CN109871755A - A kind of auth method based on recognition of face - Google Patents
A kind of auth method based on recognition of face Download PDFInfo
- Publication number
- CN109871755A CN109871755A CN201910022278.1A CN201910022278A CN109871755A CN 109871755 A CN109871755 A CN 109871755A CN 201910022278 A CN201910022278 A CN 201910022278A CN 109871755 A CN109871755 A CN 109871755A
- Authority
- CN
- China
- Prior art keywords
- user
- human face
- face photo
- descreening
- 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
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The embodiment of the invention discloses a kind of auth methods based on recognition of face, it include: the human face photo that identity-validation device acquires user by camera, detect whether the collected human face photo meets preset human face photo requirement, if the collected human face photo for meeting the human face photo requirement is then uploaded to Authentication server；The Authentication server obtains the band reticulate pattern certificate photograph of the user according to the user identifier of the user；The Authentication server inputs the descreening neural network model that training obtains in advance with reticulate pattern certificate photograph for the user's, obtains the certificate photograph of descreening；The certificate photograph of human face photo and the descreening that the identity-validation device uploads is compared and is calculated matching degree by the Authentication server, if matching degree reaches preset threshold between the two, is identified through the authentication of the user.Using the present invention, the percent of pass of authentication is improved.
Description
Technical field
The present invention relates to the communications field more particularly to a kind of auth methods based on recognition of face.
Background technique
Currently, the requirement to the authenticity of client, safety and confidence level such as many industry such as financial circles is more stringent, body
Part verification technique is also continuously improving, and user's face of the certificate photo human face photo and camera shooting that can pass through calling shines
The mode that piece compares carries out authentication, but uses the percent of pass of this alignments generally lower, because call
Certificate photo human face photo be that will affect identification with reticulate pattern, and user is when captured in real-time, it is understood that there may be the people of shooting
Face photo be not face full face, without expose complete face phenomena such as, there is also accidentally by, be not aligned with the problems such as position
It just takes pictures the phenomenon that uploading, and is manually operated and clicks the problems such as will appear faulty operation or blurring of photos when submitting photo, this
The key feature of sample, face cannot take well, cause to shoot the human face photo come and certificate photo human face photo not
Symbol, so that user needs repetitive operation that could pass through authentication.
Summary of the invention
The technical problem to be solved by the embodiment of the invention is that providing a kind of auth method based on recognition of face
And system, the percent of pass of authentication can be improved.
In order to solve the above-mentioned technical problem, first aspect of the embodiment of the present invention provides a kind of identity based on recognition of face
Verification method, comprising:
Identity-validation device acquires the human face photo of user by camera, whether detects the collected human face photo
Meet preset human face photo requirement, if the collected human face photo for meeting the human face photo requirement is then uploaded to body
Part authentication server, otherwise identity-validation device reminds user to adjust state of taking pictures, and the human face photo requires to include the people
Setting regions in face photo include the front face image of the user and include the user complete face image；
The Authentication server obtains the band reticulate pattern certificate photograph of the user according to the user identifier of the user；
The Authentication server inputs the descreening that training obtains in advance with reticulate pattern certificate photograph for the user's
Neural network model obtains the certificate photograph of descreening；
The certificate of human face photo and the descreening that the Authentication server uploads the identity-validation device
Photo compares and calculates matching degree, if matching degree reaches preset threshold between the two, is identified through the body of the user
Part verifying.
Wherein, inputting with reticulate pattern certificate photograph for the user is trained what is obtained to go by the Authentication server in advance
Before reticulate pattern neural network model further include:
The Authentication server collects reticulate pattern facial image and corresponding original facial image as sample image pair,
Form training dataset；
The Authentication server concentrates the reticulate pattern people for successively choosing each sample image centering from the training data
The reticulate pattern facial image is inputted descreening neural network model by face image, and the descreening facial image exported is described
Authentication server is according to the difference between each corresponding descreening facial image of sample image and original facial image
It is different, the parameter in the descreening neural network model is adjusted, the Authentication server iteration executes above-mentioned tune
It is had suffered journey, until meeting iterated conditional, determines the descreening neural network model that the training obtains.
Wherein, described that the collected human face photo for meeting the human face photo requirement is uploaded to Authentication server
Before further include:
The identity-validation device records the user and adjusts the collected human face photo during shooting state
Variable quantity；
If user's adjustment is taken pictures in state procedure, the variable quantity of the collected human face photo reaches default variation
Threshold value is measured, then the identity-validation device determines that the collected human face photo is living body photo.
Wherein, the method also includes:
If user's adjustment is taken pictures in state procedure, the variable quantity of the collected human face photo is not up to default to be become
Change amount threshold value, the then infra-red radiation that the identity-validation device acquisition user issues, is converted to pseudo- color thermal map, if according to institute
The profile for stating pseudo-colours thermal map judges it is face puppet color thermal map, then the identity-validation device confirms the collected people
Face photo is living body photo.
Wherein, the certificate photograph of the human face photo that the identity-validation device is uploaded and the descreening carries out pair
Than and calculate matching degree, comprising:
Five portions of eye, ear, nose, eyebrow, mouth of human face photo are obtained from the human face photo that the identity-validation device uploads
Subregion area image obtains five portions of eye, ear, nose, eyebrow, mouth of the certificate photograph of descreening from the certificate photograph of the descreening
Subregion area image, and the corresponding part area image in the human face photo and the certificate photograph of the descreening is compared
The matching degree between various pieces area image is obtained, by the correspondence portion of the human face photo and the certificate photograph of the descreening
Mean match degree between subregion area image is as the matching degree between the human face photo and the certificate photograph of the descreening.
Wherein, the Authentication server obtains the band reticulate pattern certificate of the user according to the user identifier of the user
Photo includes:
The Authentication server according to the user identifier of the user by preset certificate database service interface to
Trusted third party's certificate system sends certificate photograph acquisition request, and receives institute by the preset certificate database service interface
State the band reticulate pattern certificate photograph of the user of trusted third party's certificate system return.
Wherein, identity-validation device prompting user's adjustment state of taking pictures includes:
According to current collected human face photo, shooting angle or shooting between user's adjustment and the camera are prompted
The article of face is blocked in distance, or prompt user's removal.
Correspondingly, this motion second aspect provides a kind of system of authentication based on recognition of face, comprising:
Identity-validation device, for acquiring the human face photo of user and detecting whether the collected human face photo meets
Preset human face photo requirement, is tested if the collected human face photo for meeting the human face photo requirement is then uploaded to identity
Server is demonstrate,proved, otherwise identity-validation device reminds user to adjust state of taking pictures, and the human face photo requires to include the face and shines
Setting regions in piece include the front face image of the user and include the user complete face image；
Authentication server, for obtaining the band reticulate pattern certificate photograph of the user according to the user identifier of the user
And the user is inputted into the descreening neural network model that training obtains in advance with reticulate pattern certificate photograph, obtain descreening
Certificate photograph is also used to compare the certificate photograph of human face photo and the descreening that the identity-validation device uploads
And calculate matching degree.
The third aspect of the embodiment of the present invention provides a kind of identity-validation device, which includes processor, network interface
And memory, the processor, network interface and memory are connected with each other, wherein the network interface is by the processor
Control is used for messaging, and the memory is used to store the computer program for supporting terminal to execute the above method, the calculating
Machine program includes program instruction, and the processor is configured for calling described program instruction, executes the side of above-mentioned first aspect
Method.
Fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage medium
Matter is stored with computer program, and the computer program includes program instruction, and described program instruction makes when being executed by a processor
The method that the processor executes above-mentioned first aspect.
The embodiment of the present invention only meets the people by the collected human face photo of camera in identity-validation device
Face photo requires to upload, and effectively avoiding the photo that shooting uploads is not front face image, without complete five official rank
The generation of phenomenon, it is thus also avoided that because manual operation click submit photo when occur faulty operation or caused by blurring of photos phenomena such as
Occur, the human face photo has also carried out In vivo detection before being uploaded to Authentication server, can further demonstrate
It is the sequence of operations carried out by biologically active human body, avoids criminal using photo, video, three-dimensional mask etc.
Malice is played tricks act of authentication, calls certificate photo human face photo to carry out descreening processing from trusted third party, to effectively increase
The matching degree for shooting the human face photo and certificate photo human face photo that come, improves the confidence level and percent of pass of authentication.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow chart of auth method based on recognition of face provided in an embodiment of the present invention；
Fig. 2 be it is provided in an embodiment of the present invention judge human face photo whether be living body photo flow chart；
Fig. 3 is the schematic diagram of the setting regions of human face photo provided in an embodiment of the present invention；
Fig. 4 is the flow chart provided in an embodiment of the present invention about training descreening neural network model；
Fig. 5 is provided in an embodiment of the present invention about the human face photo progress that will shoot obtained human face photo and descreening
Compare matched flow chart；
Fig. 6 be it is provided in an embodiment of the present invention about obtain feature vector and compare obtain the flow chart of matching degree；
Fig. 7 is a kind of schematic diagram of authentication system based on recognition of face provided in an embodiment of the present invention；
Fig. 8 is the schematic diagram of the profile diagram of photo provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Referring to Figure 1, Fig. 1 is a kind of auth method based on recognition of face that one embodiment of the present of invention provides
Flow chart, comprising:
S101, identity-validation device acquire the human face photo of user by camera.
Wherein, the identity-validation device for example can be mobile phone, tablet computer, laptop, palm PC, movement
Internet device (MID, mobile internet device), wearable device (such as smartwatch (such as iwatch), intelligence
Energy bracelet, pedometer etc.) or other mountable equipment for disposing authentication applications clients.The identity-validation device to
Family, which carries out authentication, can be used for great meeting, movable safety check, security work, can be also used for important in daily life
Place such as station, airport, high speed or Daily Round Check etc..For example, criminal does evil in order to prevent when holding great meeting,
Authentication can be carried out to admission personnel using identity-validation device, permit its admission after confirming identity.
Whether S102, the detection collected human face photo meet pre-set human face photo requirement.
Wherein, the human face photo requires to include the positive dough figurine that the setting regions in the human face photo includes the user
Face image and include the user complete face image, can also include that require collected human face region size to meet pre-
If it is required that, face direction meets preset requirement or requires to reach default between collected human face region profile and photo acquisition frame
Degree of approximation etc..As shown in figure 3, the setting regions of the human face photo can be the photo acquisition of the identity-validation device
Frame, the human face photo require the photo acquisition frame for the authentication to include the face full face of the user and wrap
Complete face containing the user, the human face region length and width dimensions of collected user and the length and width dimensions difference of photo acquisition frame
Less than preset threshold.
If S103, collected human face photo do not meet pre-set human face photo requirement, the identity-validation device
User's adjustment is reminded to take pictures state until collected human face photo is uploaded to body after meeting pre-set human face photo requirement
Part authentication server.
Wherein, if it does not include the complete of user that the prompting user, which adjusts in the human face photo that the state of taking pictures includes: acquisition,
Face, the identity-validation device may remind the user that attention avoids blocking the face of itself；If the human face photo face of acquisition
When too large or too small, the identity-validation device reminds user to adjust itself distance apart from camera；If the face of acquisition
When photo face skew, the identity-validation device reminds user terminal face portion, etc..Described remind includes text prompting, language
One or more combinations of sound prompting, image prompting etc..
S104, the collected human face photo for meeting the human face photo requirement is uploaded to Authentication server.
S105, the Authentication server obtain the band reticulate pattern certificate of the user according to the user identifier of the user
Photo.
Wherein, the Authentication server is connect according to the user identifier of the user by preset certificate data service
Mouth sends certificate photograph acquisition request to the certificate system of trusted third party, and passes through the preset certificate database service interface
Receive certificate photograph of the user with reticulate pattern that trusted third party's certificate system returns.The user identifier can be body
Part card number, fingerprint, iris etc..The certificate photograph with reticulate pattern include: identity card human face photo, social security card human face photo,
The certificate photograph with reticulate pattern such as pass human face photo, correspondingly, trusted third party's certificate system includes: Ministry of Public Security's certificate
System, human resources and Department of Social Security's certificate system, entry-exit management portion certificate system etc., such as the authentication service
Device can obtain the band reticulate pattern identity card human face photo of user by the certificate database service interface of Ministry of Public Security's certificate system, or logical
The certificate database service interface for crossing human resources and Department of Social Security's certificate system obtains shining with reticulate pattern social security card face for user
Piece, the band reticulate pattern pass face that user can also be obtained by the certificate database service interface of entry-exit management portion certificate system
Photo.
Inputting with reticulate pattern certificate photograph for the user is trained what is obtained to go by S106, the Authentication server in advance
Reticulate pattern neural network model obtains the certificate photograph of descreening.
The full convolutional neural networks model that preset training obtains or mixed can be used in the descreening neural network model
Close convolutional neural networks model.It in an alternative embodiment, can be by collecting reticulate pattern facial image and corresponding original face figure
Training one is used to restore clear face image from reticulate pattern image as being used as sample image pair, and using these sample images
The band reticulate pattern certificate photograph of the current collected user is inputted process training by the full convolutional neural networks of (descreening)
Obtained descreening neural network model, can be obtained the certificate photograph of descreening.
The human face photo that S107, the Authentication server upload the identity-validation device and the descreening
Certificate photograph compares and calculates matching degree.
The matching degree can indicate the similarity between the human face photo and the certificate photograph of the descreening, i.e., instead
Reflect the identity coherence when preceding camera acquisition target user certificate photograph corresponding with the user identifier.
In an alternative embodiment, obtained from the human face photo that the identity-validation device uploads the eye of human face photo, ear,
Five nose, eyebrow, mouth partial region images, obtained from the certificate photograph of the descreening eye of certificate photograph of descreening, ear,
Five nose, eyebrow, mouth partial region images, and by the corresponding part area in the human face photo and the certificate photograph of the descreening
Area image compares to obtain the matching degree between various pieces area image, by the card of the human face photo and the descreening
Certificate photo of the Mean match degree as the human face photo and the descreening between the corresponding part area image of part photo
Matching degree between piece.
S108, the Authentication server are according to the matching degree and matching degree threshold comparison, Xiang Suoshu authentication
Equipment sends verification result, if the matching degree reaches matching degree threshold value, verification result is to be verified, otherwise verification result
Do not pass through for verifying.
S109, the Authentication server transmit verification result to the identity-validation device.
S110, the identity-validation device show verification result.
Wherein, the identity-validation device show verification result can be the screen of the identity-validation device with text or
The mode of image shows authentication as a result, can also be that the identity-validation device passes through voice informing authentication result.
Such as when user's great meeting of participation, the staff of meeting can carry out authentication to personnel participating in the meeting to prevent criminal from joining
Criminal offence is carried out with meeting, the identity-validation device acquires the face that the user meets the human face photo requirement
The Authentication server is uploaded to after photo, the Authentication server is called according to user identifier such as identification card number
The certificate photograph with reticulate pattern of the user, after the certificate photograph with reticulate pattern is carried out descreening processing, by the identity
The human face photo and the certificate photograph of calling that verifying equipment uploads compare and calculate matching degree, if the matching degree reaches
With degree threshold value, then the Authentication server, which will be verified, is sent to the identity-validation device, and the authentication is set
Standby screen, which is shown, to be verified, and user can enter meeting.
In the present embodiment, the collected human face photo of camera in the identity-validation device only meets the people
Face photo requires to upload, and effectively avoiding the photo that shooting uploads is not front face image, without complete five official rank
The generation of phenomenon, it is thus also avoided that because manual operation click submit photo when occur faulty operation or caused by blurring of photos phenomena such as
Occur, call certificate photo human face photo to carry out descreening processing from trusted third party, to effectively increase the people for shooting and
The matching degree of face photo and certificate photo human face photo improves the confidence level and percent of pass of authentication.
In an alternative embodiment, the S104 in foregoing embodiments, that is, identity-validation device uploads human face photo to body
It can also include the process that In vivo detection is carried out to human face photo before part authentication server, including as shown in Figure 2 to flow down
Journey:
S201, identity-validation device acquire human face photo by camera.
S202, the identity-validation device remind user to adjust state of itself taking pictures until the collected human face photo
Meet the human face photo requirement.
Wherein, if collected human face photo does not meet the photo requirement, the identity-validation device reminds user's tune
Whole state of itself taking pictures, wherein if the human face photo that the prompting user adjustment state of taking pictures includes: acquisition, which exists, blocks face
The phenomenon that when, the identity-validation device reminds user not block itself face；If the human face photo face of acquisition are inadequate
When completely or the face of human face photo is excessive or too small, the identity-validation device reminds user to adjust itself apart from camera
Distance；If the identity-validation device reminds user terminal face portion when the human face photo face skew of acquisition.The prompting packet
Include one or more combinations of text prompting, voice reminder, image prompting etc..
S203, the identity-validation device record the variable quantity of collected human face photo.
Wherein, the variable quantity of the collected human face photo can started pair by calculating the identity-validation device
First human face photo collected to the user and the people for being determined for compliance with human face photo requirement after user's progress authentication
Variable quantity between face photo obtains, and can also start to carry out authentication to user by recording the identity-validation device
First human face photo P1 collected to the user is tested with because human face photo does not meet identity described in the photo requirement afterwards
Card equipment reminds user to adjust the variable quantity between the human face photo P2 shot after shooting state, P2 and because of human face photo for the first time
Not meeting second of identity-validation device described in the photo requirement reminds user to adjust the human face photo shot after shooting state
Variable quantity ... the identity-validation device (n-1)th time between the P3 human face photo Pn-1 and the body for reminding to shoot after user
Part verifying equipment n-th reminds the change between the human face photo Pn for being determined for compliance with the human face photo requirement taken after user
Change amount, and calculate variable quantity of the sum of the variable quantity recorded in the above process as the collected human face photo.Into
And in an alternative embodiment, the variable quantity can be the change in location sum of the distance or equal of the face corresponding part of human face photo
Value.
S204, judge whether the variable quantity of the human face photo reaches variable quantity threshold value.
If the variable quantity of the collected human face photo reaches variable quantity threshold value, it may be considered that current collected people
Face photo is living body photo, otherwise needs to judge by another way whether current collected human face photo is living body photo.
If S205, the user take pictures in state procedure in adjustment, the variable quantity of the collected human face photo does not reach
To default variable quantity threshold value, then the infra-red radiation that the identity-validation device acquisition user issues is converted to pseudo- color thermal map.
Wherein, the infra-red radiation that the identity-validation device acquisition user issues, the photoelectricity of the identity-validation device are red
After the power signal for the infra-red radiation that user issues is converted into electric signal by external detector, the imaging of the identity-validation device is filled
Set can simulation go out the spatial distribution of user's face surface temperature, obtain puppet corresponding with the surface heat distribution of user's face
Color thermal map.
S206, face puppet color thermal map is judged whether it is according to the profile of the pseudo- color thermal map.
Wherein, the profile of the pseudo- color thermal map identity-validation device obtained and preset face puppet color thermal map
Profile compares, if similarity reaches similarity threshold, judgement is face puppet color thermal map, then confirms that the human face photo is
Otherwise living body photo re-starts the acquisition of human face photo.
If S207, the variable quantity reach default variable quantity threshold value, confirm that the human face photo is living body photo.
In the present embodiment, the identity-validation device confirms that the human face photo is after living body photo again by the face
Photo upload is played tricks using malice such as photo, video, three-dimensional masks so as to avoid criminal and is recognized to Authentication server
Card behavior.
In an alternative embodiment, the training process packet of the descreening neural network model in the S106 in foregoing embodiments
Include following below scheme as shown in Figure 4:
S401, the Authentication server collect reticulate pattern facial image and corresponding original facial image as sample graph
As right, formation training dataset；
S402, the Authentication server concentrate the net for successively choosing each sample image centering from the training data
The reticulate pattern facial image is inputted descreening neural network model, the descreening facial image exported by line facial image；
S403, the Authentication server are according to the corresponding descreening facial image of each sample image and original
Difference between facial image is adjusted the weight parameter in the descreening neural network model；
Wherein it is possible to be adjusted by gradient anti-pass to the weight parameter in the descreening neural network model.
Wherein, the weight parameter in the descreening neural network model includes peak signal-to-noise ratio value, pixel value etc..
S404, the Authentication server iteration execute above-mentioned adjustment process, until meeting iterated conditional；
Wherein, the iterated conditional may include that Y-PSNR reaches preset Y-PSNR threshold value.The peak value
Signal-to-noise ratio is for judging the similarity between the descreening image and original image, and the Y-PSNR is higher, descreening
Similarity between image and original image is higher.
For example, descreening neural network model can be trained according to 100,000 band reticulate pattern images, if iteration
Obtained descreening neural network model reaches 45 to the peak signal-to-noise ratio value that moire pattern is restored, then can be with
Think to meet iterated conditional, stop above-mentioned iteration adjustment, obtains the descreening neural network model obtained by training.
S405, the descreening neural network model that the training obtains is determined.
It is understood that after obtaining the obtained descreening model of training, the Authentication server is by the user
The obtained descreening neural network model of training in advance of the certificate photograph input with reticulate pattern, obtain the certificate photograph of descreening,
Thus facilitate and carries out the next matching with human face photo that is being uploaded to Authentication server.
In an alternative embodiment, the S107 in one embodiment, that is, Authentication server tests the identity
The card equipment human face photo uploaded and the certificate photograph of the descreening compare and calculate matching degree, including as shown in Figure 5
Following below scheme:
S501, eye, ear, nose, eyebrow, the mouth five that human face photo is obtained from the human face photo that the identity-validation device uploads
A partial region image；
S502, from the certificate photograph of the descreening obtain descreening certificate photograph five, eye, ear, nose, eyebrow, mouth
Partial region image；
S503, the corresponding part area image in the human face photo and the certificate photograph of the descreening is compared
The matching degree between various pieces area image is calculated；
S504, by being averaged between the human face photo and the corresponding part area image of the certificate photograph of the descreening
Matching degree is as the matching degree between the human face photo and the certificate photograph of the descreening.If matching degree reaches between the two
Preset threshold is then identified through the authentication of the user, does not otherwise pass through the authentication of the user.
In an alternative embodiment, the S503 in foregoing embodiments is i.e. by the certificate of the human face photo and the descreening
Corresponding part area image in photo compares the matching degree being calculated between various pieces area image, including as schemed
Following below scheme shown in 6:
S601, the Authentication server read the correspondence in the human face photo and the certificate photograph of the descreening
Partial region image.
Wherein, the partial region image is to obtain human face photo from the human face photo that the identity-validation device uploads
Eye, ear, nose, eyebrow, five partial region images of mouth and from the certificate photograph of the descreening obtain descreening certificate photo
Five partial region images of eye, ear, nose, eyebrow, mouth of piece；
S602, judge whether the corresponding region image is grayscale image.
Wherein, the grayscale image is the image indicated with gray scale, containing only luminance information, is free of color information, the gray scale
The brightness of figure is by secretly to bright, brightness change is continuous.
If S603, the corresponding region image are not grayscale images, the corresponding region image is converted into grayscale image.
Wherein, if the color of certain original point of the color of image point is that RGB (R, G, B) i.e. image is independent red by three
Color, green and blue primary components composition.So, we can be converted into gray scale by following several method:
1. floating-point arithmetic: Gray=R*0.3+G*0.59+B*0.11；
2. displacement method: Gray=(R*76+G*151+B*28) > > 8；
3. mean value method: Gray=(R+G+B)/3；
4. only taking green: Gray=G.
After the Gray value for acquiring each pixel in image by any of the above-described kind of method, by the original RGB of each pixel (R,
G, B) in R, G, B are unified to be replaced with Gray value, forms the new color RGB (Gray, Gray, Gray) of respective pixel to obtain
Grayscale image.
S604, the profile for calculating object in grayscale image, obtain a profile diagram.
Wherein, the method for obtaining contour of object in grayscale image may include: after carrying out hot-tempered processing to image, using single order
The finite difference of local derviation calculates the amplitude of the gradient of each pixel and direction in grayscale image, and to the gradient magnitude of each pixel
Non-maxima suppression is carried out, chooses high threshold th and Low threshold tl, ratio can be 2:1 or 3:1.After taking out non-maxima suppression
Image in greatest gradient amplitude Max, redefine high-low threshold value.That is: high threshold is TH=th × Max, Low threshold TL
=tl × Max.Point by the gradient magnitude of pixel less than TL is abandoned；Point by the gradient magnitude of pixel greater than TH, which is labeled, to be used as
The gradient magnitude of pixel is then greater than TL and is less than in the pixel of TH, is greater than with the gradient magnitude of pixel by contour edge point
The adjacent pixel of the pixel of TH be determined as contour edge point (such as pixel gradient magnitude greater than TH pixel it is adjacent
Pixel around upper and lower, left and right, upper left, upper right, lower-left, bottom right 8 pixels may be considered corresponding pixel points
Neighbor pixel), and then the contour edge point obtained according to the above process can determine to obtain the profile diagram of object in grayscale image.
Referring to FIG. 8, Fig. 8 is the profile diagram for obtaining a photo as example.
S605, the gray scale center for calculating separately profile diagram.
Wherein, the gray scale center refers to the center of energy point for being considered gray value in image, wherein the gray scale center
It is calculated by such as minor function:
Wherein, (r0, c0) it is the gray scale center, i indicates that the row in image, j indicate the column and g in imageijIndicate figure
The gray value of the point of the i-th row jth column as in.
S606, profile diagram will be divided into respectively to n parts of images centered on the gray scale center.
S607, the grey density values for calculating separately every a image.
Wherein, the grey density values of every a image pass through minor function such as and calculate:
Wherein, vkFor the grey density values of a copy of it image, (i', j') is the point in this part of image, and (i, j) is entire
Point in image.ri'j'Indicate that point (i', j') arrives the distance at above-mentioned gray scale center, rijIndicate that point (i, j) arrives above-mentioned gray scale center
Distance, gi'j'Indicate the gray value and g of point (i', j')ijIndicate the gray value of point (i, j).
S608, the n grey density values calculated separately are formed to a matrix as the feature vector of image, comparison
Two feature vectors obtain matching degree.
Using the partial region image of the human face photo as image A, by the part area of the certificate photograph of the descreening
Area image is respectively u and v as the feature vector of image B, described image A and described image B.Wherein, feature vector u and feature
Vector v is all that n is tieed up, i.e., all includes n grey density values.N described in the present embodiment is the positive integer more than or equal to 1.
First read the grey density values v in the feature vector u of image AaiIn the feature vector v of (i=1~n) and image B
Grey density values vbi(i=1~n).Later, the images match degree computing module in the identity-validation device utilizes function gamma
(vai, vbi)=100- | vai-vbi|÷((vai+vbi) ÷ 2) * 1000) calculate grey density values vaiWith vbiMatching degree γ
(vai, vbi), thus obtain n matching degree, i.e. γ (va1, vb1)...γ(van, vbn).Then, described image matching degree calculates mould
Block utilizes function β (n, u, v)=[∑ γ (vai, vbi)] ÷ n calculate image A and image B feature vector matching degree.
One embodiment of the present of invention additionally provides a kind of system of authentication based on recognition of face.
In an alternative embodiment, referring to FIG. 7, the authentication system based on recognition of face includes that authentication is set
Standby 701 and Authentication server 702.
The identity-validation device 701 acquire human face photo and detect the collected human face photo whether meet it is default
Human face photo requirement remind user that adjustment takes pictures state until collected face if not meeting the human face photo requirement
Photo meets the human face photo requirement.
Wherein, the human face photo requires to include the positive dough figurine that the setting regions in the human face photo includes the user
Face image and include the user complete face image.
The human face photo for meeting the human face photo requirement is uploaded to the authentication by the identity-validation device 701
Server 702.
The Authentication server 702 obtains the band reticulate pattern certificate photo of the user according to the user identifier of the user
The user is simultaneously inputted the descreening neural network model that training obtains in advance with reticulate pattern certificate photograph by piece, obtains descreening
Certificate photograph, the certificate photograph of the human face photo for being also used to upload the identity-validation device and the descreening carries out pair
Than and calculate matching degree.The Authentication server 702 is according to the matching degree and matching degree threshold comparison, Xiang Suoshu identity
It verifies equipment and sends verification result, if the matching degree reaches matching degree threshold value, verification result is to be verified, and is otherwise verified
As a result do not pass through for verifying.
The Authentication server 702 transmits verification result to the identity-validation device.
The identity-validation device 701 shows verification result.
It is understood that the system of the authentication based on recognition of face in the present embodiment can be executed as tied above
Steps flow chart some or all of in auth method described in any one embodiment that conjunction Fig. 1-Fig. 6 is introduced.
Provide a kind of identity-validation device in an embodiment of the present invention, which is characterized in that including processor and
Memory, the processor and the memory are connected with each other, wherein the memory is described for storing computer program
Computer program includes program instruction, and the processor is configured for calling described program instruction.
The processor, for detecting whether the collected human face photo meets preset human face photo requirement, if
It is the collected human face photo for meeting the human face photo requirement to be uploaded to Authentication server, otherwise authentication
Equipment reminds user's adjustment to take pictures state until collected human face photo meets the photo requirement.
The processor is also used to for the matching degree being compared with preset matching degree threshold value, if the matching degree reaches
Matching degree threshold value, then the identity-validation device shows that authentication passes through, if the matching degree is not up to preset matching degree threshold
Value, then the identity-validation device shows that authentication does not pass through.
During the memory is for recording user's adjustment shooting state, the change of the collected human face photo
Change amount.
It should be appreciated that in embodiments of the present invention, the processor can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic
Device, discrete gate or transistor logic, discrete hardware components etc..General processor can be microprocessor or this at
Reason device is also possible to any conventional processor etc..
The memory may include read-only memory and random access memory, and provide instruction sum number to processor
According to.The a part of of memory can also include nonvolatile RAM.For example, memory can also store equipment class
The information of type.
It is understood that the identity-validation device in the present embodiment can be executed as introduced previously in conjunction with Fig. 1-Fig. 6
Any one embodiment described in auth method identity-validation device some or all of execute steps flow chart.
A kind of computer readable storage medium, the computer-readable storage medium are provided in an embodiment of the present invention
Matter is stored with computer program, and the computer program includes program instruction, realization when described program instruction is executed by processor
Described program instruction is executed by processor the authentication as described in any one embodiment introduced previously in conjunction with Fig. 1-Fig. 6
Some or all of identity-validation device execution steps flow chart in method.
It deposits the inside that the computer readable storage medium can be identity-validation device described in aforementioned any embodiment
Storage unit, such as the hard disk or memory of identity-validation device.The computer readable storage medium is also possible to the identity and tests
Demonstrate,prove the External memory equipment of equipment, such as the plug-in type hard disk being equipped on the identity-validation device, intelligent memory card (Smart
Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card) etc..Further,
The computer readable storage medium can also both include the internal storage unit of the identity-validation device or deposit including outside
Store up equipment.The computer readable storage medium is for storing needed for the computer program and the identity-validation device
Other programs and data.The computer readable storage medium, which can be also used for temporarily storing, have been exported or will export
Data.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random Access
Memory, RAM) etc..
The above disclosure is only the preferred embodiments of the present invention, cannot limit the right model of the present invention with this certainly
It encloses, therefore equivalent changes made in accordance with the claims of the present invention, is still within the scope of the present invention.
Claims (10)
1. a kind of auth method based on recognition of face characterized by comprising
Identity-validation device acquires the human face photo of user by camera, detects whether the collected human face photo meets
Preset human face photo requirement, is tested if the collected human face photo for meeting the human face photo requirement is then uploaded to identity
Server is demonstrate,proved, otherwise identity-validation device reminds user to adjust state of taking pictures, and the human face photo requires to include the face and shines
Setting regions in piece include the front face image of the user and include the user complete face image；
The Authentication server obtains the band reticulate pattern certificate photograph of the user according to the user identifier of the user；
The Authentication server inputs the descreening nerve that training obtains in advance with reticulate pattern certificate photograph for the user's
Network model obtains the certificate photograph of descreening；
The certificate photograph of human face photo and the descreening that the Authentication server uploads the identity-validation device
Matching degree is compared and calculates, if matching degree reaches preset threshold between the two, the identity for being identified through the user is tested
Card.
2. as described in claim 1 based on the auth method of recognition of face, which is characterized in that the authentication service
Device inputs the user's before the descreening neural network model that training obtains in advance with reticulate pattern certificate photograph further include:
The Authentication server collects reticulate pattern facial image and corresponding original facial image as sample image pair, is formed
Training dataset；
The Authentication server concentrates the reticulate pattern face figure for successively choosing each sample image centering from the training data
The reticulate pattern facial image is inputted descreening neural network model, the descreening facial image exported by picture；
The Authentication server is according to each corresponding descreening facial image of sample image and original facial image
Between difference, the parameter in the descreening neural network model is adjusted；
The Authentication server iteration executes above-mentioned adjustment process and determines the training until meeting iterated conditional
Obtained descreening neural network model.
3. as described in claim 1 based on the auth method of recognition of face, which is characterized in that described by collected symbol
The human face photo for closing the human face photo requirement is uploaded to before Authentication server further include:
The identity-validation device records the variation that the user adjusts the collected human face photo during shooting state
Amount；
If user's adjustment is taken pictures in state procedure, the variable quantity of the collected human face photo reaches default variable quantity threshold
Value, then the identity-validation device determines that the collected human face photo is living body photo.
4. as claimed in claim 3 based on the auth method of recognition of face, which is characterized in that the method also includes:
If user's adjustment is taken pictures in state procedure, the not up to default variable quantity of the variable quantity of the collected human face photo
Threshold value, the then infra-red radiation that the identity-validation device acquisition user issues, is converted to pseudo- color thermal map, if according to the puppet
The profile of colored thermal map judges it is face puppet color thermal map, then the identity-validation device confirmation collected face shines
Piece is living body photo.
5. as described in claim 1 based on the auth method of recognition of face, which is characterized in that described to test the identity
The card equipment human face photo uploaded and the certificate photograph of the descreening compare and calculate matching degree, comprising:
Eye, ear, nose, the eyebrow, five parts of mouth area of human face photo are obtained from the human face photo that the identity-validation device uploads
Area image obtains eye, ear, nose, the eyebrow, five parts of mouth area of the certificate photograph of descreening from the certificate photograph of the descreening
Area image, and the corresponding part area image in the human face photo and the certificate photograph of the descreening is compared to obtain
Matching degree between various pieces area image, by the corresponding part area of the human face photo and the certificate photograph of the descreening
Mean match degree between area image is as the matching degree between the human face photo and the certificate photograph of the descreening.
6. as described in claim 1 based on the auth method of recognition of face, which is characterized in that the authentication service
Device includes: according to the certificate photograph with reticulate pattern that the user identifier of the user obtains the user
The Authentication server is according to the user identifier of the user by preset certificate database service interface to credible
Third party's certificate system sends certificate photograph acquisition request, and passing through can described in the preset certificate database service interface reception
Believe the band reticulate pattern certificate photograph for the user that third party's certificate system returns.
7. as described in claim 1 based on the auth method of recognition of face, which is characterized in that the identity-validation device
Prompting user adjusts the state of taking pictures
According to current collected human face photo, prompt shooting angle between user's adjustment and the camera or shooting away from
From, or the article for prompting user's removal to block face.
8. a kind of system of the authentication based on recognition of face characterized by comprising
Identity-validation device, for acquire the human face photo of user and detect the collected human face photo whether meet it is default
Human face photo requirement, if then by the collected human face photo for meeting the human face photo requirement be uploaded to authentication clothes
Business device, otherwise identity-validation device reminds user to adjust state of taking pictures, and the human face photo requires to include in the human face photo
Setting regions include the user front face image and include the user complete face image；
Authentication server, for obtaining the band reticulate pattern certificate photograph of the user according to the user identifier of the user and inciting somebody to action
The user's inputs the descreening neural network model that training obtains in advance with reticulate pattern certificate photograph, obtains the certificate of descreening
Photo is also used to that the certificate photograph of human face photo and the descreening that the identity-validation device uploads is compared and counted
Calculate matching degree.
9. a kind of identity-validation device, which is characterized in that including processor and memory, the processor and the memory phase
It connects, wherein the memory is for storing computer program, and the computer program includes program instruction, the processing
Device is configured for calling described program instruction, executes the method according to claim 1 to 7.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, the computer program include program instruction, and described program instruction executes the processor such as
The described in any item methods of claim 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910022278.1A CN109871755A (en) | 2019-01-09 | 2019-01-09 | A kind of auth method based on recognition of face |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910022278.1A CN109871755A (en) | 2019-01-09 | 2019-01-09 | A kind of auth method based on recognition of face |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109871755A true CN109871755A (en) | 2019-06-11 |
Family
ID=66917583
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910022278.1A Pending CN109871755A (en) | 2019-01-09 | 2019-01-09 | A kind of auth method based on recognition of face |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109871755A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110647805A (en) * | 2019-08-09 | 2020-01-03 | 平安科技（深圳）有限公司 | Reticulated image identification method and device and terminal equipment |
| CN111062248A (en) * | 2019-11-08 | 2020-04-24 | 宇龙计算机通信科技(深圳)有限公司 | Image detection method, device, electronic equipment and medium |
| CN111191567A (en) * | 2019-12-26 | 2020-05-22 | 深圳供电局有限公司 | Identity data processing method and device, computer equipment and storage medium |
| CN111339885A (en) * | 2020-02-19 | 2020-06-26 | 平安科技（深圳）有限公司 | User identity determination method based on iris recognition and related device |
| CN111405365A (en) * | 2020-03-12 | 2020-07-10 | 北京文香信息技术有限公司 | Identity authentication method, device, system and storage medium |
| CN111444831A (en) * | 2020-03-25 | 2020-07-24 | 深圳中科信迅信息技术有限公司 | Method for recognizing human face through living body detection |
| CN111757042A (en) * | 2020-06-28 | 2020-10-09 | 深圳市闪联信息技术有限公司 | Remote collaborative conference method and system based on face authentication |
| WO2021042544A1 (en) * | 2019-09-03 | 2021-03-11 | 平安科技（深圳）有限公司 | Facial verification method and apparatus based on mesh removal model, and computer device and storage medium |
| CN113487463A (en) * | 2021-05-25 | 2021-10-08 | 中国联合网络通信集团有限公司 | Method and system for managing flowing object |
| CN114241588A (en) * | 2022-02-24 | 2022-03-25 | 北京锐融天下科技股份有限公司 | Self-adaptive face comparison method and system |
| TWI774997B (en) * | 2019-10-05 | 2022-08-21 | 中國信託商業銀行股份有限公司 | Risk assessment method and system, service system and computer program product |
| CN117172783A (en) * | 2023-07-17 | 2023-12-05 | 湖北盈嘉集团有限公司 | Credit and debt cross checking system for confirming account receivables |
| CN111339885B (en) * | 2020-02-19 | 2024-05-28 | 平安科技（深圳）有限公司 | User identity determining method and related device based on iris recognition |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN103634120A (en) * | 2013-12-18 | 2014-03-12 | 上海市数字证书认证中心有限公司 | Method and system for real-name authentication based on face recognition |
| CN105243357A (en) * | 2015-09-15 | 2016-01-13 | 深圳市环阳通信息技术有限公司 | Identity document-based face recognition method and face recognition device |
| CN106548159A (en) * | 2016-11-08 | 2017-03-29 | 中国科学院自动化研究所 | Reticulate pattern facial image recognition method and device based on full convolutional neural networks |
| CN106778525A (en) * | 2016-11-25 | 2017-05-31 | 北京旷视科技有限公司 | Identity identifying method and device |
| CN107766844A (en) * | 2017-11-13 | 2018-03-06 | 杭州有盾网络科技有限公司 | Method, apparatus, equipment of a kind of reticulate pattern according to recognition of face |
| CN107818308A (en) * | 2017-10-31 | 2018-03-20 | 平安科技（深圳）有限公司 | A kind of recognition of face intelligence comparison method, electronic installation and computer-readable recording medium |
| CN108171032A (en) * | 2017-12-01 | 2018-06-15 | 平安科技（深圳）有限公司 | A kind of identity identifying method, electronic device and computer readable storage medium |
| CN108197557A (en) * | 2017-12-28 | 2018-06-22 | 深圳云天励飞技术有限公司 | Testimony of a witness consistency check method, terminal device and computer readable storage medium |
-
2019
- 2019-01-09 CN CN201910022278.1A patent/CN109871755A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN103634120A (en) * | 2013-12-18 | 2014-03-12 | 上海市数字证书认证中心有限公司 | Method and system for real-name authentication based on face recognition |
| CN105243357A (en) * | 2015-09-15 | 2016-01-13 | 深圳市环阳通信息技术有限公司 | Identity document-based face recognition method and face recognition device |
| CN106548159A (en) * | 2016-11-08 | 2017-03-29 | 中国科学院自动化研究所 | Reticulate pattern facial image recognition method and device based on full convolutional neural networks |
| CN106778525A (en) * | 2016-11-25 | 2017-05-31 | 北京旷视科技有限公司 | Identity identifying method and device |
| CN107818308A (en) * | 2017-10-31 | 2018-03-20 | 平安科技（深圳）有限公司 | A kind of recognition of face intelligence comparison method, electronic installation and computer-readable recording medium |
| CN107766844A (en) * | 2017-11-13 | 2018-03-06 | 杭州有盾网络科技有限公司 | Method, apparatus, equipment of a kind of reticulate pattern according to recognition of face |
| CN108171032A (en) * | 2017-12-01 | 2018-06-15 | 平安科技（深圳）有限公司 | A kind of identity identifying method, electronic device and computer readable storage medium |
| CN108197557A (en) * | 2017-12-28 | 2018-06-22 | 深圳云天励飞技术有限公司 | Testimony of a witness consistency check method, terminal device and computer readable storage medium |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110647805A (en) * | 2019-08-09 | 2020-01-03 | 平安科技（深圳）有限公司 | Reticulated image identification method and device and terminal equipment |
| CN110647805B (en) * | 2019-08-09 | 2023-10-31 | 平安科技（深圳）有限公司 | Reticulate pattern image recognition method and device and terminal equipment |
| WO2021042544A1 (en) * | 2019-09-03 | 2021-03-11 | 平安科技（深圳）有限公司 | Facial verification method and apparatus based on mesh removal model, and computer device and storage medium |
| TWI774997B (en) * | 2019-10-05 | 2022-08-21 | 中國信託商業銀行股份有限公司 | Risk assessment method and system, service system and computer program product |
| CN111062248A (en) * | 2019-11-08 | 2020-04-24 | 宇龙计算机通信科技(深圳)有限公司 | Image detection method, device, electronic equipment and medium |
| CN111191567A (en) * | 2019-12-26 | 2020-05-22 | 深圳供电局有限公司 | Identity data processing method and device, computer equipment and storage medium |
| CN111339885A (en) * | 2020-02-19 | 2020-06-26 | 平安科技（深圳）有限公司 | User identity determination method based on iris recognition and related device |
| CN111339885B (en) * | 2020-02-19 | 2024-05-28 | 平安科技（深圳）有限公司 | User identity determining method and related device based on iris recognition |
| CN111405365A (en) * | 2020-03-12 | 2020-07-10 | 北京文香信息技术有限公司 | Identity authentication method, device, system and storage medium |
| CN111444831A (en) * | 2020-03-25 | 2020-07-24 | 深圳中科信迅信息技术有限公司 | Method for recognizing human face through living body detection |
| CN111757042A (en) * | 2020-06-28 | 2020-10-09 | 深圳市闪联信息技术有限公司 | Remote collaborative conference method and system based on face authentication |
| CN113487463A (en) * | 2021-05-25 | 2021-10-08 | 中国联合网络通信集团有限公司 | Method and system for managing flowing object |
| CN114241588A (en) * | 2022-02-24 | 2022-03-25 | 北京锐融天下科技股份有限公司 | Self-adaptive face comparison method and system |
| CN114241588B (en) * | 2022-02-24 | 2022-05-20 | 北京锐融天下科技股份有限公司 | Self-adaptive face comparison method and system |
| CN117172783A (en) * | 2023-07-17 | 2023-12-05 | 湖北盈嘉集团有限公司 | Credit and debt cross checking system for confirming account receivables |
| CN117172783B (en) * | 2023-07-17 | 2024-05-07 | 湖北盈嘉集团有限公司 | Credit and debt cross checking system for confirming account receivables |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109871755A (en) | A kind of auth method based on recognition of face | |
| CN109446981B (en) | Face living body detection and identity authentication method and device | |
| US20230116801A1 (en) | Image authenticity detection method and device, computer device, and storage medium | |
| US11288504B2 (en) | Iris liveness detection for mobile devices | |
| CN108470169A (en) | Face identification system and method | |
| CN109034102A (en) | Human face in-vivo detection method, device, equipment and storage medium | |
| CN110335389A (en) | Car door unlocking method and device, system, vehicle, electronic equipment and storage medium | |
| CN108229325A (en) | Method for detecting human face and system, electronic equipment, program and medium | |
| CN104364798A (en) | Systems and method for facial verification | |
| CN106204435A (en) | Image processing method and device | |
| CN105868677A (en) | Live human face detection method and device | |
| WO2005109344A1 (en) | Method for registering iris, device for registering iris and program for registering iris | |
| CN103514440A (en) | Facial recognition | |
| CN105141842A (en) | Tamper-proof license camera system and method | |
| CN112733802B (en) | Image occlusion detection method and device, electronic equipment and storage medium | |
| CN109544560A (en) | Image processing method and device, electronic equipment and storage medium | |
| CN109871845A (en) | Certificate image extracting method and terminal device | |
| CN108154547A (en) | Image generating method and device | |
| CN208351494U (en) | Face identification system | |
| CN109815813A (en) | Image processing method and Related product | |
| CN111985281A (en) | Image generation model generation method and device and image generation method and device | |
| CN114511915B (en) | Trusted certificate photo acquisition system and method based on mobile client | |
| CN104184936B (en) | Image focusing processing method and system based on light field camera | |
| CN110348272A (en) | Method, apparatus, system and the medium of dynamic human face identification | |
| CN116074618B (en) | Intelligent photographing service method, system and storage medium for preventing portrait information leakage |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |