CN104598796B - Personal identification method and system - Google Patents
Personal identification method and system Download PDFInfo
- Publication number
- CN104598796B CN104598796B CN201510052189.3A CN201510052189A CN104598796B CN 104598796 B CN104598796 B CN 104598796B CN 201510052189 A CN201510052189 A CN 201510052189A CN 104598796 B CN104598796 B CN 104598796B
- Authority
- CN
- China
- Prior art keywords
- user
- text
- biological characteristic
- identity
- voice
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a kind of personal identification method and system, this method includes：In user's registration, the biological characteristic model of generation correspondence user's registration identity；When carrying out user identity identification, generate and show checking text, and typing user reads aloud the voice and image of checking text；The voice and image for reading aloud the user of typing checking text are identified, and obtain the dynamic effective initial time of effective initial time and speech text content, lip and lip reading content of text of voice；If the dynamic effective initial time of lip is consistent with effective initial time of voice, and lip reading content of text and speech text content are consistent with the checking text, then obtain the biological characteristic of user；The biological characteristic of acquisition is matched with the biological characteristic model in user registry database, matching score is obtained；Determine whether user identity is correct according to matching score.Using the present invention, the security of authentication can be effectively improved.
Description
Technical field
The present invention relates to identity identification technical field, and in particular to a kind of personal identification method and system.
Background technology
With the popularization of development of Mobile Internet technology, mobile payment has become non-in people's work, study, amusement, life
An often important part.Therewith, the safety problem that mobile payment is brought increasingly is highlighted, how to ensure the information security of user with
And how one important step of mobile payment is had become to the convenient effective confirmation of user identity progress.The payment of current main flow
Platform, is to use traditional pin mode mostly, or coordinate the methods such as dynamic token (token) to confirm user identity.This method
The protective effect to user profile is served to a certain extent.
But, conventional method can not be realized to be confirmed to the identity of user, is only merely that code content is carried out really
Recognize, huge economic loss can be caused if password is stolen by other people.And to have many scenes to need to use close by current user
Code, if forgetting can be easily caused using different passwords respectively；It is easy to be passed through by people if unified password is used
Other channels are stolen, such as hitting storehouse attack pattern.Therefore, conventional cipher mode is in nowadays fast-developing mobile Internet
In generation, progressively shows the deficiency of its security, validity and convenience.
Therefore, biological identification has been increasingly becoming the focus of current research.Biological identification is also known as bio-identification, is by calculating
Machine carries out personal identification identification using the intrinsic physiological characteristic of human body or behavioural characteristic.Biological identification and other authenticating parties
Method compares, and has the advantages that oneself is unique：First is security.Second is convenience --- with using key, card, order
Board or PIN are compared, and biological identification is more convenient.3rd is that asking for loss forgetting is not present in biological characteristic
Topic, also there is no concern that the problem of password, identification number are stolen.Finally, biological characteristic has good stability and persistence.
At this stage, biological identification generally uses single biological characteristic, even if being also simple using multiple biological characteristics
Each biological characteristic is identified respectively successively for ground.Therefore, attacker can still take the means broken up one by one, such as recording is attacked
Hit vocal print system, video recording attack face system etc.；System is caused easily to be broken.Greatly reduce the security of system.
The content of the invention
The embodiment of the present invention provides a kind of personal identification method and system, to improve the security of authentication.
Therefore, the embodiment of the present invention provides following technical scheme：
A kind of personal identification method, including：
In user's registration, the biological characteristic model of generation correspondence user's registration identity, and by the biological characteristic
Model is saved in user registry database；
When carrying out user identity identification, User Identity is obtained；
Generate and show checking text, and typing user reads aloud the voice and image of the checking text；
The voice and image for reading aloud the user of typing the checking text are identified, obtain the effective of the voice
Initial time and the dynamic effective initial time of speech text content, lip and lip reading content of text；
If the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and
Speech text content is consistent with the checking text, then obtains the biological characteristic of user；
Based on the User Identity, by the biological characteristic model in the biological characteristic and user registry database of acquisition
Matched, obtain matching score；
Determine whether user identity is correct according to the matching score.
Preferably, methods described also includes：
Check and whether there is the User Identity in the user registry database；；
If it is, performing the step of generating and show checking text；Otherwise, it determines user identity mistake.
Alternatively, the biological characteristic model is sound-groove model；
The biological characteristic for obtaining user includes：Obtained in the voice for reading aloud the checking text from the user of the typing
Take the vocal print feature at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, is matched
Score includes：By vocal print feature sound-groove model progress corresponding with User Identity described in user registry database
Match somebody with somebody, obtain matching score；
Or
The biological characteristic model is faceform；
The biological characteristic for obtaining user includes：Obtained in the image for reading aloud the checking text from the user of the typing
Take the face characteristic at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, is matched
Score includes：By face characteristic faceform's progress corresponding with User Identity described in user registry database
Match somebody with somebody, obtain matching score；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic for obtaining user includes：Obtained in the voice for reading aloud the checking text from the user of the typing
The face that user is obtained in taking the vocal print feature at family, and the image for reading aloud the checking text from the user of the typing is special
Levy；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, is matched
Score includes：
The vocal print feature of acquisition sound-groove model corresponding with User Identity described in user registry database is carried out
Matching, obtains the first score；
The face characteristic of acquisition faceform corresponding with User Identity described in user registry database is carried out
Matching, obtains the second score；
First score is merged with the second score, matching score is obtained.
Correspondingly, it is described to determine whether user identity correctly includes according to the matching score：
If the matching score is more than the threshold value of setting, it is determined that user identity is correct；Otherwise determine that user identity is wrong
By mistake.
Alternatively, the biological characteristic model is sound-groove model；
The biological characteristic for obtaining user includes：Obtained in the voice for reading aloud the checking text from the user of the typing
Take the vocal print feature at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, is matched
Score includes：The vocal print feature is matched with each sound-groove model in user registry database, many candidates is obtained and matches
Point；
Or
The biological characteristic model is faceform；
The biological characteristic for obtaining user includes：Obtained in the image for reading aloud the checking text from the user of the typing
Take the face characteristic at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, is matched
Score includes：The face characteristic is matched with each faceform in user registry database, many candidate matches is obtained and obtains
Point；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic for obtaining user includes：
The vocal print feature of user is obtained in the voice for reading aloud the checking text from the user of the typing, and from the record
The user entered reads aloud the face characteristic that user is obtained in the image of the checking text；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, is matched
Score includes：
By the vocal print feature and face characteristic of acquisition respectively with the vocal print feature in user registry database and faceform
Fusion matching is carried out, many candidate matches scores are obtained.
Correspondingly, it is described to determine whether user identity correctly includes according to the matching score：
Judge the corresponding user's registration identity of matching score highest characteristic model and the user identity obtained
Identify whether identical；
If it is, determining that user identity is correct；Otherwise user identity mistake is determined.
