CN103699887B - Portrait identification method and device - Google Patents
Portrait identification method and device Download PDFInfo
- Publication number
- CN103699887B CN103699887B CN201310714175.4A CN201310714175A CN103699887B CN 103699887 B CN103699887 B CN 103699887B CN 201310714175 A CN201310714175 A CN 201310714175A CN 103699887 B CN103699887 B CN 103699887B
- Authority
- CN
- China
- Prior art keywords
- score value
- identification
- face
- sequence
- value sequence
- 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.)
- Active
Links
Abstract
The invention discloses a portrait identification method and device. The portrait identification method comprises the following steps: acquiring a face video image of an identity information provider, and acquiring a face image on a valid identity certificate and a matched face image in a third-party identity data system; normalizing the face video image, the face image on the valid identity certificate and the matched face image in the third-party identity data system; acquiring face feature data from the normalized face video image, the face image on the valid identity certificate and the matched face image in the third-party identity data system for serving as first, second and third face feature data; crossly matching the first, second and third face feature data pairwise; outputting the cross matching results of the first, second and third face feature data. By adopting the scheme, the accuracy of portrait identification can be increased, and online transaction becomes safer and more reliable.
Description
Technical field
The present invention relates to image identification technical field, more particularly to a kind of portrait identification method and device.
Background technology
In order to ensure the safe and reliable of online trading, user needs to provide effective documentation of identity to supply web site operator
Audited, only audit the user passing through and just can carry out online trading.In practice, user is first by the proof of identification of scanning
Files passe to Website server, after the scanned copy obtaining the documentation of identity that user uploads, will wrap by web site operator
Include the identity information data in interior identity information and Ministry of Public Security's identity information data system such as portrait, identification card number, address
Carry out contrast identification.Identification of Images therein is exactly an important ring therein.
Existing Identification of Images is usually portrait and Ministry of Public Security's identity information on the effective proof of identification providing user
Portrait in data system carries out contrast identification.But, above-mentioned way can only identify what user uploaded to a certain extent
Whether user identity documentary evidence is consistent with the identity information data of the Ministry of Public Security.Authentication is carried out for false impersonation's identity
Behavior but cannot be carried out identifying, leaves larger hidden danger, has had a strong impact on the safety of online trading.
Content of the invention
The problem that the embodiment of the present invention solves is how more accurately to carry out Identification of Images, so that online trading is more
Safe and reliable.
For solving the above problems, embodiments provide a kind of portrait identification method, described portrait identification method bag
Include:
Captured identity information provides the facial video image of people, and obtains effective identity card that identity information provides people to provide
The facial image matching in facial image on bright and third party's identity data system；
Described in normalized, facial video image, described identity information provide the people on effective proof of identification of people's offer
The facial image matching in face image and described third party's identity data system；
From providing people the effective identity card providing through the facial video image of described normalized, described identity information
Face characteristic data is obtained in the facial image matching in facial image and described third party's identity data system on bright,
Respectively as first, second, and third face characteristic data；
Described first, second, and third face characteristic data is carried out cross-matched two-by-two；
Export the result of described first, second, and third face characteristic data cross coupling.
Alternatively, described face characteristic data includes face gray feature data, face shape characteristic and people's face
Skin textural characteristics data.
Alternatively, described described first, second, and third face characteristic data carried out cross-matched two-by-two include:
To the face gray feature data in described first, second, and third face characteristic data successively two-by-two than
Right, obtain gray feature identification score value sequence；
To the face shape characteristic in described first, second, and third face characteristic data successively two-by-two than
Right, obtain shape facility identification score value sequence；
Face dermatoglyph characteristic in described first, second, and third face characteristic data is carried out two-by-two successively
Compare, obtain textural characteristics identification score value sequence；
To described gray feature identification score value sequence, described shape recognition score value sequence and described texture recognition score value sequence
In identical precedence numerical value execute respectively multiple features fusion identification, obtain fusion recognition score value sequence.
Alternatively, described described first, second, and third face characteristic data is carried out cross-matched two-by-two, also include:
Described gray feature is identified identical bits in score value sequence, described shape recognition score value sequence, described texture recognition score value sequence
The secondary numerical value numerical value execution multiple Classifiers Combination with identical precedence in described fusion recognition score value sequence respectively, obtains comprehensive knowledge
Other score value sequence.
Alternatively, described portrait identification method also includes: the result of described cross-matched is compared with default threshold value
Relatively, when the result of described cross-matched is less than described threshold value, issue warning information.
Alternatively, the described result by described cross-matched and default threshold value are compared, when described cross-matched
Result be less than described threshold value when, issue warning information, comprising: by described fusion recognition score value sequence respectively with default first threshold
In value sequence, the numerical value of identical precedence is compared, when described fusion recognition score value sequence is less than phase in described first threshold sequence
With precedence numerical value when, issue warning information.
Alternatively, the described result by described cross-matched and default threshold value are compared, when described cross-matched
When result is less than described threshold value, issue warning information, comprising: by described comprehensive identification score value sequence respectively with default second threshold
In value sequence, the numerical value of identical precedence is compared, when described comprehensive identification score value sequence is less than phase in described Second Threshold sequence
With precedence numerical value when, issue warning information.
The embodiment of the present invention additionally provides a kind of Identification of Images device, and described Identification of Images device includes:
Collection acquiring unit, provides the facial video image of people for captured identity information, and obtains identity information offer
The facial image matching in facial image and third party's identity data system on effective proof of identification that people provides；
Normalization unit, provides having of people's offer for facial video image, described identity information described in normalized
The facial image matching in facial image and described third party's identity data system on effect proof of identification；
Extraction unit, for from providing people to carry through the facial video image of described normalized, described identity information
For effective proof of identification on facial image and described third party's identity data system in the facial image matching in obtain
Take face characteristic data, respectively as first, second, and third face characteristic data；
Matching unit, for carrying out cross-matched two-by-two by described first, second, and third face characteristic data；
Output unit, for exporting the result of described first, second, and third face characteristic data cross coupling.
Alternatively, described face characteristic data includes face gray feature data, face shape characteristic and people's face
Skin textural characteristics data.
Alternatively, described matching unit includes:
Gray-scale Matching subelement, for the face gray feature in described first, second, and third face characteristic data
Data is compared successively two-by-two, obtains gray feature identification score value sequence；
Form fit subelement, for the face shape feature in described first, second, and third face characteristic data
Data is compared successively two-by-two, obtains shape facility identification score value sequence；
Texture Matching subelement, for the face dermatoglyph in described first, second, and third face characteristic data
Characteristic is compared successively two-by-two, obtains textural characteristics identification score value sequence；
Merge coupling subelement, for described gray feature identification score value sequence, described shape recognition score value sequence and
The numerical value of the identical precedence in described texture recognition score value sequence executes multiple features fusion identification respectively, obtains fusion recognition score value
Sequence.
Alternatively, described matching unit also includes:
Comprehensive matching subelement, for described gray feature identification score value sequence, described shape recognition score value sequence, institute
State the numerical value numerical value with identical precedence in described fusion recognition score value sequence respectively of identical precedence in texture recognition score value sequence
Execution multiple Classifiers Combination, obtains comprehensively identifying score value sequence.
Alternatively, described Identification of Images device also includes:
Alarm unit, for being compared the result of described cross-matched with default threshold value, when described cross-matched
Result be less than described threshold value when, issue warning information.
Alternatively, described alarm unit includes:
First alarm part-unit, for by identical bits in described fusion recognition score value sequence and default first threshold sequence
Secondary numerical value is compared, when described fusion recognition score value sequence is less than the numerical value of identical precedence in described first threshold sequence
When, issue warning information.
Alternatively, described alarm unit includes:
Second alarm part-unit, for comprehensively identifying score value sequence and identical bits in default Second Threshold sequence by described
Secondary numerical value is compared, when described comprehensive identification score value sequence is less than the numerical value of identical precedence in described Second Threshold sequence
When, issue warning information.
Compared with prior art, the technical scheme of the embodiment of the present invention has the advantage that
By the facial video image being gathered and identity information being provided the face on effective proof of identification of people's offer
The facial image matching in image and third party's identity data system carries out cross-matched, not only can identify proof of identification
Whether the facial image on file is matched with the facial image in third party's identity data system, may recognize that identity card
The face figure in facial image and third party's identity data system on the bright facial image that people is provided, described documentation of identity
Seem no be mutually matched respectively, therefore can improve the degree of accuracy of Identification of Images so that online trading is more safe and reliable.
Further, when carrying out face characteristic identification, by score value sequence, described shape are identified to described gray feature
Identification score value sequence, described texture recognition score value sequence are held with the numerical value of identical precedence in described fusion recognition score value sequence respectively
Row multiple Classifiers Combination, obtains the comprehensive accuracy identifying score value sequence, can improving face characteristic identification further.
Further, due to being compared the result of described cross-matched with default threshold value, when described cross-matched
Result be less than described threshold value when, issue warning information, to user to point out, to take appropriate measures, it is possible to reduce by
The potential safety hazard leading in user's careless omission.
Brief description
Fig. 1 is the flow chart of a kind of portrait identification method in the embodiment of the present invention；
Fig. 2 is the flow chart of another kind of portrait identification method in the embodiment of the present invention；
Fig. 3 is the structural representation of the Identification of Images device in the embodiment of the present invention.
Specific embodiment
The technical scheme that the embodiment of the present invention adopts is by providing people by the facial video image being gathered and identity information
The facial image matching in the facial image on effective proof of identification providing and third party's identity data system is intersected
Coupling, not only can identify that the facial image on documentation of identity with the facial image in third party's identity data system is
No match, may recognize that proof of identification provide the facial image on facial image and the described documentation of identity of people with
Whether the facial image in third party's identity data system is mutually matched respectively, improves the degree of accuracy of Identification of Images so that line
Upper transaction is more safe and reliable.
Understandable for enabling the above objects, features and advantages of the present invention to become apparent from, below in conjunction with the accompanying drawings to the present invention
Specific embodiment be described in detail.
Refer to Fig. 1, the flow chart that it illustrates the portrait identification method in the embodiment of the present invention.Described Identification of Images side
Method includes:
Step s11, captured identity information provides the facial video image of people, and obtains having of identity information offer people's offer
The facial image matching in facial image and third party's identity data system on effect proof of identification.
Wherein, described identity information provides the facial video image of people can be acquired by web camera, effectively
Proof of identification can be identity card, passport etc., and third party's identity data system can be id5 identity network data system.
Step s12, described in normalized, facial video image, described identity information provide people the effective identity card providing
The facial image matching in facial image and described third party's identity data system on bright.
Step s13, from providing what people provided to have through the facial video image of described normalized, described identity information
Face is obtained in the facial image matching in facial image and described third party's identity data system on effect proof of identification
Characteristic, respectively as first, second, and third face characteristic data.
Step s14, described first, second, and third face characteristic data is carried out cross-matched two-by-two.
Step s15, the result of the described first, second, and third face characteristic data cross coupling of output.
The portrait identification method of the embodiment of the present invention, by providing people by the facial video image being gathered and identity information
The facial image matching in the facial image on effective proof of identification providing and third party's identity data system is carried out two-by-two
Cross-matched, not only can identify the facial image on documentation of identity and the face figure in third party's identity data system
Seem no match, may recognize that proof of identification provides the face figure on facial image and the described documentation of identity of people
As whether matching respectively with the facial image in third party's identity data system, improve the degree of accuracy of Identification of Images so that
Online trading is more safe and reliable.
With reference to Fig. 2, the flow chart that it illustrates another kind of portrait identification method in the embodiment of the present invention.Described Identification of Images
Method includes:
Step s21, captured identity information provides the facial video image of people, and obtains having of identity information offer people's offer
The facial image matching in facial image and third party's identity data system on effect proof of identification.
In a specific embodiment, identity information provides the facial video image of people can gather by web camera,
Facial image effectively on proof of identification can provide people the effective documentation of identity such as identity card providing from identity information
Upper acquisition, the facial image matching in third party's identity data system can obtain from such as id5 identity grid database.
Step s22, described in normalized, facial video image, described identity information provide people the effective identity card providing
The facial image matching in facial image and described third party's identity data system on bright.
Normalized image by a series of conversion, that is, can utilize the not bending moment of image to find one group of parameter can
Enough eliminate the impact that other transforming function transformation functions convert to image, pending original image is converted into corresponding sole criterion shape
Formula.This canonical form image has invariant feature to translation, rotation, scaling equiaffine conversion.Therefore, step s22 can be by institute
The described facial video image that obtains, described identity information provide the facial image and described on effective proof of identification that people provides
The facial image matching in third party's identity data system is converted into corresponding canonical form image respectively, can obtain
More accurate face characteristic data.
Step s23, from providing what people provided to have through the facial video image of described normalized, described identity information
Face is obtained in the facial image matching in facial image and described third party's identity data system on effect proof of identification
Characteristic, respectively as first, second, and third face characteristic data.
Wherein, face characteristic data can include face gray feature data, face shape characteristic and people's face skin
Textural characteristics data.
Step s24, described first, second, and third face characteristic data is carried out cross-matched two-by-two.
In being embodied as, can be by by described first face characteristic, described second face characteristic data and the
Three face characteristic data are compared successively two-by-two, obtain corresponding feature recognition score value, and composition characteristic identifies score value sequence.
Specifically, step s24 may include that first, special to the gray scale in first, second, third face characteristic data
Levy data and carry out gray feature identification two-by-two, obtain first, second, and third gray scale identification score value, composition gray scale identification point successively
Value sequence.Character shape data in first, second, and third face characteristic data is carried out with shape facility identification two-by-two, successively
Obtain first, second, and third shape recognition score value, formed shape identifies score value sequence；Special to first, second and third face
Levy the face dermatoglyph characteristic in data and carry out textural characteristics identification two-by-two, obtain first, second, and third line successively
Reason identification score value, forms texture recognition score value sequence.Then, to described gray feature identification score value sequence, described shape recognition
Score value sequence executes multiple features fusion identification respectively with the numerical value of the identical precedence in described texture recognition score value sequence, is melted
Close identification score value sequence.Step s24 is intended to obtain the fusion matching result of face characteristic data.Multiple features fusion identification therein
The multiple features fusion identification of layer can be quantified using coupling, carry out respectively mating by each characteristic obtaining different amounts first
Change value, is then normalized to quantized value, finally with a unified rule, all quantized values after normalization is counted
Calculation obtains a quantized value.In being embodied as, multiple features fusion identification may include that to first gray scale identify score value, first
Shape recognition score value and the execution multiple features fusion identification of the first texture recognition score value, obtain the first fusion recognition score value；To second
Gray scale identification score value, the second shape recognition score value and the execution multiple features fusion identification of the second texture recognition score value, obtain second and melt
Close identification score value；3rd gray scale is identified with score value, the 3rd shape recognition score value and third texture identification score value execution multiple features melt
Close identification, obtain the 3rd fusion recognition score value；Described first, second, and third fusion recognition score value composition fusion recognition score value sequence
Row.By quantifying the fusion recognition score value sequence that the multiple features fusion identification of layer draws using coupling, reflect first, second He
The synthesis result that 3rd face characteristic data is mated two-by-two.According to this synthesis result, corresponding judged result can be exported.
Step s25, the result of the described first, second, and third face characteristic data cross coupling of output.
The synthesis result that first, second, and third face characteristic data can be mated by step s25 two-by-two, that is, by institute
The first, second, and third fusion recognition score value stated is shown to user by intuitive way so that user can be easily
Solution matching result.
Step s26, the result of described cross-matched is compared with default threshold value, when the result of described cross-matched
During less than described threshold value, issue warning information.
In being embodied as, step s26 may include that by described fusion recognition score value sequence respectively with default first threshold
In value sequence, the numerical value of identical precedence is compared, when described fusion recognition score value sequence is less than phase in described first threshold sequence
With precedence numerical value when, issue warning information.Specifically, by first in the first fusion recognition score value and first threshold sequence
Secondary numerical value is compared, and the second fusion recognition score value is compared with the numerical value of the second precedence in first threshold sequence, and the 3rd
Fusion recognition score value is compared with the numerical value of the 3rd precedence in first threshold sequence.In above-mentioned comparison, work as fusion recognition
When score value sequence is less than the numerical value of identical precedence in first threshold sequence, issue warning information.Due to by described cross-matched
Result is compared with default threshold value, when the result of described cross-matched is less than described threshold value, issues warning information, to use
Family is to point out so that user more can grasp the result of Identification of Images comprehensively, convenient and practical.
In being embodied as, in order to improve the degree of accuracy of coupling further, the portrait identification method of the embodiment of the present invention is also
May include that and described gray feature is identified score value sequence, described shape recognition score value sequence, described texture recognition score value sequence
Numerical value execution multiple Classifiers Combination with identical precedence in described fusion recognition score value sequence, obtains comprehensively identifying score value sequence respectively
Row.Specifically, this step may include that and the first gray scale is identified with score value, first shape identification score value, the first texture recognition are divided
Value and the first fusion recognition score value execution multiple Classifiers Combination；To second gray scale identify score value, the second shape recognition score value, second
Texture recognition score value and the second fusion recognition score value execution multiple Classifiers Combination；3rd gray scale is identified with score value, the 3rd shape are known
Other score value, third texture identification score value and the 3rd fusion recognition score value execution multiple Classifiers Combination；Execute above-mentioned many points successively
The fusion of class device respectively obtains first, second, and third comprehensive identification score value, composition comprehensive identification score value sequence.Using multi-categorizer
Merge, the accuracy of face characteristic Data Matching can be improved further.
In being embodied as, in order to when comprehensive identification score value is less than to a certain degree, to user to point out, facilitate user to enter
One step takes appropriate measures, and the portrait identification method of the embodiment of the present invention can also include: by described comprehensive identification score value sequence
Row are compared with the numerical value of identical precedence in default Second Threshold sequence, when described comprehensive identification score value sequence is less than described
In Second Threshold sequence during the numerical value of identical precedence, issue warning information.
Specifically, this step may include that the first comprehensive identification score value and the first precedence in Second Threshold sequence
Numerical value is compared, and the second comprehensive identification score value is compared with the numerical value of the second precedence in Second Threshold sequence, and the 3rd is comprehensive
Identification score value is compared with the numerical value of the 3rd precedence in Second Threshold sequence.In above-mentioned comparison, when comprehensive identification score value
Sequence be less than first threshold sequence in identical precedence numerical value when, issue warning information, with to associative operation personnel to point out.
Refer to Fig. 3, it illustrates the structural representation of the Identification of Images device of the embodiment of the present invention.Described Identification of Images
Device includes collection acquiring unit 1, normalization unit 2, extraction unit 3, matching unit 4 and the output unit 5 being sequentially connected.
Wherein, gather acquiring unit 1, the facial video image of people is provided for captured identity information, and obtain identity letter
Breath provides the face figure matching in facial image and third party's identity data system on effective proof of identification that people provides
Picture.
Normalization unit 2, provides having of people's offer for facial video image, described identity information described in normalized
The facial image matching in facial image and described third party's identity data system on effect proof of identification.
Extraction unit 3, for from providing people to carry through the facial video image of described normalized, described identity information
For effective proof of identification on facial image and described third party's identity data system in the facial image matching in obtain
Take face characteristic data, respectively as first, second, and third face characteristic data.
Matching unit 4, for carrying out cross-matched two-by-two by described first, second, and third face characteristic data.
Output unit 5, for exporting the result of described first, second, and third face characteristic data cross coupling.
Because the facial video image being gathered and identity information are provided the effective of people's offer by above-mentioned Identification of Images device
The facial image matching in facial image on proof of identification and third party's identity data system carries out cross-matched, not only may be used
To identify whether the facial image on documentation of identity is matched with the facial image in third party's identity data system, also
Can identify that proof of identification provides the facial image on the facial image and described documentation of identity of people and third party's identity
Whether the facial image in data system is mutually matched respectively, therefore can improve the degree of accuracy of Identification of Images so that line is submitted
Easily more safe and reliable.
In order to when the matching result of facial image characteristic is less than to a certain degree, to user to point out, with further
Take appropriate measures, the Identification of Images device of the embodiment of the present invention can also include alarm unit 6.Alarm unit 6 can be used
In being compared the result of described cross-matched with default threshold value, when the result of described cross-matched is less than described threshold value
When, issue warning information.
In being embodied as, matching unit 4 can include Gray-scale Matching subelement 41, form fit subelement 42, texture
Coupling subelement 43, fusion coupling subelement 44.Wherein,
Gray-scale Matching subelement 41, for special to the face gray scale in described first, second, and third face characteristic data
Levy data to be compared two-by-two successively, obtain gray feature identification score value sequence.
Form fit subelement 42, for special to the face shape in described first, second, and third face characteristic data
Levy data to be compared two-by-two successively, obtain shape facility identification score value sequence.
Texture Matching subelement 43, for the face dermatoglyph to described first, second, and third face characteristic data
Characteristic is compared successively two-by-two, obtains textural characteristics identification score value sequence.
Merge coupling subelement 44, for described gray feature identification score value sequence, described shape recognition score value sequence
Execute multiple features fusion identification respectively with the numerical value of the identical precedence in described texture recognition score value sequence, obtain fusion recognition and divide
Value sequence.
In being embodied as, for improving the accuracy of face characteristic Data Matching further, matching unit 4 can also include
Comprehensive matching subelement 45, for described gray feature identification score value sequence, described shape recognition score value sequence, described texture
In identification score value sequence, the numerical value of identical precedence is many with the numerical value execution of identical precedence in described fusion recognition score value sequence respectively
Multiple Classifier Fusion, obtains comprehensively identifying score value sequence.
In being embodied as, alarm unit 6 can include the first alarm part-unit 61 and the second alarm part-unit 62.Its
In, the first alarm part-unit 61 can be used for described fusion recognition score value sequence is identical with default first threshold sequence
The numerical value of precedence is compared, when described fusion recognition score value sequence is less than the numerical value of identical precedence in described first threshold sequence
When, issue warning information.Second alarm part-unit 62 can be used for comprehensively identifying score value sequence and default second by described
In threshold series, the numerical value of identical precedence is compared, when described comprehensive identification score value sequence is less than in described Second Threshold sequence
During the numerical value of identical precedence, issue warning information.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
Completed with the hardware instructing correlation by program, this program can be stored in computer-readable recording medium, storage is situated between
Matter may include that rom, ram, disk or CD etc..
Above the method for the present invention and device are discussed in detail, the present invention is not limited to this.Any art technology
Personnel, without departing from the spirit and scope of the present invention, all can make various changes or modifications, and therefore protection scope of the present invention should
When being defined by claim limited range.
Claims (10)
1. a kind of portrait identification method is it is characterised in that include:
Captured identity information provides the facial video image of people, and obtains on effective proof of identification that identity information offer people provides
Facial image and third party's identity data system in the facial image matching；
Described in normalized, facial video image, described identity information provide the face figure on effective proof of identification of people's offer
The facial image matching in picture and described third party's identity data system；
From through the facial video image of described normalized, effective proof of identification of described identity information offer people's offer
Facial image and described third party's identity data system in the facial image matching in obtain face characteristic data, respectively
As first, second, and third face characteristic data；Wherein, described face characteristic data includes face gray feature data, people
Face shape characteristic and face dermatoglyph characteristic；
Described first, second, and third face characteristic data is carried out cross-matched two-by-two, comprising: to described first, second He
Face gray feature data in 3rd face characteristic data is compared successively two-by-two, obtains gray feature identification score value sequence
Row；Face shape characteristic in described first, second, and third face characteristic data is compared successively two-by-two, is obtained
Shape facility identifies score value sequence；
To the face dermatoglyph characteristic in described first, second, and third face characteristic data successively two-by-two than
Right, obtain textural characteristics identification score value sequence；To described gray feature identification score value sequence, described shape recognition score value sequence and
The numerical value of the identical precedence in described texture recognition score value sequence executes multiple features fusion identification respectively, obtains fusion recognition score value
Sequence；
Export the result of described first, second, and third face characteristic data cross coupling.
2. portrait identification method according to claim 1 it is characterised in that described by described first, second, and third people
Face characteristic carries out cross-matched two-by-two, also includes: described gray feature is identified score value sequence, described shape recognition score value
In sequence, described texture recognition score value sequence the numerical value of identical precedence respectively with identical precedence in described fusion recognition score value sequence
Numerical value execution multiple Classifiers Combination, obtain comprehensive identifying score value sequence.
3. portrait identification method according to claim 2 is it is characterised in that also include: by the result of described cross-matched
It is compared with default threshold value, when the result of described cross-matched is less than described threshold value, issue warning information.
4. portrait identification method according to claim 3 it is characterised in that the described result by described cross-matched with pre-
If threshold value be compared, when described cross-matched result be less than described threshold value when, issue warning information, comprising: will be described
Fusion recognition score value sequence is compared with the numerical value of identical precedence in default first threshold sequence respectively, when described fusion is known
When other score value sequence is less than the numerical value of identical precedence in described first threshold sequence, issue warning information.
5. portrait identification method according to claim 3 it is characterised in that the described result by described cross-matched with pre-
If threshold value be compared, when described cross-matched result be less than described threshold value when, issue warning information, comprising: will be described
Comprehensive identification score value sequence is compared with the numerical value of identical precedence in default Second Threshold sequence respectively, when described comprehensive knowledge
When other score value sequence is less than the numerical value of identical precedence in described Second Threshold sequence, issue warning information.
6. a kind of Identification of Images device is it is characterised in that include:
Gathering acquiring unit, provide the facial video image of people for captured identity information, and obtain identity information provides people to carry
For effective proof of identification on facial image and third party's identity data system in the facial image matching；
Normalization unit, provides people the effective body providing for facial video image, described identity information described in normalized
The facial image matching in facial image and described third party's identity data system that part proves；
Extraction unit, for from providing people to provide through the facial video image of described normalized, described identity information
People is obtained in the facial image matching in facial image and described third party's identity data system effectively on proof of identification
Face characteristic, respectively as first, second, and third face characteristic data；Wherein, described face characteristic data includes face
Gray feature data, face shape characteristic and face dermatoglyph characteristic；
Matching unit, for carrying out cross-matched two-by-two by described first, second, and third face characteristic data, comprising: gray scale
Coupling subelement, for carrying out successively to the face gray feature data in described first, second, and third face characteristic data
Compare two-by-two, obtain gray feature identification score value sequence；
Form fit subelement, for the face shape characteristic in described first, second, and third face characteristic data
Compared two-by-two successively, obtained shape facility identification score value sequence；
Texture Matching subelement, for the face dermatoglyph feature in described first, second, and third face characteristic data
Data is compared successively two-by-two, obtains textural characteristics identification score value sequence；
Merge coupling subelement, for described gray feature identification score value sequence, described shape recognition score value sequence and described
The numerical value of the identical precedence in texture recognition score value sequence executes multiple features fusion identification respectively, obtains fusion recognition score value sequence
Row；
Output unit, for exporting the result of described first, second, and third face characteristic data cross coupling.
7. Identification of Images device according to claim 6 is it is characterised in that described matching unit also includes:
Comprehensive matching subelement, for described gray feature identification score value sequence, described shape recognition score value sequence, described line
The numerical value of the identical precedence numerical value execution with identical precedence in described fusion recognition score value sequence respectively in reason identification score value sequence
Multiple Classifiers Combination, obtains comprehensively identifying score value sequence.
8. Identification of Images device according to claim 7 is it is characterised in that also include:
Alarm unit, for being compared the result of described cross-matched with default threshold value, when the knot of described cross-matched
When fruit is less than described threshold value, issue warning information.
9. Identification of Images device according to claim 8 is it is characterised in that described alarm unit includes:
First alarm part-unit, for by described fusion recognition score value sequence and identical precedence in default first threshold sequence
Numerical value is compared, and when described fusion recognition score value sequence is less than the numerical value of identical precedence in described first threshold sequence, sends out
Cloth warning information.
