CN101697514B - A kind of method and system of authentication - Google Patents
A kind of method and system of authentication Download PDFInfo
- Publication number
- CN101697514B CN101697514B CN200910180016.4A CN200910180016A CN101697514B CN 101697514 B CN101697514 B CN 101697514B CN 200910180016 A CN200910180016 A CN 200910180016A CN 101697514 B CN101697514 B CN 101697514B
- Authority
- CN
- China
- Prior art keywords
- user
- acoustic model
- score
- verified
- checking
- 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
Landscapes
- Collating Specific Patterns (AREA)
Abstract
A kind of method that the invention discloses authentication, including: information of registered users, this user profile includes: acoustic model and faceform；For user to be verified, gather voice and the facial image of this user to be verified, search acoustic model and the faceform of this user to be verified, use the acoustic model found that the voice of this user to be verified is verified, obtain voice messaging checking score；Use the faceform found that the facial image of this user to be verified is verified, obtain face identification score；The most successful according to voice messaging checking score and face identification score decision verification.The present invention is while utilizing face information to carry out authentication, introduce the affirmation mechanism of voice messaging, both are effectively combined, further enhance security reliability and the generality of checking, and the weights in native system can dynamically regulate, therefore, the authentication of complex environment it is suitable for.
Description
Technical field
The present invention relates to field of information security technology, particularly relate to the method and system of a kind of authentication.
Background technology
Along with the arrival of information age, the most exactly the identity of people is verified, ensure individual's letter
The safe and reliable of breath has become as a very important problem.Traditional auth method great majority are adopted
With password, but for modern, needing the password of memory to get more and more, this has resulted in one
The most awkward problem, if being disposed as identical by various passwords, then will result in degree of safety and reduces；
If arranging the most inconvenient memory of mutually different password, easily mix up.
In recent years, it is applied in the authentication of people increasingly be closed by the biological characteristic of human body itself
Note, the method carrying out authentication especially by face information.Recognition of face has many good qualities, e.g.,
Collecting device simple cheap, be prone to transmission and be easy to utilize internet to carry out remote validation etc., relative to it
For its biometric authentication means, there is significant advantage.
But, utilize face information to carry out authentication and there is also a lot of drawback:
First, it is more sensitive to light, and light is poor when, discrimination just ratio is relatively low, more pole
The situation of end is just to cannot be carried out demonstrating in dark environment.
Secondly, face information is the most easily forged by others while being easily obtained, so in security requirement
The highest system cannot be applied, only as a kind of supplementary means.
Summary of the invention
The technical problem to be solved in the present invention is to provide the method and system of a kind of authentication, solves to utilize
When face information carries out authentication, owing to face information is prone to forge relatively low the asking of the security caused
Topic, improves the safety and reliability row utilizing face information to carry out authentication.
For solving above-mentioned technical problem, the method for a kind of authentication of the present invention, including:
Information of registered users, this user profile includes: acoustic model and faceform；
For user to be verified, gather voice and the facial image of this user to be verified, search this to be verified
The acoustic model of user and faceform, use the acoustic model the found voice to this user to be verified
Verify, obtain voice messaging checking score；Use the faceform found to this user to be verified
Facial image verify, obtain face identification score；
The most successful according to voice messaging checking score and face identification score decision verification.
Further, the registration process of acoustic model includes:
From speech data, extract acoustical characteristic parameters, use the acoustical characteristic parameters training acoustics extracted
Model, acoustic model training obtained is saved in acoustic model database；
Prompting user answers a question, and receives the problem answers that this user inputs with text mode, from acoustic mode
Type descriptive data base is searched for the acoustic model title that text answers is corresponding, obtains the acoustic model name of correspondence
After title, from acoustic model database, search acoustic model, the acoustic model found is stored user
In information database.
Further, the registration process of faceform includes:
Gather the facial image of user, from the facial image gathered, extract face characteristic parameter, use
Face characteristic parameter training goes out faceform, this faceform is stored in User Information Database.
Further, the most successful according to voice messaging checking score and face identification score decision verification
Process includes: voice messaging is verified score and face identification score unification to same dimension, to this two
Individual score is weighted summation, and compares with system thresholds, if greater than system thresholds, then for closing
Method user；If no more than system thresholds, then it is disabled user.
Further, these two scores are weighted summation method particularly includes:
ω1*Scoreface+(1-ω1)*Scoreviv, wherein, ScorefaceObtain for dimension recognition of face after reunification
Point, ω1For weights, ScorevivFor dimension voice messaging checking score after reunification.
Further, ω1Determination method be: system is carried out repeatedly user checking, obtain multiple language
Sound Information Authentication score and face identification score, calculate the mean value s of this two classes score respectively1And s2, then
Further, the determination method of system thresholds is: system is carried out multiple authentication, when verifying every time,
Configure an empirical value, after checking terminates, calculate system etc. error rate, will each medium error rate of checking
Empirical value time low is as system thresholds.
Further, in registration process, if prompting user answers multiple problem, then answer for this user
The text answers of the plurality of problem sets up acoustic model respectively；
The voice of the user to be verified gathered is the voice that this user to be verified answers a question, if requiring to be tested
Card user answers multiple problem, then the language that user is answered a question by the acoustic model using each problem corresponding
Sound is verified, voice messaging verifies to be divided into that when this user to be verified answers a question for each time, voice messaging is tested
The mean value of card score.
Further, the system of a kind of authentication, including: user profile Registering modules, voice messaging
Authentication module, face recognition module and identity validation module, wherein:
User profile Registering modules, for information of registered users, this user profile includes: acoustic model and
Faceform；
Voice messaging authentication module, for gathering the voice of user to be verified, searches this user's to be verified
Acoustic model, uses the acoustic model found that the voice of this user to be verified is entered checking, obtains voice
Information Authentication score；
Face recognition module, for gathering the facial image of user to be verified, searches this user's to be verified
Faceform, uses the faceform found to verify the facial image of this user to be verified,
To face identification score；
Identity validation module, for according to voice messaging checking score and face identification score decision verification being
No success.
Further, this system also includes: characteristic extracting module and MBM, wherein:
Characteristic extracting module, for extracting acoustical characteristic parameters, the acoustics that will extract from speech data
Characteristic parameter is sent to MBM；
MBM, for using the acoustical characteristic parameters training acoustic model received, obtains training
Acoustic model be saved in acoustic model database；
During acoustic model in user profile Registering modules information of registered users, prompting user answers a question,
Receive the problem answers that this user inputs with text mode, from acoustic model descriptive data base, search for text
The acoustic model title that answer is corresponding, after obtaining the acoustic model title of correspondence, from acoustic model database
Middle lookup acoustic model, stores the acoustic model found in User Information Database.
Further, characteristic extracting module, it is additionally operable in user profile Registering modules information of registered users
Faceform time, from user profile Registering modules gather facial image extract face characteristic parameter,
The face characteristic parameter extracted is sent to MBM；
MBM, the face characteristic parameter training being also used for receiving goes out faceform, by this people
Face model stores in User Information Database.
Further, identity validation module is tested according to voice messaging checking score and face identification score decision-making
Card the most successfully process includes:
Voice messaging is verified score and face identification score unification to same dimension, to these two scores
It is weighted summation, and compares with system thresholds, if greater than system thresholds, be then validated user；
If no more than system thresholds, then it is disabled user.
In sum, the present invention, while utilizing face information to carry out authentication, introduces voice messaging
Affirmation mechanism, both are effectively combined, further enhance the security reliability of checking and universal
Property, and the weights in native system can dynamically regulate, and therefore, the identity being suitable for complex environment is tested
Card.
Accompanying drawing explanation
Fig. 1 is the flow chart of the method for authentication of the present invention；
Fig. 2 is user profile registration and the flow chart of model training in the present invention；
Fig. 3 is the flow chart of the authentication of user of the present invention；
Fig. 4 is the flow chart of decision verification of the present invention；
Fig. 5 is the Organization Chart of the system of authentication of the present invention.
Detailed description of the invention
In the embodiment of the present invention, voice messaging is confirmed, and the identification with face information carries out fusion and realizes identity
Checking, to improve the safety and reliability of authentication.
The invention mainly comprises five major parts: feature extraction, acoustic model modeling, user profile
Registration, voice and the confirmation of facial image and checking decision-making.
Below in conjunction with accompanying drawing, the detailed description of the invention of the present invention is illustrated.
101: from speech data, extract acoustical characteristic parameters, train acoustic model；
The extraction of characteristic parameter is the first step of model training and authentication, in the present embodiment, special
Levy parameter and include acoustical characteristic parameters and face characteristic parameter.
Present embodiment can use Mel fall general coefficient (MFCC) as acoustical characteristic parameters, permissible
Use the Mel coefficient on 13 rank and the general coefficient on 2 rank, the characteristic vector of totally 39 dimensions.
Present embodiment can use hidden Markov model (HMM) training algorithm, use acoustic feature
Parameter training acoustic model.
The training need of model is based on substantial amounts of speech data, in the present embodiment, can use Hub4
Chinese broadcast voice data storehouse provides speech data, as in figure 2 it is shown, the language that this speech database is provided
Sound data extract MFCC characteristic vector, then use HMM training algorithm training acoustic model (step
Rapid 201～step 202).
102: the acoustic model trained is stored in acoustic model database；
103: the registration of user profile；
The registration of user profile includes the registration of privacy concern answer and facial image, and these two parts data can
To gather simultaneously.
In present embodiment, privacy concern can show on screen in the way of word, installs on equipment
Having capture apparatus, in order to ensure the correctness of answer, user answers privacy concern by text mode, uses
During the input privacy concern answer of family, system can gather facial image simultaneously, and user profile can include 12
Individual privacy concern and 30 secondary facial images.
As in figure 2 it is shown, the privacy concern answer inputted with text mode for user, system is from acoustic mode
Type descriptive data base is searched for acoustic model title (step 203) that text answers is corresponding, is obtaining literary composition
After the acoustic model title that this answer is corresponding, from acoustic model database, find the acoustic model of correspondence
(step 204), stores in User Information Database, and uses ID to be identified.System
Acoustic model is set up in each privacy concern answer answered for user respectively.
In acoustic model database, the acoustic model of storage is the modeling to syllable, and acoustic model describes number
According in storehouse storage be the corresponding informance of Chinese character and syllable.
For the facial image obtained, system can extract characteristic parameter (step 205) from facial image,
Use face characteristic parameter training to go out faceform's (step 206), faceform is stored user's letter
In breath database, and ID is used to be identified.In present embodiment, use the local two of 3*3
Meta schema (LBP) Algorithm for Training faceform.
Acoustic model and the faceform of user collectively form user profile.
It is of course also possible to dynamically adjust acoustic model and faceform, model is made to constantly update.
104: the authentication of user, including: voice confirms and facial image confirms；
As it is shown on figure 3, the authentication process itself of user specifically includes:
301: user's typing identity ID, system extracts three randomly from 12 privacy concerns, aobvious
Show and answer to user；
302: user answers a question in the way of voice, the language of the privacy concern answer of system acquisition user
Sound；
303: extract the acoustical characteristic parameters of the voice of the privacy concern answer of this user；
304: system uses ID to search the corresponding privacy concern of this user from User Information Database
The acoustic model of answer；
305: use the sound of the voice of the corresponding acoustic model of this user privacy concern answer to extracting
Learn characteristic parameter to verify, obtain voice messaging checking score；
When proposing multiple privacy concern to user, voice messaging verifies to be divided into the average of three scores
Value.
306: sampling facial image；
307: system uses ID to search the faceform of this user from User Information Database；
308: extract the face characteristic parameter of this user's facial image；
309: use the face characteristic parameter of the faceform of this user facial image to extracting to carry out
Checking, obtains face identification score；
The method of checking is essentially identical, above User Information Database has saved the model of user,
Its real mould is exactly mathematic expectaion and the covariance of statistic, and it is corresponding that test volume is actually obtained in checking
The probability of model.
Above-mentioned voice messaging checking score and face identification score are referred to as mating mark.
105: decision verification is the most successful；
As shown in Figure 4, voice messaging checking score and face identification score are carried out regular by system respectively,
They are unified in same dimension, then score is weighted summation, compares with system thresholds
Relatively.
Ordinary circumstance, system can repeatedly be tested at test phase, thus can obtain multiple voice letter
Breath checking score and face identification score, obtain for this two class, calculate repeatedly the mathematics of score respectively
Expecting and variance, regular method is exactly when checking, deducts the mathematics of test phase by the score obtained
Expect, then divided by variance, thus by regular for score in the range of 0～1.
Concrete decision-making mechanism is as follows:
Wherein, ScorefaceBe face identification score regular after mark, ω1It is weights, ScorevivIt is
Mark after voice messaging checking score is regular, σ is the threshold value of system.
The determination method of threshold value σ is as follows:
Some people are verified by system, to select σ value.Identifying object can be divided into men and women two groups, number
Measuring identical, it is assumed that be all 10 people, then these 20 people carry out cross validation to system, and user is in certification
Oneself when be system user, and be all to emit the person of recognizing in the case of other.
System has two kinds of mistakes validation failure when: False Rejects and mistake accept.
False Rejects mean that a user is that the user in system is but rejected.
What mistake accepted mean that user is not the user in system and be accepted.
When false rejection rate (FRE) is equal with false acceptance rate (FAE), it is referred to as waiting mistake.Etc. error rate (EER)
Being the standard evaluating an identity authorization system, its value is the least, and the system recognition performance of representative is the best, so
The error rate such as the selection of threshold value should make are minimum.
Threshold calculations is as follows: the starting stage can provide an empirical value, the most constantly revises this
Empirical value.After system is verified every time, all calculate system etc. error rate, the threshold value that the error rate such as selection is minimum,
This threshold value is exactly the threshold value of requirement.
Weights ω1Determination way as follows: during above by several subscriber authentication systems, system can obtain
To several voice messagings checking score and face identification score, this two classes score is saved, respectively
Seek the mean value of two class scores, it is assumed that mean value is respectively s1And s2, then weights are:
As it is shown in figure 5, present invention also offers the system of a kind of authentication, including: user profile is noted
Volume module, characteristic extracting module, MBM, voice messaging authentication module, face recognition module and body
Part confirms module, wherein:
User profile Registering modules, for registering the answer of privacy concern, privacy concern can be as the birthday,
Matronymic word and a favorite book etc., the privacy concern that reception user inputs with text mode is answered
Case, gathers facial image, the text of privacy concern answer is searched for from acoustic model descriptive data base literary composition
The acoustic model title that this answer is corresponding, after obtaining acoustic model title, from acoustic model database
Find the acoustic model of correspondence, store in User Information Database；For the facial image gathered,
Call characteristic extracting module from facial image, extract characteristic parameter, call MBM and use face characteristic
Parameter training goes out faceform, and the faceform trained by MBM stores User Information Database
In.
Characteristic extracting module, for extracting acoustical characteristic parameters, the acoustics that will extract from speech data
Characteristic parameter is sent to MBM, and acoustical characteristic parameters can use Mel to fall general coefficient, wherein has one
The energy coefficient of dimension；From facial image, also extract face characteristic parameter, the face characteristic ginseng that will extract
Number is sent to MBM, and face characteristic parameter can use local binary feature parameter；Believe according to voice
Breath authentication module and the calling of face recognition module, extract acoustics from the language received and facial image
Characteristic parameter and face characteristic parameter, and return.
MBM, for using the acoustical characteristic parameters training acoustic model received, obtains training
Acoustic model be saved in acoustic model database；The face characteristic parameter training received also is used to go out
Faceform, stores in User Information Database by this faceform, wherein, and the modeling of acoustic model
Hidden Markov model, face modeling is used to use local binary.
Voice messaging authentication module, extracts three privacy concerns from the privacy concern of registration for randomly
Being shown to user to be verified, user to be verified uses the mode of voice to answer a question, voice messaging checking mould
Block gathers the voice of privacy of user problem answers to be verified, calls characteristic extracting module and extracts acoustic feature ginseng
Number, searches the acoustic model of this user to be verified, uses the acoustic model found to this user to be verified
The acoustical characteristic parameters of voice of privacy concern answer verify, obtain voice messaging checking score.
Face recognition module, for gathering the facial image of user to be verified, calls characteristic extracting module and carries
Take face characteristic parameter, search the faceform of this user to be verified, use the faceform couple found
The face characteristic parameter of the facial image of this user to be verified is verified, obtains face identification score.
Identity validation module, unified to same amount for voice messaging being verified score and face identification score
On guiding principle, these two scores are weighted summation, and compare with system thresholds, if greater than system
Threshold value, then be validated user；If no more than system thresholds, then it is disabled user.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for this
For the technical staff in field, the present invention can have various modifications and variations.All spirit in the present invention and
Within principle, any modification, equivalent substitution and improvement etc. made, should be included in the protection of the present invention
Within the scope of.
Claims (10)
1. a method for authentication, including:
Information of registered users, this user profile includes: acoustic model and faceform；
For user to be verified, gather voice and the facial image of this user to be verified, search this to be verified
The acoustic model of user and faceform, use the acoustic model the found voice to this user to be verified
Verify, obtain voice messaging checking score；Use the faceform found to this user to be verified
Facial image verify, obtain face identification score；
The most successful according to described voice messaging checking score and face identification score decision verification；
The registration process of described acoustic model includes:
From speech data, extract acoustical characteristic parameters, use the acoustical characteristic parameters training acoustics extracted
Model, acoustic model training obtained is saved in acoustic model database；
Prompting user answers a question, and receives the problem answers that this user inputs with text mode, from acoustic mode
Type descriptive data base is searched for the acoustic model title that text answers is corresponding, obtains the acoustic model name of correspondence
After title, from described acoustic model database, search acoustic model, the acoustic model found is stored
In User Information Database；
In registration process, if prompting user answers multiple problem, the then the plurality of problem answered for this user
Text answers set up acoustic model respectively.
2. the method for claim 1, it is characterised in that:
The registration process of described faceform includes:
Gather the facial image of user, from the facial image gathered, extract face characteristic parameter, use
Face characteristic parameter training goes out faceform, this faceform is stored in User Information Database.
3. the method for claim 1, its feature etc. is:
Described the most successful according to described voice messaging checking score and face identification score decision verification
Process includes: by described voice messaging checking score and face identification score unification in same dimension, right
These two scores are weighted summation, and compare with system thresholds, if greater than system thresholds, then
For validated user；If no more than system thresholds, then it is disabled user.
4. method as claimed in claim 3, it is characterised in that:
Described these two scores are weighted summation method particularly includes:
ω1*Scoreface+(1-ω1)*Scoreviv, wherein, ScorefaceObtain for dimension recognition of face after reunification
Point, ω1For weights, ScorevivFor dimension voice messaging checking score after reunification.
5. method as claimed in claim 4, it is characterised in that:
Described ω1Determination method be: system is carried out repeatedly user checking, obtain multiple voice messaging
Checking score and face identification score, calculate the mean value s of this two classes score respectively1And s2, then
6. method as claimed in claim 3, it is characterised in that
The determination method of described system thresholds is: system is carried out multiple authentication, when verifying every time, and configuration
One empirical value, checking terminate after, calculate system etc. error rate, will each time checking medium error rate minimum time
Empirical value as described system thresholds.
7. method as claimed in claim 3, it is characterised in that:
The voice of the user described to be verified gathered is the voice that this user to be verified answers a question, if requirement
User to be verified answers multiple problem, then user is answered a question by the acoustic model using each problem corresponding
Voice verify, described voice messaging verifies to be divided into language when this user to be verified answers a question for each time
The mean value of sound Information Authentication score.
8. a system for authentication, including: user profile Registering modules, voice messaging checking mould
Block, face recognition module and identity validation module, wherein:
Described user profile Registering modules, for information of registered users, this user profile includes: acoustic mode
Type and faceform；
Described voice messaging authentication module, for gathering the voice of user to be verified, searches this use to be verified
The acoustic model at family, uses the acoustic model found that the voice of this user to be verified is entered checking, obtains
Voice messaging checking score；
Described face recognition module, for gathering the facial image of user to be verified, searches this use to be verified
The faceform at family, uses the faceform found to test the facial image of this user to be verified
Card, obtains face identification score；
Described identity validation module, for obtaining according to described voice messaging checking score and described recognition of face
Divide decision verification the most successful；
This system also includes: characteristic extracting module and MBM, wherein:
Described characteristic extracting module, for extracting acoustical characteristic parameters, by extract from speech data
Acoustical characteristic parameters is sent to described MBM；
Described MBM, for using the acoustical characteristic parameters training acoustic model received, will training
The acoustic model obtained is saved in acoustic model database；
During acoustic model in described user profile Registering modules information of registered users, prompting user answers and asks
Topic, receives the problem answers that this user inputs with text mode, searches for from acoustic model descriptive data base
The acoustic model title that text answers is corresponding, after obtaining the acoustic model title of correspondence, from described acoustic mode
Type database is searched acoustic model, the acoustic model found is stored in User Information Database；
In registration process, if prompting user answers multiple problem, the then the plurality of problem answered for this user
Text answers set up acoustic model respectively.
9. system as claimed in claim 8, it is characterised in that:
Described characteristic extracting module, is additionally operable in described user profile Registering modules information of registered users
During faceform, from the facial image that user profile Registering modules gathers, extract face characteristic parameter, will
The face characteristic parameter extracted is sent to described MBM；
Described MBM, the face characteristic parameter training being also used for receiving goes out faceform, will
This faceform stores in User Information Database.
10. system as claimed in claim 9, it is characterised in that: described identity validation module is according to institute
State voice messaging checking score and described face identification score decision verification the most successfully process include:
By described voice messaging checking score and face identification score unification in same dimension, to these two
Score is weighted summation, and compares with system thresholds, if greater than system thresholds, is then legal
User；If no more than system thresholds, then it is disabled user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910180016.4A CN101697514B (en) | 2009-10-22 | 2009-10-22 | A kind of method and system of authentication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910180016.4A CN101697514B (en) | 2009-10-22 | 2009-10-22 | A kind of method and system of authentication |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101697514A CN101697514A (en) | 2010-04-21 |
| CN101697514B true CN101697514B (en) | 2016-08-24 |
Family
ID=42142590
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910180016.4A Active CN101697514B (en) | 2009-10-22 | 2009-10-22 | A kind of method and system of authentication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101697514B (en) |
Families Citing this family (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103037368A (en) * | 2011-09-29 | 2013-04-10 | 中国移动通信集团四川有限公司 | Method, device and system for identity authentication |
| CN103297590B (en) * | 2012-02-24 | 2016-12-14 | 中兴通讯股份有限公司 | A kind of method and system realizing equipment unblock based on audio frequency |
| CN103310339A (en) * | 2012-03-15 | 2013-09-18 | 凹凸电子（武汉）有限公司 | Identity recognition device and method as well as payment system and method |
| CN103902865B (en) * | 2012-12-25 | 2017-06-27 | 联想(北京)有限公司 | The method and electronic equipment of a kind of information processing |
| CA2907118C (en) | 2013-03-15 | 2021-02-16 | United States Postal Service | System and method of identity verification |
| CN103440686A (en) * | 2013-07-29 | 2013-12-11 | 上海交通大学 | Mobile authentication system and method based on voiceprint recognition, face recognition and location service |
| US9343068B2 (en) * | 2013-09-16 | 2016-05-17 | Qualcomm Incorporated | Method and apparatus for controlling access to applications having different security levels |
| 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 |
| CN103955637A (en) * | 2014-04-09 | 2014-07-30 | 可牛网络技术（北京）有限公司 | Identification method and device for user identity of mobile terminal |
| CN104967593B (en) * | 2014-10-15 | 2018-04-20 | 腾讯科技（深圳）有限公司 | A kind of auth method, device and system |
| US20160191512A1 (en) * | 2014-12-27 | 2016-06-30 | Mcafee, Inc. | Predictive user authentication |
| CN105049421A (en) * | 2015-06-24 | 2015-11-11 | 百度在线网络技术（北京）有限公司 | Authentication method based on use behavior characteristic of user, server, terminal, and system |
| WO2017004090A1 (en) | 2015-06-30 | 2017-01-05 | United States Postal Service | System and method of providing identity verificaiton services |
| CN106453209B (en) * | 2015-08-07 | 2020-01-21 | 阿里巴巴集团控股有限公司 | Identity verification method and device |
| CN105429969B (en) * | 2015-11-09 | 2019-08-13 | 普天智能照明研究院有限公司 | A kind of user ID authentication method and equipment |
| CN105426723A (en) * | 2015-11-20 | 2016-03-23 | 北京得意音通技术有限责任公司 | Voiceprint identification, face identification and synchronous in-vivo detection-based identity authentication method and system |
| CN105897708A (en) * | 2016-03-31 | 2016-08-24 | 宇龙计算机通信科技(深圳)有限公司 | Information protection method and mobile terminal |
| CN107241309A (en) * | 2017-05-04 | 2017-10-10 | 上海斐讯数据通信技术有限公司 | A kind of identity identifying method, authentication device and Verification System |
| CN108959874A (en) * | 2017-05-17 | 2018-12-07 | 武汉默联股份有限公司 | Information Authentication method and device |
| CN115758297A (en) * | 2017-05-25 | 2023-03-07 | 创新先进技术有限公司 | Verification method and device |
| CN107680602A (en) * | 2017-08-24 | 2018-02-09 | 平安科技（深圳）有限公司 | Voice fraud recognition methods, device, terminal device and storage medium |
| CN108263337A (en) * | 2018-01-19 | 2018-07-10 | 杭州左中右网络科技有限公司 | Self-service car lending system password matching method based on bluetooth communication |
| CN108391141B (en) * | 2018-03-19 | 2020-03-31 | 京东数字科技控股有限公司 | Method and apparatus for outputting information |
| CN110362973A (en) * | 2018-03-26 | 2019-10-22 | 中国移动通信有限公司研究院 | A kind of generation method and electronic equipment of safety problem |
| 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 |
| CN109543516A (en) * | 2018-10-16 | 2019-03-29 | 深圳壹账通智能科技有限公司 | Signing intention judgment method, device, computer equipment and storage medium |
| CN109614777B (en) * | 2018-11-23 | 2020-09-11 | 第四范式（北京）技术有限公司 | Intelligent device and user identity authentication method and device of intelligent device |
| CN111583906B (en) * | 2019-02-18 | 2023-08-15 | 中国移动通信有限公司研究院 | Role recognition method, device and terminal for voice session |
| CN109934980A (en) * | 2019-03-24 | 2019-06-25 | 哈尔滨理工大学 | A kind of identification system based on face and vocal print combined authentication |
| US11790471B2 (en) | 2019-09-06 | 2023-10-17 | United States Postal Service | System and method of providing identity verification services |
| 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 |
| CN111159444A (en) * | 2019-12-31 | 2020-05-15 | 上海能塔智能科技有限公司 | Image evidence acquisition method and device and verification method and device using vehicle |
| CN112634861A (en) * | 2020-12-30 | 2021-04-09 | 北京大米科技有限公司 | Data processing method and device, electronic equipment and readable storage medium |
| CN112989299A (en) * | 2021-03-11 | 2021-06-18 | 恒睿(重庆)人工智能技术研究院有限公司 | Interactive identity recognition method, system, device and medium |
| CN113032758B (en) * | 2021-03-26 | 2023-06-16 | 平安银行股份有限公司 | Identification method, device, equipment and storage medium for video question-answering flow |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216884A (en) * | 2007-12-29 | 2008-07-09 | 北京中星微电子有限公司 | A method and system for face authentication |
| CN101256700A (en) * | 2008-03-31 | 2008-09-03 | 浙江大学城市学院 | ATM of mixed type user identification authentication |
-
2009
- 2009-10-22 CN CN200910180016.4A patent/CN101697514B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216884A (en) * | 2007-12-29 | 2008-07-09 | 北京中星微电子有限公司 | A method and system for face authentication |
| CN101256700A (en) * | 2008-03-31 | 2008-09-03 | 浙江大学城市学院 | ATM of mixed type user identification authentication |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101697514A (en) | 2010-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101697514B (en) | A kind of method and system of authentication | |
| CN110782921B (en) | Voice evaluation method and device, storage medium and electronic device | |
| US8595007B2 (en) | Voice print recognition software system for voice identification and matching | |
| US10013972B2 (en) | System and method for identifying speakers | |
| CN106782572A (en) | The authentication method and system of speech cipher | |
| Das et al. | Development of multi-level speech based person authentication system | |
| Singh et al. | Applications of speaker recognition | |
| CN101467204B (en) | Method and system for bio-metric voice print authentication | |
| Soewito et al. | Smart mobile attendance system using voice recognition and fingerprint on smartphone | |
| CN107395352A (en) | Personal identification method and device based on vocal print | |
| 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 | |
| CN102223367B (en) | Method, device and system for accessing website of mobile subscriber | |
| CN104104664A (en) | Method, server, client and system for verifying verification code | |
| Prasad et al. | Intelligent chatbot for lab security and automation | |
| CN102413101A (en) | Voice-print authentication system having voice-print password voice prompting function and realization method thereof | |
| CN107346568A (en) | The authentication method and device of a kind of gate control system | |
| Beigi | Challenges of LargeScale Speaker Recognition | |
| CN110379433A (en) | Method, apparatus, computer equipment and the storage medium of authentication | |
| Czyżewski et al. | Analysis of results of large‐scale multimodal biometric identity verification experiment | |
| US20040075590A1 (en) | E-learning biometric identification information system | |
| CN103390406B (en) | Speaker validation method, the preparation method of speaker verification and electronic installation | |
| US10971148B2 (en) | Information providing device, information providing method, and recording medium for presenting words extracted from different word groups | |
| US10957318B2 (en) | Dynamic voice authentication | |
| CN114971658B (en) | Anti-fraud propaganda method, system, electronic equipment and storage medium |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201127 Address after: 215500 No.13, Caotang Road, Changshu, Suzhou, Jiangsu Province Patentee after: Changshu intellectual property operation center Co.,Ltd. Address before: 518057 Nanshan District Guangdong high tech Industrial Park, South Road, science and technology, ZTE building, Ministry of Justice Patentee before: ZTE Corp. |
|
| TR01 | Transfer of patent right | ||
| CP02 | Change in the address of a patent holder |
Address after: 215500 5th floor, building 4, 68 Lianfeng Road, Changfu street, Changshu City, Suzhou City, Jiangsu Province Patentee after: Changshu intellectual property operation center Co.,Ltd. Address before: No.13 caodang Road, Changshu City, Suzhou City, Jiangsu Province Patentee before: Changshu intellectual property operation center Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220726 Address after: 2105, building a, Yiyuan Century Plaza, No. 19, Haiyu North Road, Changshu, Suzhou, Jiangsu 215500 Patentee after: Changshu Jieba Property Management Co.,Ltd. Address before: 215500 5th floor, building 4, 68 Lianfeng Road, Changfu street, Changshu City, Suzhou City, Jiangsu Province Patentee before: Changshu intellectual property operation center Co.,Ltd. |
|
| TR01 | Transfer of patent right |