis business platform, production platform
To service handling window, portrait contrast is completed, result certification is realized.
Wherein, synthesis business platform be used for receive commerce system service order and perform (for example identity information veritify, it is right
Living photo/In vivo detection/identity card picture OCR identification, portrait compare), output identity card watermark according to control platform, output
Instruction gives people as control module, receives identity card OCR identification final results and export to trade company, recipient as comparing control module
The portrait comparison result of return is simultaneously exported to trade company.GBOSS modules are external system interfaces, and this method calls GBOSS systems to connect
Mouth obtains authentication idses information and identity card head image data.
Preferably, carry out identity card and achieve comparison of the registration according to the identity card picture uploaded with terminating layer, be divided into following three
The individual main stage:
1) extract human face characteristic point.The characteristic point of face face is extracted using active shape model algorithm:Canthus, eyes
Center, the corners of the mouth, nose, chin profile etc..
2) divide face subregion.The position of face organ is determined using the characteristic point extracted, according to each organ size
The sample window of appropriate pixels size is selected, the sampling block of all subregion is extracted.
3) sampling block similarity is asked for according to sampling block, will currently and reference picture is divided into sub-block, using Block- matching
Whether method obtains matching degree, be same person for differentiating that part card achieves registration according to the identity card picture uploaded with terminating layer.
Preferably, in above-mentioned the 3) stage, current and reference picture is divided into into sub-block, is obtained using the method for Block- matching
Matching degree, is using following method:One piece according in of registration is achieved as template using identity card, in the identity that terminating layer is uploaded
Search and the most like match block of this template in license piece, half-tone information contain the maximum amount of image information, line of demarcation information
Reflect in image property, line of demarcation information be difficult by extraneous illumination condition affected and produce it is violent change, therefore
Half-tone information is compared, line of demarcation information is caused its anti-gray scale and geometric distortion ability strong, existed using line of demarcation information architecture template
Relatively reliable stability can be obtained in image mosaic.Therefore searching reference block is characterized with the line of demarcation of image.
Carry out clearly marginal detection first to image, we obtain the boundary line chart of original image using following manner here
E(i,j):
S operators template both vertically and horizontally, i.e., two convolution kernel g1(x, y) and g2(x, y) is as shown in the table,
The former can detect the edge of the horizontal direction in image, and the latter then can be with the edge of vertical direction in detection image.
By two convolution kernel g1(x, y) and g2(x, y) carries out convolution algorithm, s operator mathematic(al) representations to original image f (x, y)
For:
The algorithm for being adopted is first to carry out weighted average, is then differentiated, and can replace single order local derviation with difference,
The computational methods of operator are as follows:
Preferably, each pixel in image carries out convolution algorithm with the two convolution kernels, takes its maximum and makees
For output.Operation result is the image that a width embodies line of demarcation amplitude.
Thereafter, three groups of S (X, Y) values are obtained under different contrasts and respectively under different brightness, takes which average as most
Boundary line chart E (i, j) of whole original image.
Then with each pixel neighborhood of a point periphery amount (NEA) in image defining the line of demarcation information size of this position:
Wherein, E (i, j) is the two-value boundary line chart corresponding to original image, and A is to take 1/2 side of neighborhood in point (i, j) place
It is long.With NEA (i, j) value as foundation, point (i, j) corresponding to maximum is found out as datum mark, then the point centered on this puts
The block that selection is of moderate size can just find reference characteristic block.
Current and reference picture is divided into into sub-block, general sub-block is 8 × 8 pixel sizes, it is then determined that search range,
15 × 15 pixel regions generally centered on sub-block, finally find cost function most in reference picture corresponding searching area
Little sub-block.
Cost function is mean square error (MSE) and mean absolute error (MAD), such as equation (1) and (2).Wherein N is block side
Long pixel count (being generally divided into square for convenience of search block), Cij and Rij is respectively current sub-block and the corresponding picture of reference sub-block
The gray scale of element.