10. Identification of Images device according to claim 8 is it is characterised in that described alarm unit includes:
Second alarm part-unit, for comprehensively identifying score value sequence and identical precedence in default Second Threshold sequence by described
Numerical value is compared, and when described comprehensive identification score value sequence is less than the numerical value of identical precedence in described Second Threshold sequence, sends out
Cloth warning information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310714175.4A CN103699887B (en) | 2013-12-20 | 2013-12-20 | Portrait identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310714175.4A CN103699887B (en) | 2013-12-20 | 2013-12-20 | Portrait identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103699887A CN103699887A (en) | 2014-04-02 |
| CN103699887B true CN103699887B (en) | 2017-01-18 |
Family
ID=50361410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310714175.4A Active CN103699887B (en) | 2013-12-20 | 2013-12-20 | Portrait identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103699887B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106161397A (en) * | 2015-04-21 | 2016-11-23 | 富泰华工业（深圳）有限公司 | There is the electronic installation of Anti-addiction function, Anti-addiction management system and method |
| CN105138985A (en) * | 2015-08-25 | 2015-12-09 | 北京拓明科技有限公司 | Real-name authentication method based on WeChat public number and system |
| CN106250739A (en) * | 2016-07-19 | 2016-12-21 | 柳州龙辉科技有限公司 | A kind of identity recognition device |
| CN107292620A (en) * | 2017-06-14 | 2017-10-24 | 浪潮金融信息技术有限公司 | Personal identification method and device, computer-readable recording medium, terminal |
| CN107967453A (en) * | 2017-11-24 | 2018-04-27 | 河北三川科技有限公司 | Hotel occupancy identity checking method and verifying system based on recognition of face |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102800017A (en) * | 2012-07-09 | 2012-11-28 | 高艳玲 | Identity verification system based on face recognition |
| CN203287910U (en) * | 2013-05-14 | 2013-11-13 | 苏州福丰科技有限公司 | Search system based on face recognition |
| CN103440482A (en) * | 2013-09-02 | 2013-12-11 | 北方工业大学 | Method, system and device for identifying identity document holder based on hidden video |
| CN103440327A (en) * | 2013-09-02 | 2013-12-11 | 北方工业大学 | Method and system for quick comparison of online wanted men through hidden video |
-
2013
- 2013-12-20 CN CN201310714175.4A patent/CN103699887B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102800017A (en) * | 2012-07-09 | 2012-11-28 | 高艳玲 | Identity verification system based on face recognition |
| CN203287910U (en) * | 2013-05-14 | 2013-11-13 | 苏州福丰科技有限公司 | Search system based on face recognition |
| CN103440482A (en) * | 2013-09-02 | 2013-12-11 | 北方工业大学 | Method, system and device for identifying identity document holder based on hidden video |
| CN103440327A (en) * | 2013-09-02 | 2013-12-11 | 北方工业大学 | Method and system for quick comparison of online wanted men through hidden video |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103699887A (en) | 2014-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103699887B (en) | Portrait identification method and device | |
| CN105518711B (en) | Biopsy method, In vivo detection system and computer program product | |
| CN105243357A (en) | Identity document-based face recognition method and face recognition device | |
| CN104680131A (en) | Identity authentication method based on identity certificate information and human face multi-feature recognition | |
| JP2017522635A (en) | User authentication method, apparatus for executing the same, and recording medium storing the same | |
| CN104517104A (en) | Face recognition method and face recognition system based on monitoring scene | |
| WO2018082011A1 (en) | Living fingerprint recognition method and device | |
| CN105335722A (en) | Detection system and detection method based on depth image information | |
| CN102129555A (en) | Second-generation identity card-based authentication method and system | |
| CN107680294A (en) | House property information querying method, system, terminal device and storage medium | |
| Liu et al. | Fingerprint pore matching based on sparse representation | |
| CN101681512A (en) | Vein pattern management system, vein pattern registration device, vein pattern authentication device, vein pattern registration method, vein pattern authentication method, program, and vein data struc | |
| CN104700094A (en) | Face recognition method and system for intelligent robot | |
| CN103324911A (en) | Anti-cheating system based on face recognition | |
| CN103279744A (en) | Multi-scale tri-mode texture feature-based method and system for detecting counterfeit fingerprints | |
| CN103136522A (en) | Finger vein identification technical scheme | |
| CN101711399B (en) | Vein pattern management system, vein pattern registration device, vein pattern authentication device, vein pattern registration method, vein pattern authentication method, vein data storage device | |
| Grother et al. | Biometric specifications for personal identity verification | |
| CN108399335A (en) | A kind of malicious code visual analysis method based on local entropy | |
| CN107292620A (en) | Personal identification method and device, computer-readable recording medium, terminal | |
| CN105184236A (en) | Robot-based face identification system | |
| CN106427891A (en) | Iris intelligent vehicle management system | |
| CN109145845A (en) | House permission electronics resolving method | |
| CN104915590A (en) | Human face recognition system and method applied to computer encryption | |
| CN203149598U (en) | Airport access control system based on face recognition technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |