CN108229499A - Certificate recognition methods and device, electronic equipment and storage medium - Google Patents
Certificate recognition methods and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN108229499A CN108229499A CN201711052744.8A CN201711052744A CN108229499A CN 108229499 A CN108229499 A CN 108229499A CN 201711052744 A CN201711052744 A CN 201711052744A CN 108229499 A CN108229499 A CN 108229499A
- Authority
- CN
- China
- Prior art keywords
- certificate
- image
- detected
- information
- certificate image
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/513—Sparse representations
Abstract
The embodiment of the invention discloses a kind of certificate recognition methods and device, electronic equipment and storage medium, wherein method includes：Obtain certificate image to be detected；Certificate image to be detected is handled, obtains image to be detected region for including the certificate to be detected；Feature extraction is carried out to image to be detected region according to deep neural network, obtains comprehensive characteristics；Certificate image is carried out according to the comprehensive characteristics and forges identification, obtains the recognition result of the certificate image.The embodiment of the present invention comprehensive and accurate can identify forged certificate, without carrying out the addition of additional information or equipment, implement easier.
Description
Technical field
The present invention relates to image processing techniques, especially a kind of certificate recognition methods and device, electronic equipment and storage are situated between
Matter.
Background technology
Certificate false proof is the major issue of field of anti-counterfeit technology, and certificate false proof has important application in many fields, especially
It is in internet financial field, certificate false proof technology being capable of effective protection account holder's account safety.
Invention content
The embodiment of the present invention provides a kind of technical solution of certificate identification.
A kind of certificate recognition methods provided in an embodiment of the present invention, including：Obtain certificate image to be detected；To be detected
Certificate image handled, obtain image to be detected region for including the certificate to be detected；According to deep neural network
Feature extraction is carried out to image to be detected region, obtains comprehensive characteristics；It is pseudo- that certificate image is carried out according to the comprehensive characteristics
Identification is made, obtains the recognition result of the certificate image.
In a kind of optional mode, further include：The deep neural network is trained；It is described neural to the depth
Network is trained, including：Acquire each certificate image sample, the certificate image sample be labeled with whether spurious information；According to
The deep neural network carries out feature extraction to the certificate image sample, obtains the comprehensive special of the certificate image sample
Sign；Certificate is carried out according to the comprehensive characteristics and forges identification, obtains the recognition result of the certificate image sample；Based on the card
The recognition result of part image pattern and mark whether spurious information, the deep neural network is trained.
In a kind of optional mode, when the certificate image sample is certificate image forge sample, the certificate image
Sample is forged also to be labeled with forging source markup information；It is described that the deep neural network is trained, it further includes：According to institute
It states comprehensive characteristics and obtains the forgery source predictive information that certificate image forges sample；Based on the forgery source predictive information and institute
It states and forges source markup information, the deep neural network is trained.
It is described that the deep neural network is trained in a kind of optional mode, it further includes：When collecting certificate figure
During as forging newly-increased sample, newly-increased sample being forged based on the certificate image, the deep neural network is trained, it is described
The newly-increased sample of certificate image forgery increases forgery technology newly by certificate image and obtains.
In a kind of optional mode, the comprehensive characteristics include：Local binary patterns feature, the block diagram of sparse coding are special
In sign, color characteristic, panorama sketch feature, certificate provincial characteristics, certificate minutia at least one of or it is multinomial.
In a kind of optional mode, the local binary patterns feature is used to characterize the marginal information in image to be detected；
The block diagram feature of the sparse coding is used to characterize the reflective and fuzzy message in image to be detected；The color characteristic is used for
Distinguish the certificate really shot and certificate copy；The panorama sketch feature is used to extract the spurious information of image to be detected；Institute
Certificate region is stated for determining that figure is cut in certificate region in image to be detected；The certificate minutia is used to extract image to be detected
Spurious information.
In a kind of optional mode, the recognition result of the certificate image includes：The certificate image is no to forge；Work as institute
When stating recognition result as certificate image forgery, the method further includes：According to the comprehensive characteristics, the certificate figure is exported
The forgery source of picture.
In a kind of optional mode, the method further includes：According to the forgery recognition result of certificate image and the forgery
Source, the training deep neural network.
In a kind of optional mode, the certificate image for certificate image in itself or the video frame in certificate video.
It is described that certificate image to be detected is handled in a kind of optional mode, image to be detected is obtained, including：
Denoising, enhancing, smooth and Edge contrast are carried out to certificate image, obtains image to be detected region；Alternatively, to certificate video
Carrying out video selects frame to operate, and obtains target video frame image, and denoising, enhancing, smoothly are carried out to the target video frame image
And Edge contrast, obtain image to be detected region.
It is described frame to be selected to operate certificate video progress video in a kind of optional mode, target video frame image is obtained, is wrapped
It includes：For the video frame of the certificate video, any one in following operation or arbitrary combination are carried out：Judge certificate whether position
Whether completely included in video frame images, video frame figure shared by document surface in the center of video frame images, certificate edge
As whether area ratio meets preset ratio, whether video frame images clarity meets image in default clarity threshold, video
Whether exposure meets default exposure threshold value；According to judging result, the target video frame is chosen in the certificate video
Image.
In a kind of optional mode, for certificate video to be detected, further include：If for the institute in the certificate video
The recognition result for having video frame images is that there is no forge, it is determined that there is no forge in the certificate video；If it alternatively, deposits
It is to exist to forge in the recognition result of any one video frame images, it is determined that exist in the certificate video and forge.
In a kind of optional mode, it is described obtain the recognition result of the certificate image after, further include：If it is not present
Certificate image is forged, and carries out certificate information extraction, and the certificate information of extraction is compared with information in certificate library；In response to
The certificate information of the extraction and the information matches in the certificate library, then asked by user；Or, in response to the extraction
Certificate information is mismatched with the information in the certificate library, then is not asked by user, and the certificate image is labeled as puppet
Certificate image is made, the deep neural network is trained.
A kind of certificate identification device provided in an embodiment of the present invention, including：Acquiring unit, for obtaining certificate to be detected
Image；Processing unit for handling certificate image to be detected, obtains including the to be detected of the certificate to be detected
Image-region；Extraction unit for carrying out feature extraction to image to be detected region according to deep neural network, obtains comprehensive
Close feature；Recognition unit forges identification for carrying out certificate image according to the comprehensive characteristics, obtains the knowledge of the certificate image
Other result.
In a kind of optional mode, further include：Training unit, it is described for being trained to the deep neural network
Training unit includes：Subelement is acquired, for acquiring each certificate image sample, the certificate image sample is labeled with whether forging
Information；Subelement is extracted, for carrying out feature extraction to the certificate image sample according to the deep neural network, obtains institute
State the comprehensive characteristics of certificate image sample；It identifies subelement, forges identification for carrying out certificate according to the comprehensive characteristics, obtain
The recognition result of the certificate image sample；Training performs subelement, for the recognition result based on the certificate image sample
With mark whether spurious information, the deep neural network is trained.
In a kind of optional mode, when the certificate image sample is certificate image forge sample, the certificate image
Sample is forged also to be labeled with forging source markup information；The training unit further includes：It forges source and obtains subelement, for root
The forgery source predictive information of certificate image forgery sample is obtained according to the comprehensive characteristics；Training subelement is forged, for being based on
The forgery source predictive information and the forgery source markup information, are trained the deep neural network.