If in image, the mean square deviation of the cost function MSE and MAD of each sub-block is within threshold value, by the sub-block
Typical value is set as 0, if outside threshold value, the typical value of the sub-block is set as 1, by the sub-block number that typical value is 1
Divided by the total sub-block number of image, matching degree is obtained.Matching degree is within the scope of restriction, then it is assumed that the images match, otherwise it is assumed that
Mismatch.
Require to set different restriction scopes according to different comparing, shine and terminating layer upload for identity card achieves registration
Identity card picture for, need precise alignment, be then set to 0%-5%, for business application APP use, then be set to
5%-10%.
Preferably, face characteristic can also be recognized using dynamic position signature analysis.The pre- of human face photo is carried out first
Process, remove the too high noise of photo, the method migration that the human face photo of input is detected with peripheral adjustment is into two-symbol photograph
Piece, then extract face texture, then using Local Features Analysis method processing the edge shadow and skin line of face, from
And recognize face.
Genius loci analysis be a kind of genius loci represent similar to build building blocks statistics technology, based on all of
Face all can be comprehensively formed by the construction unit block for much simplifying.Genius loci is analyzed using 32 to 50 cell block regions to distinguish
An other face, the point most commonly of selection include nose, eyes, face and specific bone curvature difference, such as cheek.These lists
First block is formed using complicated statistical technique, and they represent whole face, across multiple pixels and is represented universal
Facial contours, but to determine that a face depends not only on characteristics unit block, further depend on their geometry (such as shape
Shape and relevant position).In this way, the characteristic of face is corresponded to a kind of numeral expression of complexity by genius loci analysis
Formula, so that carry out recognition of face.
The shape of face information obtained with dynamic edge analysis method can reduce the impact of different light environments, and skin line information is carried
The details of standardized face characteristic is supplied.By two kinds of information combinations of the shape of face and skin line again with Local Features Analysis algorithm comparing
Compared with, statistics face 173 characteristic points, no matter registration and recognize face when be, under which kind of light environment, to obtain very high
Discrimination, can accurately recognize a people in million people.
Preferably, the contrast identity card achieves registration and takes and per an extraction picture, judge identity card in vivo identification
Whether it is that same people includes with live body user:Using local feature reflection method, i.e., registration is achieved to identity card and shone and vivo identification
In carry out the pretreatment of human face photo per extracting photo, the face edge by face from photo is plucked out, positioning pupil it
Between apart from EA, pupil and face's lateral edges for being closed on apart from EB, the distance between nose and two pupils NL、NR, nostril
The distance between center NB, the distance between lip upper limb and nose LC, the distance between lip lower edge and chin bottom LD,
Identity card achieves registration according to corresponding E in the identity card picture uploaded with terminating layerA/EA’、EB/EB’、NL/NL’、NR/NR’、NB/
NB' Difference and two-by-two among big number ratio all within 2%, then calculate LC/LC’、LD/LD', the ratio of its result and two
Among person, the ratio of big number is within 5%, then it is assumed that identity card achieve registration take and vivo identification in per an extraction picture is
Same people.
The above, the only present invention preferably specific embodiment, but protection scope of the present invention is not limited thereto,
Any those familiar with the art the invention discloses technical scope in, the change or replacement that can be readily occurred in,
Should all be included within the scope of the present invention.Therefore, protection scope of the present invention should be described with the protection model of claim
Enclose and be defined.
Claims (6)
1. it is a kind of that the method for realizing result certification is compared by image, including:
S1, collection user identity license piece;
S2, vivo identification is carried out to user;And
S3, contrast the identity card achieve registration takes and vivo identification in per a picture is extracted, judge identity card and live body use
Whether family is same people, if one is no, returns S1, if being all, confirms that identity card and live body user are same people.
2. the method for claim 1, it is characterised in that the collection user identity license piece of the S1 includes:
S1-1, by terminating layer login service layer service interface, service layer pushes chained address, and terminating layer downloads APP,
S1-2, the double-edged photo of identity card is shot in terminating layer, be sent to inspection in the lump with cell-phone number, IMEI, positional information
Test layer,
S1-3, inspected layer using OCR parse ID card information, extract and to terminating layer return name, ID card No., address,
Sex, Folk Information, if had more than at Liang Chu and two with real name, ID card No., address, sex, Folk Information
It is wrong, then S1-2 is returned, if wrong at only one, is corrected, and fed back correct name, ID card No., live to inspected layer
Location, sex, Folk Information,
S1-4, the correct name for being returned, ID card No., address, sex, Folk Information are sent to via inspected layer and are recognized
Card layer, authentication layers are authenticated from public security database gathering information, checking name, ID card No., address, sex, nationality
Whether information is corresponding and true, if certification does not pass through, returns failure information to terminating layer, if passing through, extracts user and exists
The identity card of public security database is achieved to register and is shone, and contrasts and judge identity card archive registration according to the body for shooting upload with terminating layer
Whether the face in part license piece is same people, if not, S1 is returned, if it is, carrying out S2;
3. the method for claim 1, it is characterised in that the S2's carries out vivo identification to user and include:
S2-1,3 composition set sequences of the random extraction from default behavior aggregate,
S2-2, the continuous action specified is completed before camera lens by voice message instruction user,
S2-3, computer disposal user action video, judge whether to complete required movement, if it is, to each from video flowing
1 photo is extracted in action, if not, return performing S2-1;
4. method as claimed in claim 2, it is characterised in that contrast in step S1-4 simultaneously judges that the identity card achieves registration
Whether it is that same people includes according to face in the identity card picture uploaded with terminating layer:1) extracted using active shape model algorithm
The characteristic point of face face:Canthus, eye center, the corners of the mouth, nose, chin profile;2) face is determined using the characteristic point extracted
The position of organ, selects the sample window of appropriate pixels size according to each organ size, extracts the sampling block of all subregion;3)
The identity card is achieved into registration sub-block is divided into according to sampling block according to the identity card picture uploaded with terminating layer, using the side of Block- matching
Method obtains sampling block matching degree, will try to achieve photo whole matching degree, for differentiating identity card after sampling block matching degree weighted average
Whether achieve registration according to the identity card picture uploaded with terminating layer is same person.
5. the method for claim 1, it is characterised in that contrast the identity card in step S3 and achieve registration and shine and live body
In identification per extract a photo, judge whether identity card and live body user are that same people includes:Using local feature reflection method,
Identity card is achieved registration takes and vivo identification in carry out the pretreatment of human face photo per an extraction picture, by face from photograph
Face edge in piece is plucked out, the distance between positioning pupil EA, pupil and face's lateral edges for being closed on apart from EB, nose
With the distance between two pupils NL、NR, the distance between nostril center NB, the distance between lip upper limb and nose LC, lip
The distance between lower edge and chin bottom LD, identity card is achieved registers according to corresponding in the identity card picture uploaded with terminating layer
EA/EA’、EB/EB’、NL/NL’、NR/NR’、NB/NB' Difference and two-by-two among big number ratio all within 2%, then calculate
LC/LC’、LD/LD', the ratio of its result with the two among big number ratio within 5%, then it is assumed that identity card achieves registration and shines
With in vivo identification per extract a photo be same people.
6. method as claimed in claim 2, it is characterised in that the terminating layer in step S1-2 is taken pictures including with PC or high photographing instrument
Or mobile phone photograph.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610898697.8A CN106529414A (en) | 2016-10-14 | 2016-10-14 | Method for realizing result authentication through image comparison |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610898697.8A CN106529414A (en) | 2016-10-14 | 2016-10-14 | Method for realizing result authentication through image comparison |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106529414A true CN106529414A (en) | 2017-03-22 |
Family
ID=58331923
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610898697.8A Pending CN106529414A (en) | 2016-10-14 | 2016-10-14 | Method for realizing result authentication through image comparison |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106529414A (en) |
Cited By (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106887091A (en) * | 2017-04-27 | 2017-06-23 | 石家庄瀚申网络科技有限公司 | It is a kind of go abroad, the self-service method handled of exit certificate and its system and device |
CN106981016A (en) * | 2017-03-30 | 2017-07-25 | 广东微模式软件股份有限公司 | A kind of remote self-help real name buys the method and system of phonecard |
CN106982426A (en) * | 2017-03-30 | 2017-07-25 | 广东微模式软件股份有限公司 | A kind of method and system for remotely realizing old card system of real name |
CN106991594A (en) * | 2017-03-30 | 2017-07-28 | 广东微模式软件股份有限公司 | A kind of long-range real name based on trade company buys the method and system of phonecard |
CN107016608A (en) * | 2017-03-30 | 2017-08-04 | 广东微模式软件股份有限公司 | The long-range account-opening method and system of a kind of identity-based Information Authentication |
CN107145873A (en) * | 2017-05-12 | 2017-09-08 | 江苏鸿信系统集成有限公司 | ID Card Image identification checking method and system based on recognition of face and OCR |
CN108009468A (en) * | 2017-10-23 | 2018-05-08 | 广东数相智能科技有限公司 | A kind of marathon race anti-cheat method, electronic equipment and storage medium |
CN108334966A (en) * | 2018-02-08 | 2018-07-27 | 国政通科技股份有限公司 | A kind of visitor's reserving method and system |
CN108600154A (en) * | 2018-03-05 | 2018-09-28 | 武汉高科国有控股集团有限公司 | A kind of government affairs remote authentication system and method |
CN108921080A (en) * | 2018-06-27 | 2018-11-30 | 北京旷视科技有限公司 | Image-recognizing method, device and electronic equipment |
CN108960195A (en) * | 2018-07-25 | 2018-12-07 | 中国建设银行股份有限公司 | Identity checking method and system |
CN109271915A (en) * | 2018-09-07 | 2019-01-25 | 北京市商汤科技开发有限公司 | False-proof detection method and device, electronic equipment, storage medium |
CN109409060A (en) * | 2018-09-26 | 2019-03-01 | 中国平安人寿保险股份有限公司 | Auth method, system and computer readable storage medium |
CN109410002A (en) * | 2018-10-29 | 2019-03-01 | 扬州笛莎服装科技有限公司 | The image collecting device of shops's customer interaction system Internet-based |
CN109446778A (en) * | 2018-10-30 | 2019-03-08 | 珠海市时杰信息科技有限公司 | Immovable Property Registration information acquisition method, computer installation and computer readable storage medium based on recognition of face |
CN109508702A (en) * | 2018-12-29 | 2019-03-22 | 安徽云森物联网科技有限公司 | A kind of three-dimensional face biopsy method based on single image acquisition equipment |
CN109522809A (en) * | 2018-10-22 | 2019-03-26 | 北京旷视科技有限公司 | Testimony of a witness checking method, server, all-in-one machine and system |
CN110097679A (en) * | 2019-05-15 | 2019-08-06 | 成都云联峰创科技有限公司 | A kind of electronic voting method |
CN110210414A (en) * | 2019-06-05 | 2019-09-06 | 北京京投信安科技发展有限公司 | The quick intersection identification technology of magnanimity face database |
CN110232322A (en) * | 2019-05-13 | 2019-09-13 | 平安科技(深圳)有限公司 | Verification method, device, computer equipment and the storage medium of identity card |
CN110336831A (en) * | 2019-07-18 | 2019-10-15 | 平安科技(深圳)有限公司 | A kind of account logon method, device and computer readable storage medium |
CN111538981A (en) * | 2020-04-24 | 2020-08-14 | 北京思特奇信息技术股份有限公司 | Certificate reading method and system for business hall |
CN111813987A (en) * | 2020-07-24 | 2020-10-23 | 台州市公安局黄岩分局 | Portrait comparison method based on police affair big data |
CN112424791A (en) * | 2018-07-25 | 2021-02-26 | 日本电气株式会社 | Information processing apparatus, information processing method, and information processing program |
CN112766927A (en) * | 2021-01-28 | 2021-05-07 | 成都晨云信息技术有限责任公司 | College new school entrance qualification examination method based on face recognition technology |
CN115001777A (en) * | 2022-05-25 | 2022-09-02 | 公安部户政管理研究中心 | Cross-service portal system management method |
CN115359544A (en) * | 2022-10-19 | 2022-11-18 | 广州卓腾科技有限公司 | Driver self-service physical examination equipment-based identity verification method, system and medium |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103440479A (en) * | 2013-08-29 | 2013-12-11 | 湖北微模式科技发展有限公司 | Method and system for detecting living body human face |
CN103489100A (en) * | 2013-10-22 | 2014-01-01 | 上海富友支付服务有限公司 | Cardless issuing card and electronic wallet application system and method based on conventional financial terminal |
CN103593598A (en) * | 2013-11-25 | 2014-02-19 | 上海骏聿数码科技有限公司 | User online authentication method and system based on living body detection and face recognition |
CN103886102A (en) * | 2014-04-09 | 2014-06-25 | 广东欧珀移动通信有限公司 | Method and system for inputting identity card number and terminal device |
CN104063690A (en) * | 2014-06-25 | 2014-09-24 | 广州卓腾科技有限公司 | Identity authentication method based on face recognition technology, device thereof and system thereof |
CN104751110A (en) * | 2013-12-31 | 2015-07-01 | 汉王科技股份有限公司 | Bio-assay detection method and device |
CN105426815A (en) * | 2015-10-29 | 2016-03-23 | 北京汉王智远科技有限公司 | Living body detection method and device |
CN105844206A (en) * | 2015-01-15 | 2016-08-10 | 北京市商汤科技开发有限公司 | Identity authentication method and identity authentication device |
CN105956162A (en) * | 2016-05-18 | 2016-09-21 | 陕西科技大学 | Electronic identity system capable of checking identity information in time |
CN105989263A (en) * | 2015-01-30 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Method for authenticating identities, method for opening accounts, devices and systems |
CN106022264A (en) * | 2016-05-19 | 2016-10-12 | 中国科学院自动化研究所 | Interactive face in vivo detection method and device based on multi-task self encoder |
-
2016
- 2016-10-14 CN CN201610898697.8A patent/CN106529414A/en active Pending
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103440479A (en) * | 2013-08-29 | 2013-12-11 | 湖北微模式科技发展有限公司 | Method and system for detecting living body human face |
CN103489100A (en) * | 2013-10-22 | 2014-01-01 | 上海富友支付服务有限公司 | Cardless issuing card and electronic wallet application system and method based on conventional financial terminal |
CN103593598A (en) * | 2013-11-25 | 2014-02-19 | 上海骏聿数码科技有限公司 | User online authentication method and system based on living body detection and face recognition |
CN104751110A (en) * | 2013-12-31 | 2015-07-01 | 汉王科技股份有限公司 | Bio-assay detection method and device |
CN103886102A (en) * | 2014-04-09 | 2014-06-25 | 广东欧珀移动通信有限公司 | Method and system for inputting identity card number and terminal device |
CN104063690A (en) * | 2014-06-25 | 2014-09-24 | 广州卓腾科技有限公司 | Identity authentication method based on face recognition technology, device thereof and system thereof |
CN105844206A (en) * | 2015-01-15 | 2016-08-10 | 北京市商汤科技开发有限公司 | Identity authentication method and identity authentication device |
CN105989263A (en) * | 2015-01-30 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Method for authenticating identities, method for opening accounts, devices and systems |
CN105426815A (en) * | 2015-10-29 | 2016-03-23 | 北京汉王智远科技有限公司 | Living body detection method and device |
CN105956162A (en) * | 2016-05-18 | 2016-09-21 | 陕西科技大学 | Electronic identity system capable of checking identity information in time |
CN106022264A (en) * | 2016-05-19 | 2016-10-12 | 中国科学院自动化研究所 | Interactive face in vivo detection method and device based on multi-task self encoder |
Non-Patent Citations (3)
Title |
---|
S1188837 上传: "基于几何特征的人脸识别", 《WWW.DOCIN.COM/P-1663662488.HTML》 * |
薛燕红: "《物联网导论》", 31 March 2014 * |
郭建 等: "基于特征块匹配的图像检索技术", 《计算机仿真》 * |
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106981016A (en) * | 2017-03-30 | 2017-07-25 | 广东微模式软件股份有限公司 | A kind of remote self-help real name buys the method and system of phonecard |
CN106982426A (en) * | 2017-03-30 | 2017-07-25 | 广东微模式软件股份有限公司 | A kind of method and system for remotely realizing old card system of real name |
CN106991594A (en) * | 2017-03-30 | 2017-07-28 | 广东微模式软件股份有限公司 | A kind of long-range real name based on trade company buys the method and system of phonecard |
CN107016608A (en) * | 2017-03-30 | 2017-08-04 | 广东微模式软件股份有限公司 | The long-range account-opening method and system of a kind of identity-based Information Authentication |
CN106887091A (en) * | 2017-04-27 | 2017-06-23 | 石家庄瀚申网络科技有限公司 | It is a kind of go abroad, the self-service method handled of exit certificate and its system and device |
CN107145873A (en) * | 2017-05-12 | 2017-09-08 | 江苏鸿信系统集成有限公司 | ID Card Image identification checking method and system based on recognition of face and OCR |
CN108009468A (en) * | 2017-10-23 | 2018-05-08 | 广东数相智能科技有限公司 | A kind of marathon race anti-cheat method, electronic equipment and storage medium |
CN108334966A (en) * | 2018-02-08 | 2018-07-27 | 国政通科技股份有限公司 | A kind of visitor's reserving method and system |
CN108600154A (en) * | 2018-03-05 | 2018-09-28 | 武汉高科国有控股集团有限公司 | A kind of government affairs remote authentication system and method |
CN108921080A (en) * | 2018-06-27 | 2018-11-30 | 北京旷视科技有限公司 | Image-recognizing method, device and electronic equipment |
CN108960195A (en) * | 2018-07-25 | 2018-12-07 | 中国建设银行股份有限公司 | Identity checking method and system |
CN112424791B (en) * | 2018-07-25 | 2024-03-19 | 日本电气株式会社 | Information processing device, information processing method, and information processing program |
CN112424791A (en) * | 2018-07-25 | 2021-02-26 | 日本电气株式会社 | Information processing apparatus, information processing method, and information processing program |
CN109271915A (en) * | 2018-09-07 | 2019-01-25 | 北京市商汤科技开发有限公司 | False-proof detection method and device, electronic equipment, storage medium |
CN109271915B (en) * | 2018-09-07 | 2021-10-08 | 北京市商汤科技开发有限公司 | Anti-counterfeiting detection method and device, electronic equipment and storage medium |
CN109409060A (en) * | 2018-09-26 | 2019-03-01 | 中国平安人寿保险股份有限公司 | Auth method, system and computer readable storage medium |
CN109522809A (en) * | 2018-10-22 | 2019-03-26 | 北京旷视科技有限公司 | Testimony of a witness checking method, server, all-in-one machine and system |
CN109522809B (en) * | 2018-10-22 | 2021-06-22 | 北京旷视科技有限公司 | People's identity verification method, server, all-in-one machine and system |
CN109410002A (en) * | 2018-10-29 | 2019-03-01 | 扬州笛莎服装科技有限公司 | The image collecting device of shops's customer interaction system Internet-based |
CN109446778A (en) * | 2018-10-30 | 2019-03-08 | 珠海市时杰信息科技有限公司 | Immovable Property Registration information acquisition method, computer installation and computer readable storage medium based on recognition of face |
CN109508702A (en) * | 2018-12-29 | 2019-03-22 | 安徽云森物联网科技有限公司 | A kind of three-dimensional face biopsy method based on single image acquisition equipment |
CN110232322A (en) * | 2019-05-13 | 2019-09-13 | 平安科技(深圳)有限公司 | Verification method, device, computer equipment and the storage medium of identity card |
CN110097679A (en) * | 2019-05-15 | 2019-08-06 | 成都云联峰创科技有限公司 | A kind of electronic voting method |
CN110210414A (en) * | 2019-06-05 | 2019-09-06 | 北京京投信安科技发展有限公司 | The quick intersection identification technology of magnanimity face database |
CN110336831A (en) * | 2019-07-18 | 2019-10-15 | 平安科技(深圳)有限公司 | A kind of account logon method, device and computer readable storage medium |
CN110336831B (en) * | 2019-07-18 | 2022-06-07 | 平安科技(深圳)有限公司 | Account login method and device and computer readable storage medium |
CN111538981A (en) * | 2020-04-24 | 2020-08-14 | 北京思特奇信息技术股份有限公司 | Certificate reading method and system for business hall |
CN111813987B (en) * | 2020-07-24 | 2024-03-08 | 台州市公安局黄岩分局 | Portrait comparison method based on police big data |
CN111813987A (en) * | 2020-07-24 | 2020-10-23 | 台州市公安局黄岩分局 | Portrait comparison method based on police affair big data |
CN112766927A (en) * | 2021-01-28 | 2021-05-07 | 成都晨云信息技术有限责任公司 | College new school entrance qualification examination method based on face recognition technology |
CN112766927B (en) * | 2021-01-28 | 2023-06-20 | 成都晨云信息技术有限责任公司 | Face recognition technology-based college entrance qualification checking method |
CN115001777A (en) * | 2022-05-25 | 2022-09-02 | 公安部户政管理研究中心 | Cross-service portal system management method |
CN115359544A (en) * | 2022-10-19 | 2022-11-18 | 广州卓腾科技有限公司 | Driver self-service physical examination equipment-based identity verification method, system and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106529414A (en) | Method for realizing result authentication through image comparison | |
US10650261B2 (en) | System and method for identifying re-photographed images | |
CN109377616B (en) | Access control system based on two-dimensional face recognition | |
WO2020155939A1 (en) | Image recognition method and device, storage medium and processor | |
US20170262472A1 (en) | Systems and methods for recognition of faces e.g. from mobile-device-generated images of faces | |
US20190138807A1 (en) | Iris liveness detection for mobile devices | |
WO2021036436A1 (en) | Facial recognition method and apparatus | |
US10922512B2 (en) | Contactless fingerprint recognition method using smartphone | |
CN108229427A (en) | A kind of identity-based certificate and the identity security verification method and system of recognition of face | |
US20100329568A1 (en) | Networked Face Recognition System | |
CN108470169A (en) | Face identification system and method | |
CN107438854A (en) | The system and method that the image captured using mobile device performs the user authentication based on fingerprint | |
CN106778607A (en) | A kind of people based on recognition of face and identity card homogeneity authentication device and method | |
CN110998598A (en) | Detection of manipulated images | |
WO2016084072A1 (en) | Anti-spoofing system and methods useful in conjunction therewith | |
CN103530659B (en) | With reference to the face identification method and attendance checking system of original and symmetrical face facial image | |
US10922399B2 (en) | Authentication verification using soft biometric traits | |
KR20170006355A (en) | Method of motion vector and feature vector based fake face detection and apparatus for the same | |
US9449217B1 (en) | Image authentication | |
CN110516649A (en) | Alumnus's authentication method and system based on recognition of face | |
CN106991448A (en) | A kind of portrait compares processing method | |
CN108446687A (en) | A kind of adaptive face vision authentication method based on mobile terminal and backstage interconnection | |
CN208351494U (en) | Face identification system | |
CN107944395A (en) | A kind of method and system based on neutral net verification testimony of a witness unification | |
CN104217504A (en) | Identity recognition self-service terminal and corresponding certificate of house property printing terminal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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: 20170322 |