CN106991390A - A kind of hand-held testimony of a witness Compare System and method based on deep learning - Google Patents
A kind of hand-held testimony of a witness Compare System and method based on deep learning Download PDFInfo
- Publication number
- CN106991390A CN106991390A CN201710201258.1A CN201710201258A CN106991390A CN 106991390 A CN106991390 A CN 106991390A CN 201710201258 A CN201710201258 A CN 201710201258A CN 106991390 A CN106991390 A CN 106991390A
- Authority
- CN
- China
- Prior art keywords
- module
- face
- testimony
- information
- identity card
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/164—Detection; Localisation; Normalisation using holistic features
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of hand-held testimony of a witness Compare System based on deep learning and method, the system includes ID card information read module, the image capture module for gathering holder scene photograph, face recognition module, testimony of a witness comparing module and display module.System reads certificate information by ID card information read module, read calls image capture module to capture the face of holder after identity card certificate information, face recognition module is used to carry out image preprocessing to photo, Face datection and face alignment, testimony of a witness comparing module is used to extract face characteristic and compares characteristic value, display module is used for the holder photo for showing collection, and ID card information, operation is reminded and comparison result.Human face detection and recognition module built in the present invention is intended using the algorithm based on deep learning.At present, the face recognition algorithms based on deep learning have been proved to have higher precision than traditional face recognition algorithms in field of face identification.
Description
Technical field
The present invention relates to computer vision, machine learning, image procossing, information security and technical field of biometric identification, tool
Body is related to one kind and is based on deep learning, for verifying the whether consistent hand-held testimony of a witness Compare System of the testimony of a witness and method, existing to improve
The information security degree of generation life.
Background technology
With the development of modern society, requirement of the entire society to information security also more and more higher.Identity card is used as mark
The perfect instrument of holder identity, very important effect is all played in each occasion.
How to confirm personal identification is a social concern urgently to be resolved hurrily.This problem includes two parts：First
Point it is the security of identity card, whether the true and false including certificate here, certificate is expired, and whether the identity belongs to blacklist etc..This
The inspection of a part is the emphasis of current identity card inspection concern.Part II refers to whether accredited personnel is exactly the affiliated people of certificate
The problem of member, i.e. testimony of a witness consistency check.In the various places for needing to carry out checking testimony of a witness uniformity, the mode that majority is taken is also
It is whether safety inspection personnel simply observe the sex of accredited personnel and meet, whether facial obvious characteristic is consistent.Manual identified has
There is larger error, the accuracy rate of identification is not high, and singly recognizes, recognition speed is very slow, the work of staff
Measure also big, unusual labor intensive, inefficiency.
In recent years, developing rapidly for recognition of face makes the testimony of a witness Compare System based on recognition of face be achieved and apply.
The scheme that testimony of a witness consistency check system on the market is taken mostly is：Facial image, identity card are gathered by high-definition camera
Reader extracts identity card certificate photo image, and background server, which is installed, compares software, and audits comparison result.This way lack
Putting is:
(1) the various pieces relative distribution of system, it is not very convenient to dispose, and moves integrally difficulty；
(2) delay has been interacted between background server；
(3) it must network.
The content of the invention
In order to solve the defect and problem of the above, system is compared the invention provides a kind of hand-held testimony of a witness based on deep learning
System and method, read technology with identity card by face recognition technology and are combined that there is provided portable, real-time hand-held testimony of a witness uniformity
Solution is recognized, the degree of identity security can be effectively improved.
The present invention provides a kind of hand-held testimony of a witness Compare System based on deep learning, and system includes ID card information and reads mould
Block, image capture module, face recognition module, testimony of a witness comparing module and display module.Wherein ID card information read module bag
Identity card Card Reader region and identity card reader chip are included, the work that module is completed is reading identity card certificate photo and identity card base
This information, wherein identity card essential information include ID card No., name, sex, the information such as date of birth.Image capture module
Call camera and obtain holder scene photograph.Face recognition module includes image preprocessing, Face datection and face alignment
Three parts.Testimony of a witness comparing module carries out feature extraction and compares the characteristic value of certificate photo and holder scene photograph.Display
The content that module is shown includes the ID card information read, and knot is reminded and compared in the image of the live holder of collection, operation
Really.
The above-mentioned hand-held testimony of a witness Compare System based on deep learning, the ID card information read module, IMAQ mould
Block, face recognition module, the relation between testimony of a witness comparing module and display module is as follows：ID card information read module will be read
To identity card certificate photo pass to face recognition module carry out Face datection and alignment etc. processing, while also by the identity read
Demonstrate,prove all information and pass to display module, holder scene photograph is passed to display module and recognition of face by image capture module respectively
Module is handled, after face recognition module is handled, and face information is sent to testimony of a witness comparing module in the photo extracted
Feature extraction and comparison are carried out, the result of comparison is returned to display module by testimony of a witness comparing module.
Invention further provides a kind of hand-held testimony of a witness comparison method based on deep learning, depth is based on using above-mentioned
The hand-held testimony of a witness Compare System of study, is carried out in accordance with the following steps：
S1 reading identity card certificate photo information
In use, identity card to be positioned over to the identity card Card Reader region of ID card information read module, built-in identity card
Reader chip will read this identity card relevant information, wherein just there is identity card certificate photo information；
S2 reading identity card essential informations
Identity card reader chip read in addition to identity card certificate photo information, also identity card essential information, this
In the identity card essential information that reads include ID card No., name, sex, date of birth etc.；
S3 calls camera
After identity card relevant information is read, using image capture module, camera is called, camera is approximately towards
Accredited human face direction；
S4 obtains holder photo
The photo of camera collection is not necessarily containing holder facial information, and the detecting step in face recognition module is used
Face in detection image, when return detects face, that is, represents to have got the photo of holder, otherwise, continues to adopt
Collection；
S5 image preprocessings
In order to improve the precision of human face detection and recognition, it is pre- that the photo that face recognition module is gathered to camera carries out image
Processing；And the certificate photo photo of identity card identity card is gathered under certain require, so wouldn't be carried out to identity card certificate photo
Pretreatment, input picture herein only has the live holder photo of collection；Image preprocessing in the face recognition module
Can be combined using a certain or a variety of methods, main methods have histogram equalization, medium filtering, gray scale stretching,
Homomorphic filtering, removal illumination and the influence of shade are normalized to the rgb color space of image；Image is located in advance
Reason can remove the irrelevant information in image, be accurately positioned face part, become apparent from image detail, further improve face
The reliability of feature extraction, matching and the identification of image；
S6 Face datections
Face datection is an important step in face recognition module；Face is carried out to the image after pretreatment
Detection, you can orient the face approximate range in image, be that ensuing link is prepared.Input picture herein is identity
The live holder photo of certificate photo and collection is demonstrate,proved, face alignment below, feature extraction and characteristic value, which compare link, is all pair
Both is handled.The method of Face datection can use a variety of methods in the face recognition module, and classification is as follows：
(1) method based on template：Such method is mainly matched with designed template and facial image to be measured, calculates two
Person's similarity, face is differentiated by threshold value；(2) method of feature based：The method of feature based is some spies according to face
Levy or characteristic point, including the various measurements such as geometric properties, space characteristics, gray feature or the relation between them constitute one
Vector, then with input of this vector as face identification system；(3) method learnt based on outward appearance：Face datection is regarded
To distinguish the pattern classification problem of " non-face sample " and " face sample ", produced by the study to face and non-face sample set
Raw grader；
S7 faces align
Face alignment is facial modeling, is automatically positioned out using face recognition module according to the facial image of input
Facial key feature points, such as eyes, nose, corners of the mouth point, eyebrow and each component outline point of face.Face alignment main at present
Technology has following a few classes：(1) ASM and AAM class methods：ASM and AAM are to carry out characteristic point for particular category objects such as faces to carry
Two taken the most classical method, ASM methods more emphasize the matching of characteristic point feature, there is a more preferable positioning precision, and AAM side
Method in terms of speed then advantageously；(2) energy function class method is counted：Facial modeling problem is really all kinds of differences
The problem of complex utilization of type information.In recent years, facial modeling problem is converted into statistics energy function by many scholars
Optimization problem solved, enrich the amalgamation mode of information, also achieve preferable actual effect；(3) regression analysis side
Method：Regression analysis is a kind of classical analysis prediction determination methods, can easily set up from input information to output information it
Between mapping relations；(4) deep learning method：Depth learning technology is realized to complex model based on multilayer neural network framework
Simulate and approach, the ability with prominent announcement data inner link and structure；
S8 feature extractions
The image of input will carry out face characteristic extraction after face alignment using testimony of a witness comparing module；It is so-called
Feature extraction, picked out to the most favourable maximally effective feature of classification, entered simultaneously using various methods exactly from image information
The appropriate dimensionality reduction of row.
S9 characteristic values are compared
Distance metric or measuring similarity can be selected during aspect ratio pair after feature extraction；It is mainly used in calculating individual
The distance of difference has Euclidean distance, Ming Shi distances (popularization of Euclidean distance), mahalanobis distance etc. between body；Distance value is bigger, individual
Difference is bigger；With distance metric on the contrary, measuring similarity is used to calculate the similarity degree between individual, main oriented quantity space cosine
Similarity, Pearson correlation coefficients, Jaccard similarity factors etc., similarity, which more levels off to, 1 shows that individual similarity is higher；
S10 comparison results are shown
Characteristic value comparison result is compared with given threshold, returned on display module the result of comparison, the i.e. testimony of a witness it is consistent or
Person's testimony of a witness is inconsistent.Use of the setting of threshold value to system plays the role of key, especially photo characteristic distance value in people
The inconsistent but closer to the distance situation of the card situation consistent but distant with the testimony of a witness；Under different application scenarios, threshold value is set
Surely it is also required to adjustment.
The present invention provides a kind of hand-held testimony of a witness Compare System and method based on deep learning, and its useful achievement is：This
Invent as integral type structure, modules are integrated in same embedded board, single-hand handling ＆ machine can be supported to operate, user
Just, it is easy to carry, system identification speed is fast, has wide range of applications；In addition the human face detection and recognition module built in the present invention is intended
Using the algorithm based on deep learning, at present in field of face identification, the face recognition algorithms based on deep learning are demonstrate,proved
It is bright that there is higher precision than traditional face recognition algorithms in field of face identification.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, the accompanying drawing in describing below is only this
Some embodiments of invention, for those of ordinary skills, on the premise of not paying creative work, can be with
Other embodiments and its accompanying drawing are obtained according to these accompanying drawing illustrated embodiments.
Fig. 1 is the high-level schematic functional block diagram of present system.
Fig. 2 is the module relationship schematic diagram of present system.
Fig. 3 is the system flow schematic diagram of present system.
Embodiment
The embodiment of the present invention is described in further detail below in conjunction with accompanying drawing.
As shown in figure 1, the hand-held testimony of a witness Compare System of the invention based on deep learning includes 5 modules：ID card information
Read module, image capture module, face recognition module, testimony of a witness comparing module and display module.The ID card information is read
The work that module is completed is reading identity card certificate photo and identity card essential information, and identity card essential information includes name, sex,
The information such as date of birth.Described image acquisition module calls camera and obtains holder scene photograph.The recognition of face mould
Block is pre-processed to identity card certificate photo and holder scene photograph, Face datection and alignment.The testimony of a witness comparing module is entered
Row feature extraction and the characteristic value for comparing certificate photo and holder scene photograph.The content that described display module is shown includes
The identity card relevant information read, the image of the live holder of collection, operation is reminded and comparison result.Modules are integrated
In same embedded board, single-hand handling ＆ machine can be supported to operate.
(1) ID card information read module：ID card information read module read certificate photo as the testimony of a witness compare it is defeated
Enter one of photo, and the identity card essential information read can be used for the simple basic condition for verifying the certificate owner, such as identity card
Number, name, sex, date of birth etc..
(2) image capture module：Image capture module purpose be gather holder scene photograph, read identity card it
After start to call camera to gather holder scene photograph.If the photo of camera collection does not detect face, after
Continuous collection.
(3) face recognition module：Face recognition module includes image preprocessing, three portions of Face datection and face alignment
Point.Wherein Face datection is a key link in Automatic face recognition system.The image of collection may not only have face
Point, so the first step of recognition of face is exactly to do Face datection, the human face region in image is detected.Face alignment is also named
Facial modeling, i.e., according to the facial image of input, be automatically positioned out facial key feature points, such as eyes, nose, the corners of the mouth
Point, eyebrow and each component outline point of face etc..Facial image pretreatment is the shadow to face such as angle in order to remove illumination
Ring.The interference information in image is removed, image detail is become apparent from, the reliability of identification is improved.
(4) testimony of a witness comparing module：Testimony of a witness comparing module includes feature extraction and characteristic value is compared.Here feature extraction refers to
Extract the characteristic value of pretreated facial image.The feature of facial image in pretreated certificate is designated as fisrt feature
Value；The feature of facial image in pretreated holder scene photograph is designated as Second Eigenvalue；Characteristic value, which is compared, refers to body
The information of part card certificate photo and holder facial image is compared, i.e., compare the First Eigenvalue and Second Eigenvalue matching, record
Its characteristic distance metric or similarity；If using distance metric value when comparing, by characteristic distance metric and setting threshold
Value is compared, if distance metric value degree is not higher than comparison threshold value, is judged that identity document is not belonging to holder, is otherwise judged identity
Card belongs to holder.If using similarity when comparing, Similarity value is compared with given threshold, if Similarity value is higher than
Comparison threshold value, judges that identity document belongs to holder, otherwise judges that identity card is not belonging to holder.
(5) display module：Display module is the interface portion of system, and the identity card that the content of display includes reading is related
Information, the image of the live holder of collection, operation is reminded and comparison result.Wherein identity card relevant information is read by reader
Take, including identity card certificate photo and identity card essential information, identity card essential information is available for inspection personnel simply to verify certificate institute
The sex for the person of having, the essential information such as date of birth.The holder photo of display is chosen to be shone for that accredited human face of comparison
Piece.The result of comparison is the whether consistent judged result of the testimony of a witness, and the testimony of a witness is consistent or the testimony of a witness is inconsistent.If reading identity
Camera lens catcher's face, display alarm are can't detect after card always：" holder face please be directed at ".
The above-mentioned hand-held testimony of a witness Compare System based on deep learning includes ID card information read module, IMAQ mould
Relation is as shown in Figure 2 between block, face recognition module, five parts of testimony of a witness comparing module and display module, module：Identity card is believed
The identity card certificate photo read is passed to face recognition module and carries out the processing such as Face datection and alignment by breath read module, simultaneously
Also all information of the identity card read (including identity card certificate photo and identity card essential information) are passed into display module.Image
Holder scene photograph is passed to acquisition module into display module respectively and face recognition module is handled.By recognition of face mould
After block processing, face information is sent to testimony of a witness comparing module and carries out feature extraction and comparison in the photo extracted.The testimony of a witness compares
The result of comparison is returned to display module by module.
The above-mentioned hand-held testimony of a witness Compare System based on deep learning, it is as shown in Figure 3 that it performs flow：
S1 reading identity card certificate photo information
In use, identity card to be positioned over to the identity card Card Reader region of ID card information read module.Built-in identity card
Reader chip will read this identity card relevant information, wherein just there is identity card certificate photo information.Identity card certificate photo information
It is the key message in system use, is also one of the input in recognition of face and testimony of a witness comparing module.
S2 reading identity card essential informations
Identity card reader chip read in addition to identity card certificate photo information, also identity card essential information.This
In the identity card essential information that reads include name, sex, date of birth, home address etc..The reading of identity card essential information
It is easy to inspection personnel simply to verify name, the essential information such as sex.
S3 calls camera
After identity card relevant information is read, image capture module calls camera, and camera is approximately towards accredited
Human face direction.If the photo for opening camera seizure was all not detected among out face, system meeting within one minute
Return to display alarm " holder face please be directed at ".
S4 obtains holder photo
The photo of camera collection is not necessarily containing holder facial information, and the detecting step in face recognition module is used
Face in detection image, when return detects face, that is, represents to have got the photo of holder, otherwise, continues to adopt
Collection.
S5 image preprocessings
In order to improve the precision of human face detection and recognition, it is pre- that the photo that face recognition module is gathered to camera carries out image
Processing.And the certificate photo photo of identity card is gathered under certain require, so wouldn't be pre-processed to identity card certificate photo.
The need for actual conditions, removal illumination is normalized by the rgb color space to image in face recognition module
With the influence of shade.
S6 Face datections
System all uses the method based on deep learning in Face datection, face alignment and characteristic extraction part.In recent years,
Deep learning becomes the focus of research because of its outstanding performance.Compared with other methods, extracted using the method for neutral net
Face characteristic, makes the feature extraction work of complexity simpler, while can learn to some recessive rules in facial image
And rule.Neutral net processing information in a parallel fashion, speed takes smaller, and recognition efficiency is higher.
S7 faces align
The problem of face aligns, relies heavily on the comprehensive fusion of information, and based on the nerve of simulation biological intelligence
The deep learning method of network technology undoubtedly has unique advantage in this aspect.
S8 feature extractions
The image of input will carry out face characteristic extraction after face alignment.Deep learning is known with traditional mode
The maximum difference of other method is that it is the automatic learning characteristic from big data, rather than using the feature of hand-designed.Good spy
The performance of PRS can be greatly improved by levying.Parameter is adjusted by hand due to relying on, and only allows to lack in the design of feature
The parameter of amount.And deep learning can from big data automatic learning characteristic expression, wherein thousands of ginseng can be included
Number.Deep learning can make full use of big data, and for new application, study obtains new effective quickly from training data
Character representation.
S9 characteristic values are compared
Extract after feature, testimony of a witness comparing module uses the characteristic value distance of Euclidean distance measurement.Distance is more than setting
Threshold value shows it is not same people, is otherwise shown to be same people.
S10 comparison results are shown
Display module returns to the result compared, and the testimony of a witness is consistent or the testimony of a witness is inconsistent.
Claims (3)
1. a kind of hand-held testimony of a witness Compare System based on deep learning, it is characterised in that including ID card information read module, figure
As acquisition module, face recognition module, testimony of a witness comparing module and display module；The work that wherein ID card information read module is completed
Work is reading identity card certificate photo and identity card essential information, identity card essential information include ID card No., name, sex,
Date of birth information；Image capture module calls camera and obtains holder scene photograph；Face recognition module includes image
Pretreatment, three parts of Face datection and face alignment；Testimony of a witness comparing module carries out feature extraction and compares certificate photo with holding
The characteristic value of witness's scene photograph；The content that display module is shown includes the ID card information that reads, collection it is live accredited
The image of people, operation is reminded and comparison result.
2. the hand-held testimony of a witness Compare System according to claim 1 based on deep learning, it is characterised in that the identity card
Information reading module, image capture module, face recognition module, the relation between testimony of a witness comparing module and display module is as follows：
The identity card certificate photo read is passed to face recognition module and carried out at Face datection and alignment by ID card information read module
Reason, while all information of the identity card read also are passed into display module, image capture module divides holder scene photograph
Supplementary biography is handled to display module and face recognition module, after face recognition module is handled, people in the photo extracted
Face information is sent to testimony of a witness comparing module and carries out feature extraction and comparison, and the result of comparison is returned to display by testimony of a witness comparing module
Module.
3. a kind of hand-held testimony of a witness comparison method based on deep learning, it is characterised in that using described in claim 1 or 2 based on
The hand-held certification Compare System of deep learning, is carried out in accordance with the following steps：
S1 reading identity card certificate photo information
In use, identity card to be positioned over to the identity card Card Reader region of ID card information read module, built-in identity card is read
Device chip will read this identity card relevant information, wherein just there is identity card certificate photo information；
S2 reading identity card essential informations
The information that identity card reader chip is read is in addition to identity card certificate photo information, also identity card essential information；
Here identity card essential information includes ID card No., name, sex, date of birth information；
S3 calls camera
After ID card information is read, using image capture module, camera is called, camera is towards accredited human face side
To；
S4 obtains holder photo
The photo of camera collection is not necessarily containing holder facial information, and the detecting step in face recognition module is used to examine
Face in altimetric image, when return detects face, that is, represents to have got the photo of holder, otherwise, continues to gather；
S5 image preprocessings
In order to improve the precision of human face detection and recognition, face recognition module needs the photo progress image gathered to camera pre-
Processing：Removal illumination and the influence of shade are normalized to the rgb color space of image；
S6 Face datections
Face datection is an important step in face recognition module；Face inspection is carried out to the image after pretreatment
Survey, you can orient the face approximate range in image, be that ensuing link is prepared；Input picture herein is identity card
Certificate photo and the live holder photo of collection；
S7 faces align
Face alignment is facial modeling, using facial image of the face recognition module according to input, is automatically positioned and appears
Portion's key feature points, including eyes, nose, the corners of the mouth point, eyebrow and each component outline point of face；
S8 feature extractions
The image of input will carry out face characteristic extraction after face alignment using testimony of a witness comparing module；
S9 characteristic values are compared
Extract after feature, the characteristic value distance of Euclidean distance measurement is used using testimony of a witness comparing module, distance is more than setting
Threshold value shows it is not same people, is otherwise shown to be same people；
S10 comparison results are shown
Display module returns to the result compared, and the testimony of a witness is consistent or the testimony of a witness is inconsistent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710201258.1A CN106991390A (en) | 2017-03-30 | 2017-03-30 | A kind of hand-held testimony of a witness Compare System and method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710201258.1A CN106991390A (en) | 2017-03-30 | 2017-03-30 | A kind of hand-held testimony of a witness Compare System and method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106991390A true CN106991390A (en) | 2017-07-28 |
Family
ID=59411916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710201258.1A Pending CN106991390A (en) | 2017-03-30 | 2017-03-30 | A kind of hand-held testimony of a witness Compare System and method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106991390A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107491750A (en) * | 2017-08-14 | 2017-12-19 | 深圳前海华夏智信数据科技有限公司 | A kind of testimony of a witness comparison method and device |
| 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 |
| CN108234591A (en) * | 2017-09-21 | 2018-06-29 | 深圳市商汤科技有限公司 | The content-data of identity-based verification device recommends method, apparatus and storage medium |
| CN108229499A (en) * | 2017-10-30 | 2018-06-29 | 北京市商汤科技开发有限公司 | Certificate recognition methods and device, electronic equipment and storage medium |
| CN108346208A (en) * | 2018-04-19 | 2018-07-31 | 深圳安邦科技有限公司 | A kind of face identification system of deep learning |
| CN108846306A (en) * | 2018-03-28 | 2018-11-20 | 中科博宏（北京）科技有限公司 | A kind of identity card Compare System and method based on deep learning recognition of face |
| CN109087429A (en) * | 2018-09-19 | 2018-12-25 | 重庆第二师范学院 | The method of library ticket testimony of a witness consistency check based on face recognition technology |
| CN109344727A (en) * | 2018-09-07 | 2019-02-15 | 苏州创旅天下信息技术有限公司 | Identity card text information detection method and device, readable storage medium storing program for executing and terminal |
| CN110214320A (en) * | 2017-08-09 | 2019-09-06 | 居米奥公司 | The certification compared using face-image |
| CN110210424A (en) * | 2019-06-05 | 2019-09-06 | 重庆两江新区管理委员会 | A kind of worker's identity information acquisition system and method based on recognition of face |
| CN110569635A (en) * | 2019-09-16 | 2019-12-13 | 山东浪潮商用系统有限公司 | service system login method based on face recognition and service system |
| CN110929711A (en) * | 2019-11-15 | 2020-03-27 | 智慧视通(杭州)科技发展有限公司 | Method for automatically associating identity information and shape information applied to fixed scene |
| CN111191563A (en) * | 2019-12-26 | 2020-05-22 | 三盟科技股份有限公司 | Face recognition method and system based on data sample and test data set training |
| CN113408421A (en) * | 2021-06-21 | 2021-09-17 | 湖北央中巨石信息技术有限公司 | Face recognition method and system based on block chain |
| CN113630460A (en) * | 2021-08-05 | 2021-11-09 | 山东卡尔电气股份有限公司 | Face recognition method and system for testimony comparison |
| CN113887277A (en) * | 2021-08-23 | 2022-01-04 | 福建数博讯信息科技有限公司 | Handheld ID card reader and information acquisition and sign-in method based on same |
| CN114511915A (en) * | 2022-04-19 | 2022-05-17 | 南昌大学 | Credible certificate photo acquisition system and method based on mobile client |
| CN114741671A (en) * | 2022-04-22 | 2022-07-12 | 广东泓胜科技股份有限公司 | Identity comparison method and system based on biological recognition and related equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902959A (en) * | 2012-04-28 | 2013-01-30 | 王浩 | Face recognition method and system for storing identification photo based on second-generation identity card |
| CN104765739A (en) * | 2014-01-06 | 2015-07-08 | 南京宜开数据分析技术有限公司 | Large-scale face database searching method based on shape space |
-
2017
- 2017-03-30 CN CN201710201258.1A patent/CN106991390A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902959A (en) * | 2012-04-28 | 2013-01-30 | 王浩 | Face recognition method and system for storing identification photo based on second-generation identity card |
| CN104765739A (en) * | 2014-01-06 | 2015-07-08 | 南京宜开数据分析技术有限公司 | Large-scale face database searching method based on shape space |
Non-Patent Citations (1)
| Title |
|---|
| 孙军: "基于计算机视觉的人证同一性研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110214320A (en) * | 2017-08-09 | 2019-09-06 | 居米奥公司 | The certification compared using face-image |
| CN110214320B (en) * | 2017-08-09 | 2022-07-15 | 居米奥公司 | Authentication using facial image comparison |
| CN107491750A (en) * | 2017-08-14 | 2017-12-19 | 深圳前海华夏智信数据科技有限公司 | A kind of testimony of a witness comparison method and device |
| CN108234591A (en) * | 2017-09-21 | 2018-06-29 | 深圳市商汤科技有限公司 | The content-data of identity-based verification device recommends method, apparatus and storage medium |
| CN108234591B (en) * | 2017-09-21 | 2021-01-05 | 深圳市商汤科技有限公司 | Content data recommendation method and device based on identity authentication device and storage medium |
| CN108229499A (en) * | 2017-10-30 | 2018-06-29 | 北京市商汤科技开发有限公司 | Certificate recognition methods and device, electronic equipment and 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 |
| CN108846306A (en) * | 2018-03-28 | 2018-11-20 | 中科博宏（北京）科技有限公司 | A kind of identity card Compare System and method based on deep learning recognition of face |
| CN108346208A (en) * | 2018-04-19 | 2018-07-31 | 深圳安邦科技有限公司 | A kind of face identification system of deep learning |
| CN109344727A (en) * | 2018-09-07 | 2019-02-15 | 苏州创旅天下信息技术有限公司 | Identity card text information detection method and device, readable storage medium storing program for executing and terminal |
| CN109087429A (en) * | 2018-09-19 | 2018-12-25 | 重庆第二师范学院 | The method of library ticket testimony of a witness consistency check based on face recognition technology |
| CN109087429B (en) * | 2018-09-19 | 2020-12-04 | 重庆第二师范学院 | Method for checking consistency of library book-borrowing testimony of witness based on face recognition technology |
| CN110210424A (en) * | 2019-06-05 | 2019-09-06 | 重庆两江新区管理委员会 | A kind of worker's identity information acquisition system and method based on recognition of face |
| CN110569635A (en) * | 2019-09-16 | 2019-12-13 | 山东浪潮商用系统有限公司 | service system login method based on face recognition and service system |
| CN110929711A (en) * | 2019-11-15 | 2020-03-27 | 智慧视通(杭州)科技发展有限公司 | Method for automatically associating identity information and shape information applied to fixed scene |
| CN111191563A (en) * | 2019-12-26 | 2020-05-22 | 三盟科技股份有限公司 | Face recognition method and system based on data sample and test data set training |
| CN113408421A (en) * | 2021-06-21 | 2021-09-17 | 湖北央中巨石信息技术有限公司 | Face recognition method and system based on block chain |
| CN113630460A (en) * | 2021-08-05 | 2021-11-09 | 山东卡尔电气股份有限公司 | Face recognition method and system for testimony comparison |
| CN113887277A (en) * | 2021-08-23 | 2022-01-04 | 福建数博讯信息科技有限公司 | Handheld ID card reader and information acquisition and sign-in method based on same |
| CN113887277B (en) * | 2021-08-23 | 2024-01-02 | 福建数博讯信息科技有限公司 | Handheld identity card reader and information acquisition and sign-in method based on same |
| CN114511915A (en) * | 2022-04-19 | 2022-05-17 | 南昌大学 | Credible certificate photo acquisition system and method based on mobile client |
| CN114511915B (en) * | 2022-04-19 | 2022-07-29 | 南昌大学 | Trusted certificate photo acquisition system and method based on mobile client |
| CN114741671A (en) * | 2022-04-22 | 2022-07-12 | 广东泓胜科技股份有限公司 | Identity comparison method and system based on biological recognition and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106991390A (en) | A kind of hand-held testimony of a witness Compare System and method based on deep learning | |
| CN102682309B (en) | Face feature registering method and device based on template learning | |
| CN103426016B (en) | Method and device for authenticating second-generation identity card | |
| CN108846306A (en) | A kind of identity card Compare System and method based on deep learning recognition of face | |
| CN106203294A (en) | The testimony of a witness unification auth method analyzed based on face character | |
| CN103530659B (en) | With reference to the face identification method and attendance checking system of original and symmetrical face facial image | |
| CN109670430A (en) | A kind of face vivo identification method of the multiple Classifiers Combination based on deep learning | |
| CN108563999A (en) | A kind of piece identity's recognition methods and device towards low quality video image | |
| CN108898125A (en) | One kind being based on embedded human face identification and management system | |
| CN106570491A (en) | Robot intelligent interaction method and intelligent robot | |
| CN108171223A (en) | A kind of face identification method and system based on multi-model multichannel | |
| CN106650574A (en) | Face identification method based on PCANet | |
| CN105335719A (en) | Living body detection method and device | |
| CN103902962A (en) | Shielding or light source self-adaption human face recognition method and device | |
| WO2022213396A1 (en) | Cat face recognition apparatus and method, computer device, and storage medium | |
| CN106650560A (en) | Identity authentication method based on identity card and identity authentication processor and system thereof | |
| CN109255319A (en) | For the recognition of face payment information method for anti-counterfeit of still photo | |
| CN107911643A (en) | Show the method and apparatus of scene special effect in a kind of video communication | |
| CN108846269A (en) | One kind is towards manifold identity identifying method and identification authentication system | |
| CN111862413A (en) | Method and system for realizing epidemic situation resistant non-contact multidimensional identity rapid identification | |
| CN109711309A (en) | A kind of method whether automatic identification portrait picture closes one's eyes | |
| CN110705454A (en) | Face recognition method with living body detection function | |
| CN107944395A (en) | A kind of method and system based on neutral net verification testimony of a witness unification | |
| Hernandez-Ortega et al. | FaceQvec: Vector quality assessment for face biometrics based on ISO compliance | |
| CN106156739B (en) | A kind of certificate photo ear detection and extracting method based on face mask analysis |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170728 |