In a kind of optional mode, the training unit further includes：Newly-increased sample training subelement collects card for working as
When part image forge increases sample newly, newly-increased sample is forged based on the certificate image, the deep neural network is trained,
The newly-increased sample of certificate image forgery increases forgery technology newly by certificate image and obtains.
In a kind of optional mode, the comprehensive characteristics include：Local binary patterns feature, the block diagram of sparse coding are special
In sign, color characteristic, panorama sketch feature, certificate provincial characteristics, certificate minutia at least one of or it is multinomial.
In a kind of optional mode, the local binary patterns feature is used to characterize the marginal information in image to be detected；
The block diagram feature of the sparse coding is used to characterize the reflective and fuzzy message in image to be detected；The color characteristic is used for
Distinguish the certificate really shot and certificate copy；The panorama sketch feature is used to extract the spurious information of image to be detected；Institute
Certificate region is stated for determining that figure is cut in certificate region in image to be detected；The certificate minutia is used to extract image to be detected
Spurious information.
In a kind of optional mode, the recognition result of the certificate image includes：The certificate image is no to forge；Work as institute
When stating recognition result as certificate image forgery, described device further includes：Source output unit is forged, for according to described comprehensive
Feature is closed, exports the forgery source of the certificate image.
In a kind of optional mode, described device further includes：As a result training unit, for being known according to the forgery of certificate image
Other result and the forgery source, the training deep neural network.
In a kind of optional mode, the certificate image for certificate image in itself or the video frame in certificate video.
In a kind of optional mode, the processing unit includes：Certificate image handle subelement, for certificate image into
Row denoising, enhancing, smooth and Edge contrast, obtain image to be detected region；Alternatively, certificate video handles subelement, use
Frame is selected to operate in carrying out video to certificate video, obtain target video frame image, and the target video frame image is gone
It makes an uproar, enhance, smooth and Edge contrast, obtaining image to be detected region.
In a kind of optional mode, the certificate video processing subelement is specifically used for：For regarding for the certificate video
Frequency frame carries out any one in following operation or arbitrary combination：Judge whether certificate is located at center, the certificate of video frame images
Whether edge is completely included in video frame images, whether video frame images area ratio shared by document surface meets default ratio
Whether example, video frame images clarity meet whether image exposure degree in default clarity threshold, video meets default exposure
Threshold value；According to judging result, the target video frame image is chosen in the certificate video.
In a kind of optional mode, for certificate video to be detected, further include：Determination subelement is forged, if for needle
Recognition result to all video frame images in the certificate video is that there is no forge, it is determined that in the certificate video
There is no forgeries；Alternatively, if the recognition result there are any one video frame images is to exist to forge, it is determined that the certificate regards
Exist in frequency and forge.
In a kind of optional mode, it is described obtain the recognition result of the certificate image after, further include：It compares single
Member if to be forged for there is no certificate images, carries out certificate information extraction, and by information in the certificate information of extraction and certificate library
It is compared；Response unit for the information matches in the certificate information in response to the extraction and the certificate library, then passes through
User asks；Or, mismatched in response to the certificate information of the extraction with the information in the certificate library, then it please not by user
It asks, and the certificate image is labeled as forged certificate image, the deep neural network is trained.
A kind of computer readable storage medium provided in an embodiment of the present invention, is stored thereon with computer program, the program
The step of certificate recognition methods is realized when being executed by processor.
A kind of electronic equipment provided in an embodiment of the present invention, including memory, processor and storage on a memory and can
The computer program run on a processor, the processor realize the step of the certificate recognition methods when performing described program
Suddenly.
Certificate recognition methods and device, the electronic equipment and storage medium provided based on the above embodiment of the present invention, relatively
Identification is forged in the prior art certificate need to be carried out by additional information or special hardware equipment, by certificate image to be detected
Feature extraction is carried out, obtains the characteristic information of multiple classifications, it is pseudo- so as to carry out certificate according to the comprehensive characteristics information of multiple classifications
Identification is made, comprehensive and accurate can identify forged certificate, without carrying out the addition of additional information or equipment, is implemented simpler
Just.It in an optional mode, is modeled by the powerful descriptive power of deep neural network so as to forging characteristic information
Resolution it is more accurate, can preferably distinguish true identity information and forged identity information；And pass through forged certificate source
It distinguishes, service port can be supplied to use using spurious information source as additional forgery class label.
Below by drawings and examples, technical scheme of the present invention is described in further detail.
Description of the drawings
The attached drawing of a part for constitution instruction describes the embodiment of the present invention, and is used to explain together with description
The principle of the present invention.
With reference to attached drawing, according to following detailed description, the present invention can be more clearly understood, wherein：
Fig. 1 is the flow chart of certificate recognition methods one embodiment of the present invention.
Fig. 2 is the flow chart of another embodiment of certificate recognition methods of the present invention
Fig. 3 is the flow chart of training deep neural network in the embodiment of the present invention.
Fig. 4 is the structure diagram of certificate identification device one embodiment of the present invention.
Fig. 5 is the structure diagram of another embodiment of certificate identification device of the present invention.
Fig. 6 is the structure diagram of electronic equipment one embodiment of the present invention.
Specific embodiment
Carry out the various exemplary embodiments of detailed description of the present invention now with reference to attached drawing.It should be noted that：Unless in addition have
Body illustrates that the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
The range of invention.
Simultaneously, it should be appreciated that for ease of description, the size of the various pieces shown in attached drawing is not according to reality
Proportionate relationship draw.
It is illustrative to the description only actually of at least one exemplary embodiment below, is never used as to the present invention
And its application or any restrictions that use.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as part of specification.
It should be noted that：Similar label and letter represents similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need to that it is further discussed.
There is important application in many fields of the certificate false proof technology in life, particularly in internet financial field, card
Part anti-counterfeiting technology being capable of effective protection account holder's account safety.However the modeling for carrying out certificate spurious information is very complicated.One
Aspect, the acquisition of information of forgery is than relatively limited, and such as in internet arena, spurious information comes solely from picture or video file, and
It can not be distinguished according to the additional information of holder.On the other hand, the difference in source is forged according to certificate, feature also not phase
Together, the complexity of feature means the difficulty bigger of modeling.In the implementation of the present invention, inventor has found, existing card
Part method for anti-counterfeit is needed by many additional informations or by special hardware device, with the development of internet finance, there is an urgent need for
It grasps more convenient, efficient certificate false proof and makes identifying schemes.
Fig. 1 is the flow chart of certificate recognition methods one embodiment of the present invention.As shown in Figure 1, the embodiment method includes
S101-S104。
S101：Obtain certificate image to be detected.
In the business such as internet finance, need to carry out user's real-name authentication.User shoots upper subpoena by mobile terminal
Part picture or video, to further determine that the identity of user.For example, in user carries out stock account-opening, user is needed
Upload identity card picture.At this point, the user is by carrying out picture upload after the terminal takings such as mobile phone.
S102：Certificate image to be detected is handled, obtains image to be detected region for including certificate to be detected.
It after getting certificate image to be detected, is handled, identification requirement is forged so as to obtain meeting certificate
Image to be detected region of certificate to be detected.
In the embodiment of the present invention, certificate image can be video of the certificate image in (certificate picture) or certificate video in itself
Frame.
For certificate picture to be detected, the image procossings such as denoising, enhancing, smooth and sharpening are carried out to certificate image, are obtained
To image to be detected region.Image procossing is in image analysis, before carrying out feature extraction, segmentation and matching to input picture
The processing carried out, the main purpose of image procossing are to eliminate information unrelated in image, restore useful real information, enhance
Detectability for information about and the data that simplify to the maximum extent, so as to improve feature extraction, image segmentation, matching and identification
Reliability.
For the certificate video of certificate to be detected, video is carried out to certificate video first, frame is selected to operate, obtain target video
Frame image；Then the image procossings such as denoising, enhancing, smooth and sharpening are carried out to target video frame image, obtain image to be detected
Region.Similarly, image procossing is in image analysis, to what is carried out before input picture progress feature extraction, segmentation and matching
Processing, the main purpose of image procossing are to eliminate information unrelated in image, restore useful real information, enhancing is for information about
Detectability and to the maximum extent simplify data, so as to improve feature extraction, image segmentation, matching and identification reliability.
Wherein, select video the standard that frame operates that can include several following：Whether certificate is located at picture centre, certificate side
Whether edge is completely included in image, image area ratio height, image definition and image exposure degree shared by document surface.
Video is carried out by above-mentioned standard to select frame, the image of satisfactory quality can be selected, for example, the image selected can be：
Certificate is located at picture centre, certificate edge is completely included into image, certificate accounts for image scaled about 1/2~3/4, image clearly
Degree is higher, exposure is higher.
Therefore, the detailed process that video selects frame to operate, obtain target video frame image being carried out to certificate video can include
Following steps：(1) for the video frame of certificate video, any one in following operation or arbitrary combination are carried out：Judge certificate
Whether the center of video frame images is located at, whether certificate edge includes regarding in video frame images, shared by document surface by complete
Whether frequency frame image area ratio meets preset ratio, whether video frame images clarity meets default clarity threshold, video
Whether middle image exposure degree meets default exposure threshold value；(2) according to judging result, the mesh is chosen in the certificate video
Mark video frame images.
For the certificate video of certificate to be detected：It is if equal for the recognition result of all video frame images in certificate video
It is forged to be not present, it is determined that there is no forge in certificate video；If the alternatively, identification knot there are any one video frame images
Fruit is to exist to forge, it is determined that exists in certificate video and forges.
S103：Feature extraction is carried out to image to be detected region according to deep neural network, obtains comprehensive characteristics.
In the embodiment of the present invention, comprehensive characteristics can be understood as including multiple class another characteristics, wherein each category Properties
For characterizing some characteristic in image to be detected region.For example, comprehensive characteristics include but not limited to：Local binary patterns are special
Sign, the block diagram feature of sparse coding, color characteristic, panorama sketch feature, certificate provincial characteristics, certificate minutia.Feature is believed
It is more to cease classification, can more carry out forgery identification comprehensively.Wherein：Local binary patterns feature is used to characterize the side in image to be detected
Edge information；The block diagram feature of sparse coding is used to characterize the reflective and fuzzy message in image to be detected；Color characteristic is used for
Distinguish the certificate really shot and certificate copy；Panorama sketch feature is used to extract the spurious information of image to be detected；Certificate area
Domain is used to determine that figure is cut in certificate region in image to be detected；Certificate minutia is used to extract the spurious information of image to be detected.
S104：Certificate image is carried out according to comprehensive characteristics and forges identification, obtains the recognition result of certificate image.
In the embodiment of the present invention, forgery includes two kinds of situations：Certificate is forged；Image forge.Certificate forgery refers to by right
Certificate itself feature recognition in image, judgement obtain whether certificate is forged in itself；Image forge refers to image is identified,
Judge image whether be by other images by some processing (such as：Copy, PS, reproduction) after obtained image, judge to scheme
It seem no forgery.
In a kind of optional mode, the recognition result of the certificate image includes：Certificate image is no to forge；When identification is tied
When fruit is certificate image forge, the method further includes：According to comprehensive characteristics, the forgery source of certificate image is exported.It is wherein pseudo-
Source is made to include but not limited to：Copy, PS (photoshop is edited), reproduction picture, etc..When recognition result is certificate image
During forgery, this method may also include：According to the forgery recognition result of certificate image and source is forged, training deep neural network
(concrete mode of training deep neural network can be found in hereafter step S2011-S2014).
As it can be seen that identification is forged relative in the prior art certificate need to be carried out by additional information or special hardware equipment, this
In inventive embodiments scheme, by certificate image to be detected carry out feature extraction, obtain the characteristic information of multiple classifications, so as to
According to multiple classifications characteristic information carry out certificate forge identification, comprehensive and accurate can identify forged certificate, without into
The addition of row additional information or equipment is implemented easier.Moreover, the powerful descriptive power by deep neural network can be effective
Identify the various forged certificates for forging source.
Fig. 2 is the flow chart of another embodiment of certificate recognition methods of the present invention.On the basis of Fig. 1 embodiments, Fig. 2 is implemented
Example, which is described, forges feature extraction and identification model using deep neural network as certificate, and spy is performed for certificate image to be detected
Sign extraction and certificate forge the operation of identification.The realization process of the embodiment can simply understand as follows：It predefines and forges spy
Information aggregate (training pattern) is levied, by carrying out feature extraction, and judge whether included in image to be detected to image to be detected
The characteristic information in characteristic information set is forged, determines whether certificate is forged certificate.
Deep learning forms more abstract high-rise expression attribute classification or feature by combining low-level feature, to find number
According to distributed nature represent.Neural network possesses input layer, output layer and a hidden layer, and the feature vector of input passes through hidden
The transformation containing layer reaches output layer, and classification results are obtained in output layer.Deep neural network (Deep Neural Networks,
DNN it) can be understood as the neural network with deep learning ability.For example, the modeling process of deep neural network is, it is first
First, prepare data, it would be desirable to which the sample data of classification is stored in the form of each column in matrix, by the category label of each sample
A row vector is stored as by data order；Secondly, network configuration, parameter initialization and conversion are carried out；Then, loss letter is write
Number；Finally, sample data is trained.
As shown in Fig. 2, the embodiment method includes S201-S204.
S201：Certificate is obtained by training deep neural network and forges feature extraction and identification model.
It is the flow chart of training deep neural network in the embodiment of the present invention referring to Fig. 3 in an optional mode,
Deep neural network is trained by following steps S2011-S2014.
S2011：Acquire each certificate image sample, certificate image sample be labeled with whether spurious information.
For example, the certificate image data set (database) manually marked can be pre-established, by the certificate figure of all kinds of certificates
As being stored in the data set.In modeling process, image pattern is obtained from the data set.
Wherein, spurious information refers to that each category feature by image reflects that image is likely to the feature forged, also referred to as
Forge clue.Such as spurious information includes but not limited to：It is reflective and fuzzy message in the image reflected by HSC features, logical
It crosses RGB feature and can significantly distinguish the certificate really shot and certificate copy information, image can be extracted by LARGE
In most apparent spurious information (hack), the edges of reflective reproduction device screen moire fringes can be extracted based on SMALL features
Etc. spurious informations, image PS (photoshop is edited), reproduction screen moire fringes, reflective etc. can be extracted by TINY features
Spurious information.
S2012：Feature extraction is carried out to certificate image pattern according to deep neural network, obtains the comprehensive of certificate image sample
Close feature.
Comprehensive characteristics include and are not limited to following arbitrary multinomial：Local binary patterns (Local Binary Pattern,
LBP) feature, block diagram (Histogram of Sparse Code, HSC) feature of sparse coding, color (RedGreen
Blue, RGB) feature, panorama sketch (LARGE) feature, certificate region (SMALL) feature, certificate details (TINY) feature.
Wherein, by LBP features, the marginal information in image to be detected can be protruded；It, can be brighter by HSC features
Reflective and fuzzy message in aobvious reflection image to be detected；By RGB feature, the card really shot can be more obviously distinguished
Part and certificate copy information；LARGE features are full figure features, based on LARGE features, can be extracted in image to be detected most
Apparent spurious information (hack)；Certificate region (SMALL) is certificate region several times size in image to be detected (such as 1.5
Times size) region cut figure, comprising the part that certificate, certificate and background suit, based on SMALL features, can extract it is reflective,
The spurious informations such as the edge of reproduction device screen moire fringes；Certificate detail view (TINY) is that figure is cut in the region of evidence obtaining part size, is wrapped
Containing certificate, based on TINY features, the forgeries such as image PS (photoshop is edited), reproduction screen moire fringes, reflective can be extracted
Information.
S2013：Certificate is carried out according to comprehensive characteristics and forges identification, obtains the recognition result of certificate image sample.
S2014：Recognition result based on certificate image sample and mark whether spurious information, to deep neural network into
Row training.
The spurious information of the forged certificate included in above-mentioned various features can be learnt by training by deep neural network
It arrives, can be detected after the image input deep neural network comprising these spurious informations any later, it is possible to judge
It is otherwise real docu-ment image for forged certificate image.
In an optional mode, when certificate image sample is certificate image forge sample, certificate image forges sample
It is also labeled with forging source markup information；Deep neural network is trained, is further included：Certificate figure is obtained according to comprehensive characteristics
Forgery source predictive information as forging sample；Based on source predictive information and the forgery source markup information is forged, to depth
Degree neural network is trained.
In an optional mode, deep neural network is trained, is further included：It is forged newly when collecting certificate image
When increasing sample, newly-increased sample is forged based on certificate image, deep neural network is trained, certificate image forges newly-increased sample
Forgery technology is increased newly by certificate image to obtain.
S202：Feature extraction and identification model are forged according to certificate, feature extraction is carried out to certificate image to be detected.
In an optional mode, feature extraction is carried out to certificate image to be detected by following steps：According to " certificate
The classification of the determining forgery characteristic information of forgery feature extraction and identification model ", correspondence are extracted from certificate image to be detected
Comprehensive characteristics information of all categories.
For example, it is assumed that the forgery characteristic information that certificate forgery feature extraction and identification model determine includes：LBP、HSC、
Six class of RGB, LARGE, SMALL, TINY, then, correspondence extracts the six category features information from certificate image to be detected.
S203：The forgery that certificate image is carried out according to comprehensive characteristics identifies.
In an optional mode, realize that carrying out certificate according to the characteristic information of multiple classifications forges knowledge by following steps
Not：By the comprehensive characteristics of multiple classifications, it is input to certificate and forges feature extraction and identification model, detect certificate image to be detected
In whether include one or more forgery characteristic informations.
Still such as above-mentioned example, for extracting LBP, HSC, RGB, LARGE, SMALL, TINY six in certificate image to be detected
This six category features information is input to certificate and forged in feature extraction and identification model by category feature information, by it is saved this
Six classes forge the certificate forgery feature extraction of characteristic information and classification is identified to features above information in identification model, if institute
Extraction feature matches with one or more in the forgery characteristic information that is preserved in model, it is determined that in certificate image to be detected
Include one or more forgery characteristic informations.
S204：Obtain the recognition result of certificate image.
Still such as above-mentioned example, if it is determined that comprising one or more forgery features in certificate image to be detected, then export
Certificate recognition result be：Similar prompt messages such as " certificate image are forged certificate image ".In addition, due in modeling process
In, the source of all kinds of forged certificate images is marked, therefore, when the situation that certificate recognition result is forged certificate image
Under, forged certificate image sources can be exported simultaneously, for example, output forged certificate image is copy or reproduction etc..
It is above-mentioned when it is forged certificate image that recognition result, which prompts certificate image to be detected, in an optional mode
Method may also include the steps of：According to comprehensive characteristics, the forgery source of certificate image is exported.It further includes on this basis：Root
According to the forgery recognition result of certificate image and forgery source, training deep neural network.As a result, by this certificate figure to be detected
As sample data, being updated to the model of deep neural network.In the prior art, the side that certificate false proof makes detection is carried out
Method is once it is determined that be just not easy to have updated afterwards, it is difficult to cope with emerging forgery situation.The present invention is based on deep neural networks to come
Certificate false proof detection is carried out, it, can be in time by emerging spurious information to depth nerve when there is new forgery sample
Network is trained, it is easy to the training of formula is finely adjusted in rear end, timely update deep neural network, makes deep neural network
Learn in time to various emerging spurious informations, reach the standard grade in time, so as to be carried out to emerging forgery situation rapidly
Anti-counterfeiting detection.
In an optional mode, after the recognition result of the certificate image is obtained, further include：If there is no certificates
Image forge carries out certificate information extraction, and the certificate information of extraction is compared with information in certificate library；In response to extraction
Certificate information and the certificate library in information matches, then asked by user；Or, the certificate information in response to extraction and institute
The information stated in certificate library mismatches, then is not asked by user, and certificate image is labeled as forged certificate image, to depth
Neural network is trained.
Certificate recognition methods through the embodiment of the present invention when business platform needs certificate information, carries out certificate first
Anti-counterfeiting detects, and when video to be detected or picture are detected by certificate false proof, certificate is extracted from video to be detected or picture
Number, identity information of the image as registered user in certificate photograph.Idiographic flow is described as follows：Business platform can receive
To certificate video or image to be detected when login or checking request, is obtained, after selecting frame operation and pretreatment operation, by
Trained deep neural network extracts multi-class characteristic information, detects in image to be detected with the presence or absence of spurious information, treats
Detection image carries out certificate false proof and makes detection, and feeds back certificate false proof testing result to business platform.If testing result is shown not
There are spurious informations, then carry out certificate identification and information extraction, verification information correctness are compared with information in library, if information
Correctly, then logging request passes through, and the service request sent to the user that request logs in is handled.If testing result display exists
Spurious information, then logging request does not pass through, and can selectively return to spurious information source, for further distinguishing.
As it can be seen that in the embodiments of the present invention, modeled by the powerful descriptive power of deep neural network so that
It is more accurate to the resolution for forging characteristic information, it can preferably distinguish true identity information and forged identity information；And pass through
The differentiation in forged certificate source can be supplied to service port to make using spurious information source as additional forgery class label
With.
Fig. 4 is the structure diagram of certificate identification device one embodiment of the present invention.The device of the embodiment can be used for real
The existing above-mentioned each method embodiment of the present invention.As shown in figure 4, the device of the embodiment includes:
Acquiring unit 401, for obtaining certificate image to be detected；
Processing unit 402 for handling certificate image to be detected, obtains including the certificate to be detected
Image to be detected region；
Extraction unit 403 for carrying out feature extraction to image to be detected region according to deep neural network, obtains
Comprehensive characteristics；
Recognition unit 404 forges identification for carrying out certificate image according to the comprehensive characteristics, obtains the certificate image
Recognition result.
In an optional mode, the comprehensive characteristics information includes：Local binary patterns feature, the column of sparse coding
In figure feature, color characteristic, panorama sketch feature, certificate provincial characteristics, certificate minutia at least one of or it is multinomial.
In an optional mode, the local binary patterns feature is used to characterize the marginal information in image to be detected；
The block diagram feature of the sparse coding is used to characterize the reflective and fuzzy message in image to be detected；The color characteristic is used for
Distinguish the certificate really shot and certificate copy；The panorama sketch feature is used to extract the spurious information of image to be detected；Institute
Certificate region is stated for determining that figure is cut in certificate region in image to be detected；The certificate minutia is used to extract image to be detected
Spurious information.
Fig. 5 is the structure diagram of another embodiment of certificate identification device of the present invention.The device of the embodiment can be used for
Realize the above-mentioned each method embodiment of the present invention.As shown in figure 5, the device of the embodiment includes:
Acquiring unit 501, for obtaining certificate image to be detected；
Processing unit 502 for handling certificate image to be detected, obtains including the certificate to be detected
Image to be detected region；
Extraction unit 503 for carrying out feature extraction to image to be detected region according to deep neural network, obtains
Comprehensive characteristics；
Recognition unit 504 forges identification for carrying out certificate image according to the comprehensive characteristics, obtains the certificate image
Recognition result.
In an optional mode, described device further includes：Training unit 505, for the deep neural network into
Row training, the training unit 505 include：
Subelement 5051 is acquired, for acquiring each certificate image sample, the certificate image sample is labeled with whether forging
Information；
Subelement 5052 is extracted, is carried for carrying out feature to the certificate image sample according to the deep neural network
It takes, obtains the comprehensive characteristics of the certificate image sample；
It identifies subelement 5053, forges identification for carrying out certificate according to the comprehensive characteristics, obtain the certificate image
The recognition result of sample；
Training perform subelement 5054, for the recognition result based on the certificate image sample and mark whether forgery
Information is trained the deep neural network.
In an optional mode, when the certificate image sample is certificate image forge sample, the certificate image
Sample is forged also to be labeled with forging source markup information；The training unit 505 further includes：
It forges source and obtains subelement 5055, for obtaining the forgery of certificate image forgery sample according to the comprehensive characteristics
Source predictive information；
Training subelement 5056 is forged, for being based on the forgery source predictive information and the forgery source mark letter
Breath, is trained the deep neural network.
In an optional mode, the training unit 505 further includes：
Newly-increased sample training subelement 5057, for when collecting certificate image and forging newly-increased sample, based on the card
Part image forge increases sample newly and the deep neural network is trained, and the certificate image forges newly-increased sample and passes through certificate
Image increases forgery technology newly and obtains.
In an optional mode, the comprehensive characteristics include：Local binary patterns feature, the block diagram of sparse coding are special
In sign, color characteristic, panorama sketch feature, certificate provincial characteristics, certificate minutia at least one of or it is multinomial.
In an optional mode, the local binary patterns feature is used to characterize the marginal information in image to be detected；
The block diagram feature of the sparse coding is used to characterize the reflective and fuzzy message in image to be detected；The color characteristic is used for
Distinguish the certificate really shot and certificate copy；The panorama sketch feature is used to extract the spurious information of image to be detected；Institute
Certificate region is stated for determining that figure is cut in certificate region in image to be detected；The certificate minutia is used to extract image to be detected
Spurious information.
In an optional mode, the recognition result of the certificate image includes：The certificate image is no to forge；Work as institute
When stating recognition result as certificate image forgery, described device further includes：
Source output unit 506 is forged, for according to the comprehensive characteristics, exporting the forgery source of the certificate image.
In an optional mode, described device further includes：
As a result training unit 507, for the forgery recognition result according to certificate image and the forgery source, described in training
Deep neural network.
In an optional mode, the certificate image for certificate image in itself or the video frame in certificate video.
In an optional mode, the processing unit 502 includes：
Certificate image handles subelement 5021, for carrying out denoising, enhancing, smooth and Edge contrast to certificate image, obtains
To image to be detected region；Alternatively,
Certificate video handles subelement 5022, frame is selected to operate for carrying out video to certificate video, obtains target video frame
Image, and denoising, enhancing, smooth and Edge contrast are carried out to the target video frame image, obtain image to be detected area
Domain.
In an optional mode, the certificate video processing subelement 5022 is specifically used for：
For the video frame of the certificate video, any one in following operation or arbitrary combination are carried out：Judge certificate
Whether the center of video frame images is located at, whether certificate edge includes regarding in video frame images, shared by document surface by complete
Whether frequency frame image area ratio meets preset ratio, whether video frame images clarity meets default clarity threshold, video
Whether middle image exposure degree meets default exposure threshold value；
According to judging result, the target video frame image is chosen in the certificate video.
In an optional mode, for certificate video to be detected, described device further includes：
Determination subelement 508 is forged, if for the recognition result for all video frame images in the certificate video
It is that there is no forge, it is determined that there is no forge in the certificate video；Alternatively, if there are any one video frame images
Recognition result is to exist to forge, it is determined that exists in the certificate video and forges.
In an optional mode, which is characterized in that it is described obtain the recognition result of the certificate image after, it is described
Device further includes：
Comparing unit 509 if to be forged for there is no certificate images, carries out certificate information extraction, and by the certificate of extraction
Information is compared with information in certificate library；
Response unit 510 for the information matches in the certificate information in response to the extraction and the certificate library, then leads to
Cross user's request；Or, mismatched in response to the certificate information of the extraction with the information in the certificate library, then do not pass through user
Request, and the certificate image is labeled as forged certificate image, the deep neural network is trained.
The embodiment of the present invention additionally provides a kind of electronic equipment, such as can be mobile terminal, personal computer (PC), put down
Plate computer, server etc..Below with reference to Fig. 6, it illustrates suitable for being used for realizing the terminal device of the embodiment of the present application or service
The structure diagram of the electronic equipment 600 of device：As shown in fig. 6, computer system 600 includes one or more processors, communication
Portion etc., one or more of processors are for example：One or more central processing unit (CPU) 601 and/or one or more
Image processor (GPU) 613 etc., processor can according to the executable instruction being stored in read-only memory (ROM) 602 or
From the executable instruction that storage section 608 is loaded into random access storage device (RAM) 603 perform various appropriate actions and
Processing.Communication unit 612 may include but be not limited to network interface card, and the network interface card may include but be not limited to IB (Infiniband) network interface card,
Processor can communicate with read-only memory 602 and/or random access storage device 630 to perform executable instruction,
It is connected by bus 604 with communication unit 612 and is communicated through communication unit 612 with other target devices, is implemented so as to complete the application
The corresponding operation of any one method that example provides, for example, obtaining certificate image to be detected；Certificate image to be detected is carried out
Processing, obtains image to be detected region for including the certificate to be detected；According to deep neural network to the mapping to be checked
As region progress feature extraction, comprehensive characteristics are obtained；Certificate image is carried out according to the comprehensive characteristics and forges identification, is obtained described
The recognition result of certificate image.
In addition, in RAM 603, it can also be stored with various programs and data needed for device operation.CPU601、ROM602
And RAM603 is connected with each other by bus 604.In the case where there is RAM603, ROM602 is optional module.RAM603 is stored
Executable instruction is written in executable instruction into ROM602 at runtime, and it is above-mentioned logical that executable instruction performs processor 601
The corresponding operation of letter method.Input/output (I/O) interface 605 is also connected to bus 604.Communication unit 612 can be integrally disposed,
It may be set to be with multiple submodule (such as multiple IB network interface cards), and in bus link.
I/O interfaces 605 are connected to lower component：Importation 606 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loud speaker etc.；Storage section 608 including hard disk etc.；
And the communications portion 609 of the network interface card including LAN card, modem etc..Communications portion 609 via such as because
The network of spy's net performs communication process.Driver 610 is also according to needing to be connected to I/O interfaces 605.Detachable media 611, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on driver 610, as needed in order to be read from thereon
Computer program be mounted into storage section 608 as needed.
Need what is illustrated, framework as shown in Figure 6 is only a kind of optional realization method, can root during concrete practice
The component count amount and type of above-mentioned Fig. 6 are selected, are deleted, increased or replaced according to actual needs；It is set in different function component
Put, can also be used it is separately positioned or integrally disposed and other implementations, such as GPU and CPU separate setting or can be by GPU collection
Into on CPU, communication unit separates setting, can also be integrally disposed on CPU or GPU, etc..These interchangeable embodiments
Each fall within protection domain disclosed by the invention.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product, it is machine readable including being tangibly embodied in
Computer program on medium, computer program are included for the program code of the method shown in execution flow chart, program code
It may include the corresponding instruction of corresponding execution method and step provided by the embodiments of the present application, for example, receiving the picture of certificate to be detected
Or video；After picture or video to certificate to be detected are handled, certificate image to be detected is obtained；To certificate image to be detected
Feature extraction is carried out, obtains the characteristic information of multiple classifications；Certificate is carried out according to the characteristic information of the multiple classification and forges knowledge
Not, the recognition result of the certificate is obtained.In such embodiments, the computer program can by communications portion 609 from
It is downloaded and installed on network and/or is mounted from detachable media 611.In the computer program by central processing unit
(CPU) during 601 execution, the above-mentioned function of being limited in the present processes is performed.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above method embodiment can pass through
The relevant hardware of program instruction is completed, and aforementioned program can be stored in a computer read/write memory medium, the program
When being executed, step including the steps of the foregoing method embodiments is performed；And aforementioned storage medium includes：ROM, RAM, magnetic disc or light
The various media that can store program code such as disk.
Methods and apparatus of the present invention may be achieved in many ways.For example, can by software, hardware, firmware or
Software, hardware, firmware any combinations realize methods and apparatus of the present invention.The said sequence of the step of for the method
Merely to illustrate, the step of method of the invention, is not limited to sequence described in detail above, special unless otherwise
It does not mentionlet alone bright.In addition, in some embodiments, the present invention can be also embodied as recording program in the recording medium, these programs
Including being used to implement machine readable instructions according to the method for the present invention.Thus, the present invention also covering stores to perform basis
The recording medium of the program of the method for the present invention.Description of the invention provides for the sake of example and description, and not
It is exhaustively or limits the invention to disclosed form.Many modifications and variations are for those of ordinary skill in the art
For be obvious.Selection and description embodiment are to more preferably illustrate the principle of the present invention and practical application, and make ability
The those of ordinary skill in domain is it will be appreciated that of the invention so as to design the various embodiments with various modifications suitable for special-purpose.
Claims (10)
1. a kind of certificate recognition methods, which is characterized in that including：
Obtain certificate image to be detected；
Certificate image to be detected is handled, obtains image to be detected region for including the certificate to be detected；
Feature extraction is carried out to image to be detected region according to deep neural network, obtains comprehensive characteristics；
Certificate image is carried out according to the comprehensive characteristics and forges identification, obtains the recognition result of the certificate image.
2. it according to the method described in claim 1, it is characterized in that, further includes：
The deep neural network is trained；
It is described that the deep neural network is trained, including：
Acquire each certificate image sample, the certificate image sample be labeled with whether spurious information；
Feature extraction is carried out to the certificate image sample according to the deep neural network, obtains the certificate image sample
Comprehensive characteristics；
Certificate is carried out according to the comprehensive characteristics and forges identification, obtains the recognition result of the certificate image sample；
Recognition result based on the certificate image sample and mark whether spurious information, the deep neural network is carried out
Training.
3. according to the method described in claim 2, it is characterized in that, when the certificate image sample is certificate image forge sample
When, the certificate image forges sample and is also labeled with forging source markup information；It is described that the deep neural network is instructed
Practice, further include：
The forgery source predictive information of certificate image forgery sample is obtained according to the comprehensive characteristics；
Based on the forgery source predictive information and the forgery source markup information, the deep neural network is instructed
Practice.
4. according to the method described in claim 2, it is characterized in that, described be trained the deep neural network, also wrap
It includes：
When collecting certificate image and forging newly-increased sample, newly-increased sample is forged to depth nerve based on the certificate image
Network is trained, and the newly-increased sample of certificate image forgery increases forgery technology newly by certificate image and obtains.
5. according to claim 2-4 any one of them methods, which is characterized in that the comprehensive characteristics include：Local binary mould
In formula feature, the block diagram feature of sparse coding, color characteristic, panorama sketch feature, certificate provincial characteristics, certificate minutia
At least one is multinomial.
6. the method according to right wants 5, which is characterized in that the local binary patterns feature is used to characterize image to be detected
In marginal information；The block diagram feature of the sparse coding is used to characterize the reflective and fuzzy message in image to be detected；Institute
Color characteristic is stated for distinguishing the certificate really shot and certificate copy；The panorama sketch feature is used to extract image to be detected
Spurious information；The certificate region is used to determine that figure is cut in certificate region in image to be detected；The certificate minutia is used for
Extract the spurious information of image to be detected.
7. according to claim 1-4 any one of them methods, which is characterized in that the recognition result of the certificate image includes：
The certificate image is no to forge；When the recognition result is forged for the certificate image, the method further includes：
According to the comprehensive characteristics, the forgery source of the certificate image is exported.
8. a kind of certificate identification device, which is characterized in that including：
Acquiring unit, for obtaining certificate image to be detected；
Processing unit for handling certificate image to be detected, obtains including the to be detected of the certificate to be detected
Image-region；
Extraction unit for carrying out feature extraction to image to be detected region according to deep neural network, obtains comprehensive spy
Sign；
Recognition unit forges identification for carrying out certificate image according to the comprehensive characteristics, obtains the identification of the certificate image
As a result.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The step of any one of claim 1-7 the methods are realized during row.
10. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes the step of any one of claim 1-7 the methods when performing described program
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711052744.8A CN108229499A (en) | 2017-10-30 | 2017-10-30 | Certificate recognition methods and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711052744.8A CN108229499A (en) | 2017-10-30 | 2017-10-30 | Certificate recognition methods and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108229499A true CN108229499A (en) | 2018-06-29 |
Family
ID=62655722
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711052744.8A Pending CN108229499A (en) | 2017-10-30 | 2017-10-30 | Certificate recognition methods and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108229499A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109255299A (en) * | 2018-08-13 | 2019-01-22 | 北京市商汤科技开发有限公司 | Identity identifying method and device, electronic equipment and storage medium |
| CN109359502A (en) * | 2018-08-13 | 2019-02-19 | 北京市商汤科技开发有限公司 | False-proof detection method and device, electronic equipment, storage medium |
| CN110070130A (en) * | 2019-04-23 | 2019-07-30 | 杭州智趣智能信息技术有限公司 | A kind of certificate information recognition methods, transposition, equipment and medium |
| CN110309859A (en) * | 2019-06-06 | 2019-10-08 | 阿里巴巴集团控股有限公司 | A kind of image true-false detection method, device and electronic equipment |
| CN110443237A (en) * | 2019-08-06 | 2019-11-12 | 北京旷视科技有限公司 | Certificate recognition methods, device, electronic equipment and computer readable storage medium |
| CN110533643A (en) * | 2019-08-21 | 2019-12-03 | 阿里巴巴集团控股有限公司 | Certificate identification method and device |
| CN110717450A (en) * | 2019-10-09 | 2020-01-21 | 深圳大学 | Training method and detection method for automatically identifying copied image of original document |
| WO2020034733A1 (en) * | 2018-08-13 | 2020-02-20 | 北京市商汤科技开发有限公司 | Identity authentication method and apparatus, electronic device, and storage medium |
| CN111127398A (en) * | 2019-11-25 | 2020-05-08 | 深圳前海微众银行股份有限公司 | Method and device for detecting certificate photo duplication, terminal equipment and storage medium |
| CN111259894A (en) * | 2020-01-20 | 2020-06-09 | 普信恒业科技发展（北京）有限公司 | Certificate information identification method and device and computer equipment |
| CN111275042A (en) * | 2020-01-21 | 2020-06-12 | 支付宝实验室(新加坡)有限公司 | Counterfeit certificate identification method and device and electronic equipment |
| CN111291586A (en) * | 2018-12-06 | 2020-06-16 | 北京市商汤科技开发有限公司 | Living body detection method, living body detection device, electronic apparatus, and computer-readable storage medium |
| CN111324874A (en) * | 2020-01-21 | 2020-06-23 | 支付宝实验室(新加坡)有限公司 | Certificate authenticity identification method and device |
| CN111353434A (en) * | 2020-02-28 | 2020-06-30 | 北京市商汤科技开发有限公司 | Information identification method, device, system, electronic equipment and storage medium |
| US10769263B1 (en) | 2019-05-07 | 2020-09-08 | Alibaba Group Holding Limited | Certificate verification |
| CN111652223A (en) * | 2019-08-21 | 2020-09-11 | 阿里巴巴集团控股有限公司 | Certificate identification method and device |
| CN111767828A (en) * | 2020-06-28 | 2020-10-13 | 京东数字科技控股有限公司 | Certificate image copying and identifying method and device, electronic equipment and storage medium |
| WO2020224296A1 (en) * | 2019-05-07 | 2020-11-12 | 创新先进技术有限公司 | Certificate verification and identity verification methods and apparatuses, and devices |
| CN112528998A (en) * | 2021-02-18 | 2021-03-19 | 成都新希望金融信息有限公司 | Certificate image processing method and device, electronic equipment and readable storage medium |
| WO2022089124A1 (en) * | 2020-10-29 | 2022-05-05 | 腾讯科技（深圳）有限公司 | Certificate authenticity identification method and apparatus, computer-readable medium, and electronic device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156853A (en) * | 2010-02-11 | 2011-08-17 | 深圳鼎识科技有限公司 | Certificate reading system and certificate reading method |
| CN104079587A (en) * | 2014-07-21 | 2014-10-01 | 深圳天祥质量技术服务有限公司 | Certificate identification device and certificate check system |
| CN105118048A (en) * | 2015-07-17 | 2015-12-02 | 北京旷视科技有限公司 | Method and device for identifying copying certificate image |
| US20170078281A1 (en) * | 2015-09-15 | 2017-03-16 | Solid Year Co., Ltd. | User Login Method and System Capable of Analyzing User Face Validity |
| CN106778607A (en) * | 2016-12-15 | 2017-05-31 | 国政通科技股份有限公司 | A kind of people based on recognition of face and identity card homogeneity authentication device and method |
| CN106991390A (en) * | 2017-03-30 | 2017-07-28 | 电子科技大学 | A kind of hand-held testimony of a witness Compare System and method based on deep learning |
| CN106991451A (en) * | 2017-04-14 | 2017-07-28 | 武汉神目信息技术有限公司 | A kind of identifying system and method for certificate picture |
-
2017
- 2017-10-30 CN CN201711052744.8A patent/CN108229499A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156853A (en) * | 2010-02-11 | 2011-08-17 | 深圳鼎识科技有限公司 | Certificate reading system and certificate reading method |
| CN104079587A (en) * | 2014-07-21 | 2014-10-01 | 深圳天祥质量技术服务有限公司 | Certificate identification device and certificate check system |
| CN105118048A (en) * | 2015-07-17 | 2015-12-02 | 北京旷视科技有限公司 | Method and device for identifying copying certificate image |
| US20170078281A1 (en) * | 2015-09-15 | 2017-03-16 | Solid Year Co., Ltd. | User Login Method and System Capable of Analyzing User Face Validity |
| CN106778607A (en) * | 2016-12-15 | 2017-05-31 | 国政通科技股份有限公司 | A kind of people based on recognition of face and identity card homogeneity authentication device and method |
| CN106991390A (en) * | 2017-03-30 | 2017-07-28 | 电子科技大学 | A kind of hand-held testimony of a witness Compare System and method based on deep learning |
| CN106991451A (en) * | 2017-04-14 | 2017-07-28 | 武汉神目信息技术有限公司 | A kind of identifying system and method for certificate picture |
Non-Patent Citations (2)
| Title |
|---|
| 张宜浩、文俊浩: "《基于半监督学习的个性化推荐算法研究》", 31 May 2016, 重庆大学出版社 * |
| 葛晓滨 等: "《移动电子商务教程》", 30 September 2014, 中国科学技术大学出版社 * |
Cited By (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359502A (en) * | 2018-08-13 | 2019-02-19 | 北京市商汤科技开发有限公司 | False-proof detection method and device, electronic equipment, storage medium |
| JP7165746B2 (en) | 2018-08-13 | 2022-11-04 | ベイジン・センスタイム・テクノロジー・デベロップメント・カンパニー・リミテッド | ID authentication method and device, electronic device and storage medium |
| KR102406432B1 (en) * | 2018-08-13 | 2022-06-08 | 베이징 센스타임 테크놀로지 디벨롭먼트 컴퍼니 리미티드 | Identity authentication methods and devices, electronic devices and storage media |
| JP2021516819A (en) * | 2018-08-13 | 2021-07-08 | ベイジン センスタイム テクノロジー デベロップメント カンパニー， リミテッド | ID authentication method and devices, electronic devices and storage media |
| CN109255299A (en) * | 2018-08-13 | 2019-01-22 | 北京市商汤科技开发有限公司 | Identity identifying method and device, electronic equipment and storage medium |
| WO2020034733A1 (en) * | 2018-08-13 | 2020-02-20 | 北京市商汤科技开发有限公司 | Identity authentication method and apparatus, electronic device, and storage medium |
| KR20200118842A (en) * | 2018-08-13 | 2020-10-16 | 베이징 센스타임 테크놀로지 디벨롭먼트 컴퍼니 리미티드 | Identity authentication method and device, electronic device and storage medium |
| CN111291586A (en) * | 2018-12-06 | 2020-06-16 | 北京市商汤科技开发有限公司 | Living body detection method, living body detection device, electronic apparatus, and computer-readable storage medium |
| CN110070130A (en) * | 2019-04-23 | 2019-07-30 | 杭州智趣智能信息技术有限公司 | A kind of certificate information recognition methods, transposition, equipment and medium |
| WO2020224296A1 (en) * | 2019-05-07 | 2020-11-12 | 创新先进技术有限公司 | Certificate verification and identity verification methods and apparatuses, and devices |
| US10769263B1 (en) | 2019-05-07 | 2020-09-08 | Alibaba Group Holding Limited | Certificate verification |
| US11086977B2 (en) | 2019-05-07 | 2021-08-10 | Advanced New Technologies Co., Ltd. | Certificate verification |
| US11392679B2 (en) | 2019-05-07 | 2022-07-19 | Advanced New Technologies Co., Ltd. | Certificate verification |
| US10909227B2 (en) | 2019-05-07 | 2021-02-02 | Advanced New Technologies Co., Ltd. | Certificate verification |
| CN110309859B (en) * | 2019-06-06 | 2023-04-07 | 创新先进技术有限公司 | Image authenticity detection method and device and electronic equipment |
| CN110309859A (en) * | 2019-06-06 | 2019-10-08 | 阿里巴巴集团控股有限公司 | A kind of image true-false detection method, device and electronic equipment |
| CN110443237A (en) * | 2019-08-06 | 2019-11-12 | 北京旷视科技有限公司 | Certificate recognition methods, device, electronic equipment and computer readable storage medium |
| CN110443237B (en) * | 2019-08-06 | 2023-06-30 | 北京旷视科技有限公司 | Certificate identification method, device, electronic equipment and computer readable storage medium |
| CN111652223A (en) * | 2019-08-21 | 2020-09-11 | 阿里巴巴集团控股有限公司 | Certificate identification method and device |
| CN111652223B (en) * | 2019-08-21 | 2023-04-18 | 创新先进技术有限公司 | Certificate identification method and device |
| CN110533643A (en) * | 2019-08-21 | 2019-12-03 | 阿里巴巴集团控股有限公司 | Certificate identification method and device |
| CN110533643B (en) * | 2019-08-21 | 2023-06-30 | 创新先进技术有限公司 | Certificate authentication method and device |
| CN110717450A (en) * | 2019-10-09 | 2020-01-21 | 深圳大学 | Training method and detection method for automatically identifying copied image of original document |
| CN111127398A (en) * | 2019-11-25 | 2020-05-08 | 深圳前海微众银行股份有限公司 | Method and device for detecting certificate photo duplication, terminal equipment and storage medium |
| CN111127398B (en) * | 2019-11-25 | 2023-09-26 | 深圳前海微众银行股份有限公司 | Certificate photo copying detection method and device, terminal equipment and storage medium |
| CN111259894B (en) * | 2020-01-20 | 2023-07-07 | 普信恒业科技发展（北京）有限公司 | Certificate information identification method and device and computer equipment |
| CN111259894A (en) * | 2020-01-20 | 2020-06-09 | 普信恒业科技发展（北京）有限公司 | Certificate information identification method and device and computer equipment |
| CN111324874A (en) * | 2020-01-21 | 2020-06-23 | 支付宝实验室(新加坡)有限公司 | Certificate authenticity identification method and device |
| CN111275042B (en) * | 2020-01-21 | 2023-07-18 | 支付宝实验室(新加坡)有限公司 | Identification method and device for forged certificate and electronic equipment |
| CN111275042A (en) * | 2020-01-21 | 2020-06-12 | 支付宝实验室(新加坡)有限公司 | Counterfeit certificate identification method and device and electronic equipment |
| WO2021169384A1 (en) * | 2020-02-28 | 2021-09-02 | 北京市商汤科技开发有限公司 | Information identification method and apparatus, and system, electronic device, storage medium and computer program |
| CN111353434A (en) * | 2020-02-28 | 2020-06-30 | 北京市商汤科技开发有限公司 | Information identification method, device, system, electronic equipment and storage medium |
| TWI766458B (en) * | 2020-02-28 | 2022-06-01 | 大陸商北京市商湯科技開發有限公司 | Information identification method and apparatus, electronic device, and storage medium |
| JP2022524672A (en) * | 2020-02-28 | 2022-05-10 | 北京市商▲湯▼科技▲開▼▲發▼有限公司 | Information recognition methods and devices, systems, electronic devices, storage media and computer programs |
| CN111767828A (en) * | 2020-06-28 | 2020-10-13 | 京东数字科技控股有限公司 | Certificate image copying and identifying method and device, electronic equipment and storage medium |
| CN111767828B (en) * | 2020-06-28 | 2023-12-08 | 京东科技控股股份有限公司 | Certificate image reproduction identification method and device, electronic equipment and storage medium |
| WO2022089124A1 (en) * | 2020-10-29 | 2022-05-05 | 腾讯科技（深圳）有限公司 | Certificate authenticity identification method and apparatus, computer-readable medium, and electronic device |
| CN112528998A (en) * | 2021-02-18 | 2021-03-19 | 成都新希望金融信息有限公司 | Certificate image processing method and device, electronic equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108229499A (en) | Certificate recognition methods and device, electronic equipment and storage medium | |
| CN108229330A (en) | Face fusion recognition methods and device, electronic equipment and storage medium | |
| CN109255299A (en) | Identity identifying method and device, electronic equipment and storage medium | |
| CN108921061A (en) | A kind of expression recognition method, device and equipment | |
| CN108229341A (en) | Sorting technique and device, electronic equipment, computer storage media, program | |
| CN108427874A (en) | Identity identifying method, server and computer readable storage medium | |
| CN108229485A (en) | For testing the method and apparatus of user interface | |
| CN108256591A (en) | For the method and apparatus of output information | |
| CN112907595B (en) | Surface defect detection method and device | |
| CN115186303B (en) | Financial signature safety management method and system based on big data cloud platform | |
| CN107944398A (en) | Based on depth characteristic association list diagram image set face identification method, device and medium | |
| CN109784415A (en) | The method and device of image-recognizing method and device, training convolutional neural networks | |
| CN108509994A (en) | character image clustering method and device | |
| CN109598671A (en) | Image generating method, device, equipment and medium | |
| CN113642621A (en) | Zero sample image classification method based on generation countermeasure network | |
| CN108766464A (en) | Digital audio based on mains frequency fluctuation super vector distorts automatic testing method | |
| CN113657896A (en) | Block chain transaction topological graph analysis method and device based on graph neural network | |
| CN109614982A (en) | Product analysis method, apparatus, computer equipment and storage medium | |
| Bai et al. | Fake generated painting detection via frequency analysis | |
| Mata-Montero et al. | A texture and curvature bimodal leaf recognition model for identification of costa rican plant species | |
| CN108268889A (en) | To carved gravure true-false detection method, detection platform and detecting system | |
| CN110197140A (en) | Material checking method and equipment based on Text region | |
| CN111462388A (en) | Bill inspection method and device, terminal equipment and storage medium | |
| Hauri | Detecting signatures in scanned document images | |
| Yu et al. | A multi-scale feature selection method for steganalytic feature GFR |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180629 |