Preferably, methods described also includes：
After user's registration request is received, generate and show Login script, and typing user reads aloud the Login script
Voice and image；
The voice and image for reading aloud the user of typing the Login script are identified, obtain the effective of the voice
Initial time and the dynamic effective initial time of speech text content, lip and lip reading content of text；
If the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and
Speech text content is consistent with the Login script, then allows registration；Otherwise registration is forbidden；
After registration is allowed, then perform the step of generation corresponds to the biological characteristic model of user's registration identity.
Alternatively, the checking text and/or Login script are static text or stochastic and dynamic text.
A kind of identification system, including：
Registering modules, in user's registration, generating the biological characteristic model of correspondence user's registration identity, and will
The biological characteristic model is saved in user registry database；
Identity acquisition module, for when carrying out user identity identification, obtaining User Identity；
Text processing module, for generating and showing checking text；
Typing module, the voice and image of the checking text are read aloud for typing user；
Identification module, the voice and image for reading aloud the checking text for the user to typing is identified, and obtains institute
The dynamic effective initial time of the effective initial time and speech text content, lip of predicate sound and lip reading content of text, and judge institute
State whether the dynamic effective initial time of lip is consistent with effective initial time of voice, and in lip reading content of text and speech text
Hold and whether be consistent with the checking text；；
Biological characteristic extraction module, for the effective initial time and effective initial time phase of voice moved in the lip
Symbol, and lip reading content of text and speech text content with it is described checking text be consistent when, obtain user biological characteristic；
Matching module, for based on the User Identity, the biology that the biological characteristic extraction module is obtained to be special
Levy and matched with the biological characteristic model in user registry database, obtain matching score；
Judge module, for determining whether user identity is correct according to the matching score.
Preferably, the system also includes：
Module is checked, the User Identity is whether there is in the user registry database for checking；If it is,
The text processing module is then triggered to generate and show checking text；Otherwise user identity mistake is determined.
Alternatively, the biological characteristic model is sound-groove model；The biological characteristic extraction module is specifically for from described
The user of typing reads aloud the vocal print feature that user is obtained in the voice of the checking text；The matching judgment module is by the life
The vocal print feature sound corresponding with the User Identity that the identity acquisition module is obtained that thing characteristic extracting module is obtained
Line model is matched, and obtains matching score；
Or
The biological characteristic model is faceform；The biological characteristic extraction module is specifically for the use from the typing
Read aloud the face characteristic that user is obtained in the image of the checking text in family；The matching judgment module carries the biological characteristic
The face characteristic faceform corresponding with the User Identity that the identity acquisition module is obtained that modulus block is obtained enters
Row matching, obtains matching score；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic extraction module includes：
First extraction unit, the sound for obtaining user in reading aloud the voice for verifying text from the user of the typing
Line feature；
Second extraction unit, the people for obtaining user in reading aloud the image for verifying text from the user of the typing
Face feature；
The matching module includes：
First matching unit, vocal print feature and the identity for first extraction unit to be obtained obtain mould
The corresponding sound-groove model of identity that block is obtained is matched, and obtains the first score；
Second matching unit, face characteristic and the identity for second extraction unit to be obtained obtain mould
The corresponding faceform of identity that block is obtained is matched, and obtains the second score；
Integrated unit, for first score to be merged with second score, obtains matching score.
Correspondingly, the judge module, specifically for when the matching score is more than the threshold value set, determining user's body
Part is correct；Otherwise user identity mistake is determined.
Alternatively, the biological characteristic model is sound-groove model；
The biological characteristic extraction module, in reading aloud the voice for verifying text from the user of the typing
Obtain the vocal print feature of user；
A matching module, specifically for each sound-groove model in the vocal print feature and user registry database is carried out
Match somebody with somebody, much candidate matching score；
Or
The biological characteristic model is faceform；
The biological characteristic extraction module, in reading aloud the image for verifying text from the user of the typing
Obtain the face characteristic of user；
A matching module, specifically for each faceform in the face characteristic and user registry database is carried out
Match somebody with somebody, obtain many candidate matches scores；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic extraction module includes：
First extraction unit, the sound for obtaining user in reading aloud the voice for verifying text from the user of the typing
Line feature；
Second extraction unit, the people for obtaining user in reading aloud the image for verifying text from the user of the typing
Face feature；
The matching module, extracts single specifically for the vocal print feature for obtaining first extraction unit and described second
The face characteristic that member is obtained carries out merging matching with the vocal print feature in user registry database and faceform respectively, obtains many
Candidate matches score.
Correspondingly, the judge module, specifically for judging the corresponding registered user of matching score highest characteristic model
Identify whether identical with the identity of acquisition；If it is, determining that user identity is correct；Otherwise user identity mistake is determined.
Preferably, the text processing module, is additionally operable to after the Registering modules receive user's registration request, generation
And show Login script；
The typing module, is additionally operable to voice and image that typing user reads aloud the Login script；
The identification module, the voice and image for being additionally operable to read aloud the user of typing the Login script is identified,
Effective initial time and lip reading content of text that the effective initial time and speech text content, lip of the voice are moved are obtained, and
Judge whether the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and voice
Whether content of text is consistent, and when being consistent, determination allows registration, otherwise forbids registration with the checking text；
The Registering modules specifically for after the identification module determines to allow registration, generation correspondence user's registration body
The biological characteristic model of part mark, and the biological characteristic model is saved in user registry database.
Personal identification method provided in an embodiment of the present invention and system, when carrying out user identity identification, generate and show
Text is verified, then typing user reads aloud the voice and image of checking text, and the voice and image are identified, obtain described
The dynamic effective initial time of the effective initial time and speech text content, lip of voice and lip reading content of text；It is dynamic in the lip
Effective initial time be consistent with effective initial time of voice and lip reading content of text and speech text content with it is described
In the case that checking text is consistent, identity legitimacy checking further is carried out to user using the biological characteristic of user.Due to
Dynamic biological characteristic and static biological characteristic are combined in identification procedure, therefore efficiently avoid user profile quilt
Risk that is counterfeit, stealing, substantially increases the security of authentication；And the biological characteristic of user is inherent, it is not necessary to
Especially memory, so that improving user identity tests the convenience recognized, validity.
Brief description of the drawings
, below will be to institute in embodiment in order to illustrate more clearly of the embodiment of the present application or technical scheme of the prior art
The accompanying drawing needed to use is briefly described, it should be apparent that, drawings in the following description are only one described in the present invention
A little embodiments, for those of ordinary skill in the art, can also obtain other accompanying drawings according to these accompanying drawings.
Fig. 1 is the flow chart of personal identification method of the embodiment of the present invention；
Fig. 2A is characteristic matching and a kind of flow chart of authentication in the inventive method；
Fig. 2 B are characteristic matching and another flow chart of authentication in the inventive method；
Fig. 2 C are characteristic matching and another flow chart of authentication in the inventive method；
Fig. 2 D are characteristic matching and another flow chart of authentication in the inventive method；
Fig. 3 is a kind of flow chart of user's registration in the embodiment of the present invention；
Fig. 4 is a kind of structural representation of identification system of the embodiment of the present invention；
Fig. 5 A are a kind of concrete structure schematic diagrams of matching module in the embodiment of the present invention；
Fig. 5 B are another concrete structure schematic diagrams of matching module in the embodiment of the present invention；
Fig. 5 C are another concrete structure schematic diagrams of matching module in the embodiment of the present invention.
Embodiment
In order that those skilled in the art more fully understand the scheme of the embodiment of the present invention, below in conjunction with the accompanying drawings and implement
Mode is described in further detail to the embodiment of the present invention.
As shown in figure 1, being the flow chart of personal identification method provided in an embodiment of the present invention, comprise the following steps：
Step 101, in user's registration, the biological characteristic model of generation correspondence user's registration identity, and will be described
Biological characteristic model is saved in user registry database.
The biological characteristic model can be sound-groove model, and/or faceform, naturally it is also possible to be other biological characteristics
Model, such as, fingerprint, iris etc. are not limited this embodiment of the present invention.The registration process of user will be said in detail below
It is bright.
Step 102, when carrying out user identity identification, User Identity is obtained.
Step 103, generate and show checking text, and typing user reads aloud the voice and image of the checking text.
It should be noted that the checking text can be fixed text or stochastic and dynamic text.Tested in display
While demonstrate,proving text, the prompt message for pointing out user to read aloud the checking text can be shown, such as can be that " please read aloud
Show word " etc..During user reads aloud the checking text, start recording and record a video, voice when typing user reads aloud
And image.
Step 104, the voice and image for reading aloud the user of typing the checking text are identified, and obtain the voice
Effective initial time and the dynamic effective initial time of speech text content, lip and lip reading content of text.
Effective initial time of voice can pass through VAD (Voice Activity Detector, voice activation detection) etc.
Technology realizes that speech recognition can utilize existing speech recognition technology.
The dynamic positioning of lip and feature extraction are the prerequisites of lip reading identification, can specifically use some existing identification sides
Method, such as, method-variable formwork and Snake methods based on shape, the method based on image-principal component analysis method are based on
The method of motion-optical flow analysis method etc..
Step 105, judge whether the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip
Whether language content of text and speech text content are consistent with the checking text；If it is, performing step 106；Otherwise, hold
Row step 109.
The dynamic effective initial time of the lip is consistent the effective initial time for referring to that lip is dynamic with effective initial time of voice
Differed with effective initial time of voice in the range of allowing, such as in 1s.
Lip reading content of text and speech text content with it is described checking text be consistent refer to lip reading content of text with it is described
Verify that the matching degree and speech text content of text and the matching degree of the checking text are satisfied by certain requirement, such as it is, right
Lip reading content of text is to verifying the matching degree and speech text content of text with verifying that the matching degree of text sets corresponding respectively
Threshold value, when calculating obtained matching degree and being more than corresponding threshold value, determine that content is consistent, otherwise determine content not phase
Symbol.
By judging whether the dynamic effective initial time of lip is consistent with effective initial time of voice, it is ensured that to be identified to use
The execute-in-place at family, it is to avoid the attack of the mode such as recording, video recording.
Step 106, the biological characteristic of user is obtained.
The specific category of the biological characteristic needs biological characteristic used during with user's registration identical, such as, vocal print,
Face, fingerprint, iris etc., can be any one in these biological characteristics or its any combination.
Step 107, based on the User Identity, by the biology in the biological characteristic and user registry database of acquisition
Characteristic model is matched, and obtains matching score.
The species of biological characteristic of the specific matching process of biological characteristic to being extracted is related, can specifically use existing one
A little matching ways, are not limited this embodiment of the present invention.
Step 108, user's body is determined according to the matching score whether part is correct.
Step 109, user identity mistake is determined.
Carrying out identification by the present invention program needs first to carry out the registration of user identity, in user's registration, obtains
User's registration identity, and according to the biological characteristic of user, generate to should enrollment status mark biological characteristic model.With
Family enrollment status mark can be numeral, letter, special symbol or its combination etc..
In another embodiment of the inventive method, when carrying out user identity identification, user's note can also be first checked
It whether there is the identity in volume database.
The checking process of above-mentioned identity can be carried out before above-mentioned steps 103, i.e. if user's registration identity
Mark includes the User Identity obtained, then step 103 is performed again, otherwise, it determines user identity mistake.
The checking process of above-mentioned identity can also be carried out between above-mentioned steps 105 and step 106, i.e., judging
The dynamic effective initial time of lip is stated with effective initial time of voice to be consistent, and lip reading content of text and speech text content
After whether being consistent with the checking text, the inspection of above-mentioned identity is carried out, if there is institute in user registry database
Identity is stated, then the step of performing the biological characteristic for obtaining user；Otherwise, it determines user identity mistake.
It should be noted that when being matched, User Identity that can be by the biological characteristic of acquisition only with acquisition
Corresponding biological model is matched, and obtains matching score, then judges whether the matching score is more than the threshold value of setting again,
If it is, determining that user identity is correct, user identity mistake is otherwise determined；The biological characteristic of acquisition and user can also be noted
All corresponding biological characteristic models match in volume database, obtain many candidate matches scores, then judge wherein to match again
Whether the corresponding registered user's mark of characteristic model of highest scoring is identical with the User Identity obtained；If it is, really
Determine user identity correct；Otherwise user identity mistake is determined.
It is previously noted that the biological characteristic model can be sound-groove model, and/or faceform, naturally it is also possible to be it
Its biological characteristic model, such as, and fingerprint, iris etc..
If, can be by the biological characteristic of extraction and corresponding biological characteristic model using single biological characteristic model
Matched, obtain matching score, such as：If the biological characteristic model is sound-groove model, can in above-mentioned steps 106
To obtain the vocal print feature of user, for then carrying out vocal print feature in reading aloud the voice for verifying text from the user of typing
Match somebody with somebody；If the biological characteristic model is faceform, described test can be read aloud from the user of typing in above-mentioned steps 106
Demonstrate,prove the face characteristic that user is obtained in the image of text.
If, can be in above-mentioned steps 106 respectively by the different biological features of acquisition using multi-biological characteristic model
Matched with corresponding biological characteristic model.Below exemplified by integrating using vocal print feature and face characteristic, to above two
Different matching ways are described in detail.
As shown in Figure 2 A, it is characteristic matching and a kind of flow of authentication in personal identification method of the embodiment of the present invention
Figure, comprises the following steps：
Step A1, the vocal print feature of acquisition sound-groove model corresponding with User Identity is matched, first is obtained
Score；
Step A2, the face characteristic of acquisition faceform corresponding with User Identity is matched, second is obtained
Score；
Step A3, the first score is merged with the second score, matching score is obtained.
Specific amalgamation mode can be implemented using various ways such as direct summation, weighted sum, averages to this present invention
Example is not limited.
Step A4, judges whether the matching score is more than the threshold value of setting, if it is, performing step A5；Otherwise, hold
Row step A6；
Step A5, determines that user identity is correct；
Step A6, determines user identity mistake.
As shown in Figure 2 B, it is characteristic matching and a kind of flow of authentication in personal identification method of the embodiment of the present invention
Figure, comprises the following steps：
Step B1, the vocal print feature of acquisition sound-groove model corresponding with User Identity is matched, first is obtained
Score；
Step B2, judges whether first score is more than voice print matching threshold value；If it is, performing step B3；Otherwise,
Perform step B6；
Step B3, the face characteristic of acquisition faceform corresponding with User Identity is matched, second is obtained
Score；
Step B4, judges whether second score is more than face matching threshold；If it is, performing step B5；Otherwise,
Perform step B6；
Step B5, determines that user identity is correct；
Step B6, determines user identity mistake.
It should be noted that in actual applications, can also advanced row face characteristic matching, vocal print is then carried out again special
The matching levied；The matching of vocal print feature and face characteristic can certainly be carried out simultaneously, and two kinds of matching results are then judged respectively
Corresponding condition whether is satisfied by, if be satisfied by, it is determined that user identity is correct, otherwise determines user identity mistake.
As shown in Figure 2 C, it is characteristic matching and a kind of flow of authentication in personal identification method of the embodiment of the present invention
Figure, comprises the following steps：
Step C1, the vocal print feature of acquisition is matched with each sound-groove model in user registry database, obtains many
Individual first matching score；
Step C2, faceform to be matched is determined based on the first matching score for imposing a condition is met；
First matching score imposed a condition that meets is：First matching score is more than first of first threshold
With score；Or the first matching score of the maximum preceding setting number of first matching score；
Step C3, the face characteristic of acquisition is matched with the faceform to be matched, obtains multiple second
With score；
Step C4, the first corresponding matching score of each second matching score is merged, many candidate matches are obtained
Score；
Step C5, judges that identity of the corresponding registered user's mark of matching score highest characteristic model with obtaining is
It is no identical；If it is, performing step C6；Otherwise step C7 is performed；
Step C6, determines that user identity is correct；
Step C7, determines user identity mistake.
In actual applications, equally then the matching of vocal print feature can be carried out again with the matching of advanced row face characteristic, it is right
This embodiment of the present invention is not limited.
As shown in Figure 2 D, it is characteristic matching and a kind of flow of authentication in personal identification method of the embodiment of the present invention
Figure, comprises the following steps：
Step D1, the vocal print feature of acquisition is matched with each sound-groove model in user registry database, obtains many
Individual first matching score, and the first matching score for meeting first condition is selected as the first score to be fused；
Step D2, the face characteristic of acquisition is matched with each faceform in user registry database, obtains many
Individual second matching score, and the second matching score for meeting second condition is selected as the second score to be fused；
Step D3, registered user's mark corresponding to first score to be fused is corresponding with second score to be fused
Registered user mark take union；
Step D4, identifies corresponding first matching score and the second matching score to registered user that is described and concentrating and carries out
Fusion, obtains many candidate matches scores；
Step D5, judges that identity of the corresponding registered user's mark of matching score highest characteristic model with obtaining is
It is no identical；If it is, performing step D6；Otherwise step D7 is performed；
Step D6, determines that user identity is correct；
Step D7, determines user identity mistake.
Above-mentioned first condition and second condition are respectively：Matching score is more than given threshold；Or matching score maximum
The matching score of preceding setting number.
In order to be further ensured that the security of user's registration, in an alternative embodiment of the invention, in user's registration, may be used also
It is identified with the registration behavior for user, prevents that attacker's fake user information from being registered.
As shown in figure 3, be a kind of flow chart of user's registration in the embodiment of the present invention, comprise the following steps：
Step 301, after user's registration request is received, generate and show Login script, and typing user reads aloud described
The voice and image of Login script.
Step 302, the voice and image for reading aloud the user of typing the Login script are identified, and obtain the voice
Effective initial time and the dynamic effective initial time of speech text content, lip and lip reading content of text.
Equally, the Login script can be fixed text or stochastic and dynamic text.Moreover, extract above
Verify that text can be identical with the Login script, can also be different.
Step 303, judge whether the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip
Whether language content of text and speech text content are consistent with the Login script；If it is, performing step 304；Otherwise, hold
Row step 306.
Step 304, it is determined that allowing registration.
Step 305, after registration is allowed, the biology of correspondence user's registration identity is generated according to the biological characteristic of user
Characteristic model.
Step 306, registration is forbidden.
That is, after registration is allowed, correspondence user's registration identity is generated further according to the biological characteristic of the user of acquisition
The biological characteristic model of mark.And if the dynamic effective initial time of the lip is not consistent with effective initial time of voice, or
Person's lip reading content of text and speech text content are not consistent with the Login script, then show the information errors that user provides,
In this case, the user's registration, or prompting user is forbidden to re-register.
In addition, it is necessary to which explanation, in actual applications, in above-mentioned steps 305, may not be from the voice
And/or the biological characteristic of user is extracted in image, but the biological characteristic of other forms is extracted, this embodiment of the present invention is not done
Limit.
It can be seen that, personal identification method provided in an embodiment of the present invention, when carrying out user identity identification, generation checking text
This, then typing user reads aloud the voice and image of checking text, and the voice and image are identified, the voice is obtained
Effective initial time and the dynamic effective initial time of speech text content, lip and lip reading content of text；The lip move it is effective
Initial time is consistent with effective initial time of voice and lip reading content of text and speech text content are literary with the checking
Originally in the case of being consistent, identity legitimacy checking further is carried out to user using the biological characteristic of user.Due to knowing in identity
Combine dynamic biological characteristic and static biological characteristic during not, thus efficiently avoid user profile by it is counterfeit,
The risk stolen, substantially increases the security of authentication；And the biological characteristic of user is inherent, it is not necessary to especially remember
Recall, so that improving user identity tests the convenience recognized, validity.
Correspondingly, the embodiment of the present invention also provides a kind of identification system, as shown in figure 4, being a kind of knot of the system
Structure schematic diagram.The system includes：
Registering modules 400, in user's registration, generating the biological characteristic model of correspondence user's registration identity,
And the biological characteristic model is saved in user registry database 500；
Identity acquisition module 401, for when carrying out user identity identification, obtaining User Identity；
Text processing module 402, for generating and showing checking text；
Typing module 403, the voice and image of the checking text are read aloud for typing user；
Identification module 404, the voice and image for reading aloud the checking text for the user to typing is identified, and obtains
The dynamic effective initial time of the effective initial time and speech text content, lip of the voice and lip reading content of text, and judge
Whether the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and speech text
Whether content is consistent with the checking text；
Biological characteristic extraction module 405, for the effective initial time and effective initial time of voice moved in the lip
Be consistent, and lip reading content of text and speech text content with it is described checking text be consistent when, obtain user biological characteristic；
Matching module 406, for based on the User Identity, by the life corresponding with the ID of the biological characteristic of extraction
Thing characteristic model is matched, and obtains matching score；
Judge module 407, for determining whether user identity is correct according to the matching score.
In another embodiment of present system, the system can also further comprise：：
Module (not shown) is checked, the user identity mark is whether there is in the user registry database for checking
Know；Generate if it is, triggering the text processing module 402 and show checking text；Otherwise user identity mistake is determined.
It should be noted that in embodiments of the present invention, the biological characteristic model can be sound-groove model, face mould
Type, Fingerprint Model, iris model etc., can be single biological characteristic models, and a variety of biological characteristic moulds can also be included simultaneously
Type.
Such as, if the biological characteristic model is sound-groove model, correspondingly, biological characteristic extraction module 405 can be from
The user of the typing reads aloud the vocal print feature that user is obtained in the voice of the checking text；If the biological characteristic model
For faceform, correspondingly, biological characteristic extraction module 405 can read aloud the figure of the checking text from the user of the typing
The face characteristic of user is obtained as in；If the biological characteristic model includes：Sound-groove model and faceform are correspondingly, raw
Thing characteristic extracting module 405 can include：First extraction unit, for reading aloud the checking text from the user of the typing
The vocal print feature of user is obtained in voice；Second extraction unit, for reading aloud the checking text from the user of the typing
The face characteristic of user is obtained in image.
In actual applications, the matching module 406 can be noted using various ways to the biological characteristic of acquisition with user
Biological characteristic model in volume database is matched.Such as, the biological characteristic that biological characteristic extraction module 405 can be obtained
Only with the User Identity of acquisition, corresponding biological model is matched, and obtains matching score, or biological characteristic is extracted
The biological characteristic that module 405 is obtained matches to corresponding biological characteristic model all in user registry database, obtains wait more
Select matching score.Correspondingly, the difference of the matching way specifically used according to matching module 406, judge module 407 can also have
Different judgment modes.
Below exemplified by integrating using vocal print feature and face characteristic, different matching ways and judgment mode are carried out detailed
Describe in detail bright.
As shown in Figure 5A, it is a kind of concrete structure schematic diagram of matching module in the embodiment of the present invention.
In this embodiment, the matching module includes：
First matching unit 511, the vocal print feature for first extraction unit to be obtained is obtained with the identity
The corresponding sound-groove model of identity that modulus block is obtained is matched, and obtains the first score；
Second matching unit 512, the face characteristic for second extraction unit to be obtained is obtained with the identity
The corresponding faceform of identity that modulus block is obtained is matched, and obtains the second score；
Integrated unit 513, for the first score to be merged with the second score, obtains matching score.
Correspondingly, the matching score that the matching module based on the embodiment is obtained, above-mentioned judge module 407 specifically for
When the matching score is more than the threshold value of setting, determine that user identity is correct；Otherwise user identity mistake is determined.
In actual applications, the matching module can also only include above-mentioned first matching unit 511 and second matching list
Member 512.Correspondingly, above-mentioned judge module 407 specifically judges that whether first score is more than voice print matching threshold value, and judge
Whether second score is more than face matching threshold；If it is, determining that user identity is correct；Otherwise determine that user identity is wrong
By mistake.Certainly, in actual applications, above-mentioned judge module 407 can also be sentenced after the first matching unit 511 obtains the first score
Whether first score of breaking is more than voice print matching threshold value, if it is, the second matching unit of triggering performs corresponding matching behaviour
Make；Otherwise user identity mistake is determined.
As shown in Figure 5 B, it is another concrete structure schematic diagram of matching module in the embodiment of the present invention.
In this embodiment, the matching module includes：
3rd matching unit 521, for the vocal print feature and user registry database for obtaining first extraction unit
In each sound-groove model matched, obtain multiple first matching scores；
First determining unit 522, for determining face mould to be matched based on meeting the first matching score imposed a condition
Type；First matching score imposed a condition that meets is：First matching score is matched more than the first of first threshold
Point；Or the first matching score of the maximum preceding setting number of first matching score；
4th matching unit 523, for the face characteristic of acquisition to be matched with the faceform to be matched, is obtained
To multiple second matching scores；
Score integrated unit 524, for the first corresponding matching score of each second matching score to be merged, is obtained
To many candidate matches scores.
It should be noted that in actual applications, can also by the advanced row face characteristic of the 4th matching unit 523
Match somebody with somebody, sound-groove model to be matched is then determined according to matching score by the first determining unit 521, then again by the 3rd matching unit
521 carry out the matching of vocal print feature.
Correspondingly, many candidate matches scores that the matching module based on the embodiment is obtained, above-mentioned judge module 407 is specific
For judging whether the corresponding registered user's mark of matching score highest characteristic model is identical with the identity obtained；If
It is, it is determined that user identity is correct；Otherwise user identity mistake is determined.
As shown in Figure 5 C, it is another concrete structure schematic diagram of matching module in the embodiment of the present invention.
In this embodiment, the matching module includes：
First match selection unit 531, for vocal print feature and the user's registration number for obtaining first extraction unit
Matched according to each sound-groove model in storehouse, obtain multiple first matching scores, and select to meet the first matching of first condition
Score is used as the first score to be fused；
Second match selection unit 532, for face characteristic and the user's registration number for obtaining second extraction unit
Matched according to each faceform in storehouse, obtain multiple second matching scores, and select to meet the second matching of second condition
Score is used as the second score to be fused；
Union unit 533, for first score to be fused corresponding registered user mark and described to be fused the
The corresponding registered user's mark of two scores takes union；
Union integrated unit 534, for identifying corresponding first matching score and the to registered user that is described and concentrating
Two matching scores are merged, and obtain many candidate matches scores；
Above-mentioned first condition and second condition are respectively：Matching score is more than given threshold；Or matching score maximum
The matching score of preceding setting number.
Correspondingly, many candidate matches scores that the matching module based on the embodiment is obtained, above-mentioned judge module 407 is specific
For judging whether the corresponding registered user's mark of matching score highest characteristic model is identical with the identity obtained；If
It is, it is determined that user identity is correct；Otherwise user identity mistake is determined.
The specific implementation of above-mentioned each model function can refer to the description in above the inventive method embodiment, herein not
Repeat again.
In addition, it is necessary to explanation, in this practical application, one kind that Registering modules 301 can only extract user is biological
Feature, can also extract a variety of biological characteristics, generate corresponding biological characteristic model.
In order to be further ensured that the security of user's registration, in another embodiment of present system, in user's registration,
Registration behavior that can also be by above-mentioned each module for user is identified, and prevents that attacker's fake user information from being registered.
Specifically, in user's registration, above-mentioned each module also performs following operation：
The text processing module 402 generates after the Registering modules receive user's registration request and shows registration
Text；
The typing module 403, is additionally operable to voice and image that typing user reads aloud the Login script；
The identification module 404, the voice and image for being additionally operable to read aloud the user of typing the Login script is known
Not, effective initial time and lip reading content of text that the effective initial time and speech text content, lip of the voice are moved are obtained,
And judge whether the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and language
Whether sound content of text is consistent, and notify the Registering modules to allow registration when being consistent with the checking text, otherwise
The Registering modules 301 are notified to forbid registration.
Correspondingly, the Registering modules are specifically for after the identification module 404 determines to allow registration, generation is to application
The biological characteristic model of family enrollment status mark, and the biological characteristic model is saved in user registry database.
Identification system provided in an embodiment of the present invention, in user's registration, text is verified in generation, then typing user
Read aloud checking text voice and image, the voice and image are identified, obtain the voice effective initial time and
The dynamic effective initial time of speech text content, lip and lip reading content of text；The effective initial time and voice moved in the lip
Effective initial time be consistent and lip reading content of text and speech text content with the checking text situation about being consistent
Under, identity legitimacy checking further is carried out to user using the biological characteristic of user.Due to being combined in identification procedure
Dynamic biological characteristic and static biological characteristic, therefore user profile is efficiently avoid by risk counterfeit, steal, greatly
The big security for improving authentication；And the biological characteristic of user is inherent, it is not necessary to especially memory, so as to improve
User identity tests the convenience recognized, validity.
The scheme of the embodiment of the present invention is connected a variety of checkings by way of cross validation, can be with omnibearing stereo
Portray the biological characteristic of user likes；Meanwhile, there is good defense to subsystem attack, attacker is difficult real-time, various dimensions
Ground imitates user biological feature, significantly enhances the security of system.
Each embodiment in this specification is described by the way of progressive, identical similar portion between each embodiment
Divide mutually referring to what each embodiment was stressed is the difference with other embodiment.It is real especially for system
Apply for example, because it is substantially similar to embodiment of the method, so describing fairly simple, related part is referring to embodiment of the method
Part explanation.System embodiment described above is only schematical, wherein described illustrate as separating component
Unit can be or may not be physically separate, the part shown as unit can be or may not be
Physical location, you can with positioned at a place, or can also be distributed on multiple NEs.Can be according to the actual needs
Some or all of module therein is selected to realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying
In the case of creative work, you can to understand and implement.
The embodiment of the present invention is described in detail above, embodiment used herein is carried out to the present invention
Illustrate, the explanation of above example is only intended to help to understand method and apparatus of the invention；Simultaneously for the one of this area
As technical staff, according to the present invention thought, will change in specific embodiments and applications, to sum up institute
State, this specification content should not be construed as limiting the invention.
Claims (15)
1. a kind of personal identification method, it is characterised in that including：
In user's registration, the biological characteristic model of generation correspondence user's registration identity, and by the biological characteristic model
It is saved in user registry database；
When carrying out user identity identification, User Identity is obtained；
Generate and show checking text, and typing user reads aloud the voice and image of the checking text；
The voice and image for reading aloud the user of typing the checking text are identified, obtain effective starting of the voice
Time and the dynamic effective initial time of speech text content, lip and lip reading content of text；
If the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and voice
Content of text is consistent with the checking text, then obtains the biological characteristic of user；
Based on the User Identity, the biological characteristic model in the biological characteristic and user registry database of acquisition is carried out
Matching, obtains matching score；
Determine whether user identity is correct according to the matching score.
2. according to the method described in claim 1, it is characterised in that methods described also includes：
Check and whether there is the User Identity in the user registry database；
If it is, performing the step of generating and show checking text；Otherwise, it determines user identity mistake.
3. method according to claim 2, it is characterised in that
The biological characteristic model is sound-groove model；
The biological characteristic for obtaining user includes：Obtain and use in the voice for reading aloud the checking text from the user of the typing
The vocal print feature at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, obtains matching score
Including：Vocal print feature sound-groove model corresponding with User Identity described in user registry database is matched,
Obtain matching score；
Or
The biological characteristic model is faceform；
The biological characteristic for obtaining user includes：Obtain and use in the image for reading aloud the checking text from the user of the typing
The face characteristic at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, obtains matching score
Including：Face characteristic faceform corresponding with User Identity described in user registry database is matched,
Obtain matching score；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic for obtaining user includes：Obtain and use in the voice for reading aloud the checking text from the user of the typing
The face characteristic of user is obtained in the vocal print feature at family, and the image for reading aloud the checking text from the user of the typing；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, obtains matching score
Including：
The vocal print feature of acquisition sound-groove model corresponding with User Identity described in user registry database is matched,
Obtain the first score；
The face characteristic of acquisition faceform corresponding with User Identity described in user registry database is matched,
Obtain the second score；
First score is merged with the second score, matching score is obtained.
4. method according to claim 3, it is characterised in that described that whether user identity is determined according to the matching score
Correctly include：
If the matching score is more than the threshold value of setting, it is determined that user identity is correct；Otherwise user identity mistake is determined.
5. method according to claim 2, it is characterised in that
The biological characteristic model is sound-groove model；
The biological characteristic for obtaining user includes：Obtain and use in the voice for reading aloud the checking text from the user of the typing
The vocal print feature at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, obtains matching score
Including：The vocal print feature is matched with each sound-groove model in user registry database, many candidate matching scores are obtained；
Or
The biological characteristic model is faceform；
The biological characteristic for obtaining user includes：Obtain and use in the image for reading aloud the checking text from the user of the typing
The face characteristic at family；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, obtains matching score
Including：The face characteristic is matched with each faceform in user registry database, many candidate matches scores are obtained；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic for obtaining user includes：
The vocal print feature of user is obtained in the voice for reading aloud the checking text from the user of the typing, and from the typing
User reads aloud the face characteristic that user is obtained in the image of the checking text；
The biological characteristic by acquisition is matched with the biological characteristic model in user registry database, obtains matching score
Including：
The vocal print feature and face characteristic of acquisition are carried out with the vocal print feature in user registry database and faceform respectively
Fusion matching, obtains many candidate matches scores.
6. method according to claim 5, it is characterised in that described that whether user identity is determined according to the matching score
Correctly include：
Judge the corresponding user's registration identity of matching score highest characteristic model and the User Identity obtained
It is whether identical；
If it is, determining that user identity is correct；Otherwise user identity mistake is determined.
7. the method according to any one of claim 1 to 6, it is characterised in that methods described also includes：
After user's registration request is received, generate and show Login script, and typing user reads aloud the language of the Login script
Sound and image；
The voice and image for reading aloud the user of typing the Login script are identified, obtain effective starting of the voice
Time and the dynamic effective initial time of speech text content, lip and lip reading content of text；
If the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and voice
Content of text is consistent with the Login script, then allows registration；Otherwise registration is forbidden；
After registration is allowed, then perform the step of generation corresponds to the biological characteristic model of user's registration identity.
8. method according to claim 7, it is characterised in that the checking text and/or Login script are static text
Or stochastic and dynamic text.
9. a kind of identification system, it is characterised in that including：
Registering modules, in user's registration, generating the biological characteristic model of correspondence user's registration identity, and will be described
Biological characteristic model is saved in user registry database；
Identity acquisition module, for when carrying out user identity identification, obtaining User Identity；
Text processing module, for generating and showing checking text；
Typing module, the voice and image of the checking text are read aloud for typing user；
Identification module, the voice and image for reading aloud the checking text for the user to typing is identified, and obtains institute's predicate
The dynamic effective initial time of the effective initial time and speech text content, lip of sound and lip reading content of text, and judge the lip
Whether dynamic effective initial time is consistent with effective initial time of voice, and lip reading content of text and speech text content are
It is no to be consistent with the checking text；；
Biological characteristic extraction module, effective initial time for being moved in the lip is consistent with effective initial time of voice, and
And lip reading content of text and speech text content with it is described checking text be consistent when, obtain user biological characteristic；
Matching module, for based on the User Identity, the biological characteristic that the biological characteristic extraction module is obtained with
Biological characteristic model in user registry database is matched, and obtains matching score；
Judge module, for determining whether user identity is correct according to the matching score.
10. system according to claim 9, it is characterised in that the system also includes：
Module is checked, the User Identity is whether there is in the user registry database for checking；If it is, touching
Sending out text processing module described generates and shows checking text；Otherwise user identity mistake is determined.
11. system according to claim 10, it is characterised in that
The biological characteristic model is sound-groove model；The biological characteristic extraction module is bright specifically for the user from the typing
The vocal print feature of user is obtained in the voice for reading the checking text；The biological characteristic is extracted mould by the matching judgment module
The vocal print feature sound-groove model progress corresponding with the User Identity that the identity acquisition module is obtained that block is obtained
Match somebody with somebody, obtain matching score；
Or
The biological characteristic model is faceform；The biological characteristic extraction module is bright specifically for the user from the typing
The face characteristic of user is obtained in the image for reading the checking text；The biological characteristic is extracted mould by the matching judgment module
Face characteristic faceform's progress corresponding with the User Identity that the identity acquisition module is obtained that block is obtained
Match somebody with somebody, obtain matching score；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic extraction module includes：
First extraction unit, the vocal print for obtaining user in reading aloud the voice for verifying text from the user of the typing is special
Levy；
Second extraction unit, the face for obtaining user in reading aloud the image for verifying text from the user of the typing is special
Levy；
The matching module includes：
First matching unit, vocal print feature and the identity acquisition module for first extraction unit to be obtained are obtained
The corresponding sound-groove model of identity taken is matched, and obtains the first score；
Second matching unit, face characteristic and the identity acquisition module for second extraction unit to be obtained are obtained
The corresponding faceform of identity taken is matched, and obtains the second score；
Integrated unit, for first score to be merged with second score, obtains matching score.
12. system according to claim 11, it is characterised in that
The judge module, specifically for when the matching score is more than the threshold value set, determining that user identity is correct；Otherwise
Determine user identity mistake.
13. system according to claim 10, it is characterised in that
The biological characteristic model is sound-groove model；
The biological characteristic extraction module, specifically for being obtained in reading aloud the voice for verifying text from the user of the typing
The vocal print feature of user；
The matching module, specifically for the vocal print feature is matched with each sound-groove model in user registry database,
Much candidate matching scores；
Or
The biological characteristic model is faceform；
The biological characteristic extraction module, specifically for being obtained in reading aloud the image for verifying text from the user of the typing
The face characteristic of user；
The matching module, specifically for the face characteristic is matched with each faceform in user registry database,
Obtain many candidate matches scores；
Or
The biological characteristic model includes：Sound-groove model and faceform；
The biological characteristic extraction module includes：
First extraction unit, the vocal print for obtaining user in reading aloud the voice for verifying text from the user of the typing is special
Levy；
Second extraction unit, the face for obtaining user in reading aloud the image for verifying text from the user of the typing is special
Levy；
The matching module, is obtained specifically for the vocal print feature and second extraction unit for obtaining first extraction unit
The face characteristic taken carries out merging matching with the vocal print feature in user registry database and faceform respectively, obtains many candidates
Matching score.
14. system according to claim 13, it is characterised in that
The judge module, specifically for judging the corresponding registered user's mark of matching score highest characteristic model and acquisition
Whether identity is identical；If it is, determining that user identity is correct；Otherwise user identity mistake is determined.
15. the system according to any one of claim 9 to 14, it is characterised in that
The text processing module, is additionally operable to after the Registering modules receive user's registration request, generates and show registration
Text；
The typing module, is additionally operable to voice and image that typing user reads aloud the Login script；
The identification module, the voice and image for being additionally operable to read aloud the user of typing the Login script is identified, and obtains
The dynamic effective initial time of the effective initial time and speech text content, lip of the voice and lip reading content of text, and judge
Whether the dynamic effective initial time of the lip is consistent with effective initial time of voice, and lip reading content of text and speech text
Whether content is consistent, and when being consistent, determination allows registration, otherwise forbids registration with the checking text；
The Registering modules specifically for after the identification module determines to allow registration, generation correspondence user's registration identity mark
The biological characteristic model of knowledge, and the biological characteristic model is saved in user registry database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510052189.3A CN104598796B (en) | 2015-01-30 | 2015-01-30 | Personal identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510052189.3A CN104598796B (en) | 2015-01-30 | 2015-01-30 | Personal identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104598796A CN104598796A (en) | 2015-05-06 |
| CN104598796B true CN104598796B (en) | 2017-08-25 |
Family
ID=53124574
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510052189.3A Active CN104598796B (en) | 2015-01-30 | 2015-01-30 | Personal identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104598796B (en) |
Families Citing this family (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106302339A (en) * | 2015-05-25 | 2017-01-04 | 腾讯科技（深圳）有限公司 | Login validation method and device, login method and device |
| CN104966053B (en) * | 2015-06-11 | 2018-12-28 | 腾讯科技（深圳）有限公司 | Face identification method and identifying system |
| CN106651340B (en) * | 2015-11-02 | 2021-06-29 | 创新先进技术有限公司 | Settlement method and device |
| CN105590104A (en) * | 2015-12-31 | 2016-05-18 | 联想(北京)有限公司 | Recognition method and apparatus, and electronic equipment |
| CN105554026A (en) * | 2016-01-12 | 2016-05-04 | 中北大学 | Electronic record information security management system |
| CN105718874A (en) * | 2016-01-18 | 2016-06-29 | 北京天诚盛业科技有限公司 | Method and device of in-vivo detection and authentication |
| CN105938528B (en) * | 2016-04-03 | 2019-03-12 | 长春工业大学 | A kind of control system of computer events |
| CN105975826A (en) * | 2016-04-20 | 2016-09-28 | 中国石油大学(华东) | Enterprise economic management information security system |
| CN107404381A (en) * | 2016-05-19 | 2017-11-28 | 阿里巴巴集团控股有限公司 | A kind of identity identifying method and device |
| CN105869644A (en) * | 2016-05-25 | 2016-08-17 | 百度在线网络技术（北京）有限公司 | Deep learning based voiceprint authentication method and device |
| CN106098069B (en) * | 2016-06-21 | 2019-12-20 | 佛山科学技术学院 | Identity authentication method and terminal equipment |
| CN106126914A (en) * | 2016-06-22 | 2016-11-16 | 宝鸡市中心医院 | Clinical information management system |
| CN106295299A (en) * | 2016-08-15 | 2017-01-04 | 歌尔股份有限公司 | The user registering method of a kind of intelligent robot and device |
| CN107819807A (en) * | 2016-09-14 | 2018-03-20 | 腾讯科技（深圳）有限公司 | A kind of Information Authentication method, apparatus and equipment |
| CN106407783A (en) * | 2016-09-23 | 2017-02-15 | 武汉必腾移动科技有限责任公司 | An online verification method and system |
| CN115758297A (en) * | 2017-05-25 | 2023-03-07 | 创新先进技术有限公司 | Verification method and device |
| CN108418805A (en) * | 2018-02-05 | 2018-08-17 | 浙江齐享科技有限公司 | A kind of user authentication method and system for sharing Moped Scooter |
| CN108334768A (en) * | 2018-03-19 | 2018-07-27 | 黄冈职业技术学院 | A kind of computer system user auth method |
| CN108446737B (en) * | 2018-03-21 | 2022-07-05 | 百度在线网络技术（北京）有限公司 | Method and device for identifying objects |
| CN108510364A (en) * | 2018-03-30 | 2018-09-07 | 杭州法奈昇科技有限公司 | Big data intelligent shopping guide system based on voiceprint identification |
| CN108712381A (en) * | 2018-04-16 | 2018-10-26 | 出门问问信息科技有限公司 | A kind of auth method and device |
| CN109299594B (en) * | 2018-09-04 | 2022-11-22 | 创新先进技术有限公司 | Identity verification method and device |
| CN109409204B (en) * | 2018-09-07 | 2021-08-06 | 北京市商汤科技开发有限公司 | Anti-counterfeiting detection method and device, electronic equipment and storage medium |
| CN109271915B (en) * | 2018-09-07 | 2021-10-08 | 北京市商汤科技开发有限公司 | Anti-counterfeiting detection method and device, electronic equipment and storage medium |
| CN109461437B (en) * | 2018-11-28 | 2023-05-09 | 平安科技（深圳）有限公司 | Verification content generation method and related device for lip language identification |
| CN111292734B (en) * | 2018-12-06 | 2024-03-08 | 阿里巴巴集团控股有限公司 | Voice interaction method and device |
| CN109635533B (en) * | 2018-12-07 | 2021-09-03 | 中国移动通信集团江苏有限公司 | Identity recognition method, device, equipment and medium |
| CN109801636A (en) * | 2019-01-29 | 2019-05-24 | 北京猎户星空科技有限公司 | Training method, device, electronic equipment and the storage medium of Application on Voiceprint Recognition model |
| CN111669350A (en) * | 2019-03-05 | 2020-09-15 | 阿里巴巴集团控股有限公司 | Identity verification method, verification information generation method, payment method and payment device |
| CN110766841A (en) * | 2019-06-12 | 2020-02-07 | 天津新泰基业电子股份有限公司 | User registration and verification method and device |
| CN110955874A (en) * | 2019-10-12 | 2020-04-03 | 深圳壹账通智能科技有限公司 | Identity authentication method, identity authentication device, computer equipment and storage medium |
| CN110891049A (en) * | 2019-10-12 | 2020-03-17 | 平安国际智慧城市科技股份有限公司 | Video-based account login method, device, medium and electronic equipment |
| CN110853646B (en) * | 2019-11-20 | 2023-11-03 | 深圳前海微众银行股份有限公司 | Conference speaking role distinguishing method, device, equipment and readable storage medium |
| CN110717164A (en) * | 2019-12-16 | 2020-01-21 | 国网电子商务有限公司 | Intelligent multidimensional weighting identity authentication and risk control method and system |
| CN111048113B (en) * | 2019-12-18 | 2023-07-28 | 腾讯科技（深圳）有限公司 | Sound direction positioning processing method, device, system, computer equipment and storage medium |
| CN111326175A (en) * | 2020-02-18 | 2020-06-23 | 维沃移动通信有限公司 | Prompting method for interlocutor and wearable device |
| CN113743160A (en) * | 2020-05-29 | 2021-12-03 | 北京中关村科金技术有限公司 | Method, apparatus and storage medium for biopsy |
| CN111768178A (en) * | 2020-06-30 | 2020-10-13 | 银联商务股份有限公司 | Method and related device for examining and approving business member |
| CN112329750B (en) * | 2021-01-05 | 2021-04-20 | 北京远鉴信息技术有限公司 | Digital sequence-based identity recognition method and device, electronic equipment and medium |
| CN112927688B (en) * | 2021-01-25 | 2022-05-10 | 思必驰科技股份有限公司 | Voice interaction method and system for vehicle |
| CN113313231B (en) * | 2021-04-28 | 2024-05-24 | 上海淇玥信息技术有限公司 | LipNet algorithm-based identity recognition method and system and electronic equipment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100363938C (en) * | 2005-10-31 | 2008-01-23 | 浙江大学 | Multi-model ID recognition method based on scoring difference weight compromised |
| CN102413100A (en) * | 2010-09-25 | 2012-04-11 | 盛乐信息技术（上海）有限公司 | Voice-print authentication system having voice-print password picture prompting function and realization method thereof |
| CN103973441B (en) * | 2013-01-29 | 2016-03-09 | 腾讯科技（深圳）有限公司 | Based on user authen method and the device of audio frequency and video |
| CN103475490B (en) * | 2013-09-29 | 2018-02-27 | 广州网易计算机系统有限公司 | A kind of auth method and device |
-
2015
- 2015-01-30 CN CN201510052189.3A patent/CN104598796B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104598796A (en) | 2015-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104598796B (en) | Personal identification method and system | |
| US10482230B2 (en) | Face-controlled liveness verification | |
| US10223512B2 (en) | Voice-based liveness verification | |
| EP2965249B1 (en) | Method and system for distinguishing humans from machines | |
| EP3147768A1 (en) | Screen interface unlocking method and screen interface unlocking device | |
| CN102402985A (en) | Voiceprint authentication system for improving voiceprint identification safety and method for realizing the same | |
| CN102543084A (en) | Online voiceprint recognition system and implementation method thereof | |
| CN102413100A (en) | Voice-print authentication system having voice-print password picture prompting function and realization method thereof | |
| JP2007522551A (en) | Multi-select challenge-response user authentication system and method | |
| CN107886958A (en) | A kind of express delivery cabinet pickup method and device based on vocal print | |
| WO2010047816A1 (en) | Speaker verification methods and apparatus | |
| CN103177238B (en) | Terminal and user identification method | |
| WO2014166362A1 (en) | Method, server, client and system for verifying verification codes | |
| CN103678977A (en) | Method and electronic device for protecting information security | |
| CN104158664A (en) | Identity authentication method and system | |
| CN109273009A (en) | Access control method, device, computer equipment and storage medium | |
| CN106357411A (en) | Identity verification method and device | |
| US20100095371A1 (en) | Visual authentication systems and methods | |
| KR101754954B1 (en) | Certification system and method using autograph and voice | |
| KR101972800B1 (en) | Method and apparatus for authenticating handwritten signature | |
| CN111090846B (en) | Login authentication method, login authentication device, electronic equipment and computer readable storage medium | |
| KR102604319B1 (en) | Speaker authentication system and method | |
| JP2000250861A (en) | System and method for non-contact type ic card log-in | |
| JP2012103792A (en) | Authentication device and authentication system | |
| CN105897429A (en) | Authentication method, authentication device and terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |