CN105224850A - Combined right-discriminating method and intelligent interactive system - Google Patents
Combined right-discriminating method and intelligent interactive system Download PDFInfo
- Publication number
- CN105224850A CN105224850A CN201510698472.3A CN201510698472A CN105224850A CN 105224850 A CN105224850 A CN 105224850A CN 201510698472 A CN201510698472 A CN 201510698472A CN 105224850 A CN105224850 A CN 105224850A
- Authority
- CN
- China
- Prior art keywords
- recognition
- authentication
- correspondence
- ascore
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
Abstract
The invention discloses a kind of combined right-discriminating method, and comprise the following steps: step 1-1-1: choose any one recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore; Step 1-2-1: choose any another kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore; Step 2-1-1: define weight respectively to each recognition method in above step, obtains the weights △ A relevant to its direct proportion according to characteristic of correspondence information matches value Ascore simultaneously respectively; Step 3-1-1: compute weighted to the recognition result of the different recognition method in above step, when its summed result is greater than preset value, then authentication is passed through.Under ensureing the prerequisite of security, substantially increase serious forgiveness, the impalpable phenomenon occurred when avoiding authentication.
Description
Technical field
The present invention relates to a kind of method for authenticating, more particularly, it relates to a kind of combined right-discriminating method and intelligent interactive system.
Background technology
Identity recognizing technology is derivative important technology with electronic product development, be widely used in rights management, Item Management, multiple field such as attendance management and interactive maintenance, at present, , comparatively proven technique has fingerprint identification technology, face recognition technology and sound groove recognition technology in e etc., owing to identifying according to the unique features of different verifier, thus realize the technique effect of accurate authentication, no matter which kind of technical scheme above-mentioned, all need to set up the corresponding data (fingerprint image data with authorized person's feature in advance, vocal print coded data, face image data etc.), then by the data prestored and the collected data of user's single authentication are compared, the unique point quantity identical according to both calculates corresponding score value, if score value is greater than preset value, just complete authentication, enter subsequent operation, otherwise, then failed authentication.
But, in actual authentication process, if it is too high that preset value is arranged, so being bound to security will be higher, because this feature is difficult to imitated and replaces, but so, due to gather error, body state difference all easily cause the unique point lazy weight that can overlap, cause corresponding score to be less than preset value, authentication has been difficult to, user is brought inconvenience; If preset value setting is too low, so this feature just easily imitated and replacement, when other users carry out authentication, is also difficult to play one and preferably distinguishes effect, make to authorize threshold to reduce, impact security.
Enlighten in conjunction with two kinds of authentication modes and above technology although also exist, but be only when a kind of authentication mode cannot complete, realize authentication by another kind of authentication mode, although authentication mode so equally provides convenience, be equally also difficult to make a breakthrough in security.
Summary of the invention
For the deficiency that prior art exists, the first object of the present invention is a kind of combined right-discriminating method providing security and serious forgiveness all higher;
The second object of the present invention is the intelligent interactive system providing a kind of security and serious forgiveness all higher.
For achieving the above object, the invention provides following technical scheme:
For realizing the first object of the present invention, a kind of combined right-discriminating method, and comprise the following steps:
Step 1-1-1: choose any one recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore;
Step 1-2-1: choose any another kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore;
Step 2-1-1: define weight respectively to each recognition method in above step, obtains the weights △ A relevant to its direct proportion according to characteristic of correspondence information matches value Ascore simultaneously respectively;
Step 3-1-1: compute weighted to the recognition result of the different recognition method in above step, when its summed result is greater than preset value, then authentication is passed through.
First, acquisition of information is carried out by step 1-1-1, user can by present comparatively ripe and in the recognition method that resolution is higher one, identify, and according to the registration of unique point, obtain characteristic information matching value, same, exactly acquisition of information is carried out to another recognition method by step 1-2-1, according to these at least two kinds of recognition method, obtain different characteristic information matching values, because the characteristic information matching value difference in different modes is larger, so characteristic information matching value is changed into corresponding weights by one, like this in each recognition method, the obtainable score value of each unique point coincidence converts, and then it is assigned weight, by the ranking operation disclosed in step 3-1-1, obtain end product, certainly, the size of preset value can be set according to actual conditions, to meet authentication condition, so, by at least two kinds of authentication modes, authentication is carried out to it, and obtain last score in the mode of weighting, so just effectively can evade some authentications caused because of environmental factor or physical factors cannot successful situation, simultaneously, the method of at least two kinds of authentications, be difficult to be obtained feature by other people, so security have also been obtained guarantee, its principle is as follows:
Such setting, if single mode authentication in existing, in order to meet security, it is relatively high that the threshold value passed through arranges by needs, the coincidence such as completing 800 unique points just just can complete authentication, but by design of the present invention, wherein there is the coincidence of 600 unique points in a kind of authentication mode single, and there is the coincidence of 500 unique points in another kind of authentication mode single, convert the characteristic information matching value that two kinds of modes obtain to weights respectively, now corresponding weight value is multiplied by corresponding weight and still can be greater than preset value, just authentication can be passed through, so, both compare, security can be guaranteed equally, although the threshold of 500 unique points and 600 unique point coincidence needs is lower, if but in three kinds of authentication modes two kinds have the unique point of quantity like this to overlap, user also can be determined very accurately, like this, under the guaranteed prerequisite of security, complete authentication and just become a relatively easy operation, because the change of user's state, change of environment for use etc., all can not affect but the authentication success ratio of a mode.
Further, described preset value is proportional to the value of weight maximum in selected recognition method.By such setting, using maximum weighted value as preset value, judge using a kind of recognition method that weight is the highest as benchmark, such reliability is higher.
Further, also comprise step 1-1-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-1-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-2-1;
Step 1-1-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-1-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-2-1, step 1-1-2 and step 1-1-3 order can be exchanged.
By such setting, when an authentication, corresponding upper and lower bound threshold value is set, if exceed this scope, so just judge to authorize or refusal mandate, all can not enter next link again, certainly, this upper limit threshold and lower threshold compared to existing technology in threshold value should have difference, upper limit threshold can arrange relatively high, when single mode authentication uniquely can determine user, and just can be directly same by authentication, when single mode authentication can get rid of the user of all correspondences, next step authentication can not be entered so equally.
Further, in step 2-1-1, △ A=(Ascore-Amin)/(Amax-Amin).By such setting, number of parameters is reduced, the adjustment that the size of Amax and Amin is carried out its precision can be regulated, one can be played when mating simultaneously and can play preferably authentication effect.
Further, in step 2-1-1, the weight ratio of fingerprint recognition, recognition of face and Application on Voiceprint Recognition is respectively 4:3.5:2.5.These 3 kinds of information are everyone distinctive information.But according to recognition of face, fingerprint recognition and the know-why of Application on Voiceprint Recognition and the scene of reality, although as different in everyone voiceprint, but the extraction of voiceprint is easily subject to the interference of outside noise, the change of own bodies simultaneously also can affect the effect of Application on Voiceprint Recognition, so the weight of recognition method in weight authentication arithmetic is not identical in 3.The level of security of fingerprint recognition is better than recognition of face, and recognition of face is better than Application on Voiceprint Recognition.Weight refers to the weight relationship between the weight relationship between every two kinds of algorithms, 3 kinds of algorithms.As when system uses fingerprint recognition and recognition of face, the weight x% of fingerprint recognition matching value, the weight of recognition of face matching value is y%, x%>y%.This is the algorithmic characteristic of Integrated comparative fingerprint recognition and recognition of face, and recognition of face is for fingerprint recognition, and easily by external environmental interference, as ambient light intensity, so weighted value reduces relatively, the weight of Application on Voiceprint Recognition is then relatively lower.Calculate the degree of ripeness of ratio juris and current algorithm according to three kinds and be subject to the extraneous factor effects such as environment, three kinds of authentication mode score value weight proportion relations: fingerprint recognition: recognition of face: Application on Voiceprint Recognition=4:3.5:2.5.
Further, also comprise step 1-3-1: choose the last a kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore.By the setting of step 1-3-1, security can be improved further and improve serious forgiveness.
Further, also comprise step 1-2-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-2-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-3-1;
Step 1-2-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-2-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-3-1, step 1-2-2 and step 1-2-3 order can be exchanged.Same,
By such setting, when second time authentication, corresponding upper and lower bound threshold value is set, if exceed this scope, so just judge to authorize or refusal mandate, all can not enter next link again, certainly, this upper limit threshold and lower threshold compared to existing technology in threshold value should have difference, upper limit threshold can arrange relatively high, when single mode authentication uniquely can determine user, and just can be directly same by authentication, when single mode authentication can get rid of the user of all correspondences, next step authentication can not be entered so equally.
Further, also comprise step 1-3-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-3-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-1-1; Step 1-3-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-3-3 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 2-1-1, step 1-3-2 and step 1-3-3 order can be exchanged.
For realizing the second object of the present invention, a kind of intelligent interactive system, includes combinations thereof method for authenticating, comprises the following steps:
First, acquisition of information is carried out by step 1-1-1, user can by present comparatively ripe and in the recognition method that resolution is higher one, identify, and according to the registration of unique point, obtain characteristic information matching value, same, exactly acquisition of information is carried out to another recognition method by step 1-2-1, according to these at least two kinds of recognition method, obtain different characteristic information matching values, because the characteristic information matching value difference in different modes is larger, so characteristic information matching value is changed into corresponding weights by one, like this in each recognition method, the obtainable score value of each unique point coincidence converts, and then it is assigned weight, by the ranking operation disclosed in step 3-1-1, obtain end product, certainly, the size of preset value can be set according to actual conditions, to meet authentication condition, so, by at least two kinds of authentication modes, authentication is carried out to it, and obtain last score in the mode of weighting, so just effectively can evade some authentications caused because of environmental factor or physical factors cannot successful situation, simultaneously, the method of at least two kinds of authentications, be difficult to be obtained feature by other people, so security have also been obtained guarantee, its principle is as follows:
Such setting, if single mode authentication in existing, in order to meet security, it is relatively high that the threshold value passed through arranges by needs, the coincidence such as completing 800 unique points just just can complete authentication, but by design of the present invention, the coincidence that two authentication modes all complete 400 unique points just can complete authentication, so, both compare, security can be guaranteed equally, although the threshold of 400 unique point coincidence needs is lower, if but in three kinds of authentication modes two kinds have the unique point of quantity like this to overlap, user also can be determined very accurately, like this, under the guaranteed prerequisite of security, complete authentication and just become a relatively easy operation, because the change of user's state, change of environment for use etc., all can not affect but the authentication success ratio of a mode.
Further, described preset value is proportional to the value of weight maximum in selected recognition method.By such setting, using maximum weighted value as preset value, judge using a kind of recognition method that weight is the highest as benchmark, such reliability is higher.
Further, also comprise step 1-1-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-1-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-2-1;
Step 1-1-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-1-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-2-1, step 1-1-2 and step 1-1-3 order can be exchanged.
By such setting, when an authentication, corresponding upper and lower bound threshold value is set, if exceed this scope, so just judge to authorize or refusal mandate, all can not enter next link again, certainly, this upper limit threshold and lower threshold compared to existing technology in threshold value should have difference, upper limit threshold can arrange relatively high, when single mode authentication uniquely can determine user, and just can be directly same by authentication, when single mode authentication can get rid of the user of all correspondences, next step authentication can not be entered so equally.
Further, in step 2-1-1, △ A=(Ascore-Amin)/(Amax-Amin).By such setting, number of parameters is reduced, the adjustment that the size of Amax and Amin is carried out its precision can be regulated, one can be played when mating simultaneously and can play preferably authentication effect.
Further, in step 2-1-1, the weight ratio of fingerprint recognition, recognition of face and Application on Voiceprint Recognition is respectively 4:3.5:2.5.These 3 kinds of information are everyone distinctive information.But according to recognition of face, fingerprint recognition and the know-why of Application on Voiceprint Recognition and the scene of reality, although as different in everyone voiceprint, but the extraction of voiceprint is easily subject to the interference of outside noise, the change of own bodies simultaneously also can affect the effect of Application on Voiceprint Recognition, so the weight of recognition method in weight authentication arithmetic is not identical in 3.The level of security of fingerprint recognition is better than recognition of face, and recognition of face is better than Application on Voiceprint Recognition.Weight refers to the weight relationship between the weight relationship between every two kinds of algorithms, 3 kinds of algorithms.As when system uses fingerprint recognition and recognition of face, the weight x% of fingerprint recognition matching value, the weight of recognition of face matching value is y%, x%>y%.This is the algorithmic characteristic of Integrated comparative fingerprint recognition and recognition of face, and recognition of face is for fingerprint recognition, and easily by external environmental interference, as ambient light intensity, so weighted value reduces relatively, the weight of Application on Voiceprint Recognition is then relatively lower.Calculate the degree of ripeness of ratio juris and current algorithm according to three kinds and be subject to the extraneous factor effects such as environment, three kinds of authentication mode score value weight proportion relations: fingerprint recognition: recognition of face: Application on Voiceprint Recognition=4:3.5:2.5.
Further, also comprise step 1-3-1: choose the last a kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore.By the setting of step 1-3-1, security can be improved further and improve serious forgiveness.
Further, also comprise step 1-2-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-2-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-3-1;
Step 1-2-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-2-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-3-1, step 1-2-2 and step 1-2-3 order can be exchanged.Same,
By such setting, when second time authentication, corresponding upper and lower bound threshold value is set, if exceed this scope, so just judge to authorize or refusal mandate, all can not enter next link again, certainly, this upper limit threshold and lower threshold compared to existing technology in threshold value should have difference, upper limit threshold can arrange relatively high, when single mode authentication uniquely can determine user, and just can be directly same by authentication, when single mode authentication can get rid of the user of all correspondences, next step authentication can not be entered so equally.
Further, also comprise step 1-3-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-3-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-1-1; Step 1-3-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-3-3 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 2-1-1, step 1-3-2 and step 1-3-3 order can be exchanged.
By adopting technique scheme, under ensureing the prerequisite of security, substantially increase serious forgiveness, the impalpable phenomenon occurred when avoiding authentication.
Accompanying drawing explanation
Fig. 1 is this method for authenticating process flow diagram one;
Fig. 2 is this method for authenticating flowchart 2;
Fig. 3 is this method for authenticating flow chart 3;
Fig. 4 is this method for authenticating process flow diagram four;
Fig. 5 is this method for authenticating process flow diagram five.
Embodiment
Referring to figs. 1 through Fig. 5, the embodiment of the present invention is described further.
First intelligent interactive system principle is sketched, intelligent interactive Operation system setting has several recognition method, include but not limited to fingerprint, vocal print, face, password, gesture, speech recognition, but core of the present invention is, under the one pattern of intelligent interactive system (pattern higher to security requirement), this combined right-discriminating method is used to realize the function of authorization.
Be below intelligent interactive system preferred embodiment, according to the operative scenario of robot, the duty of system is divided into some mode of operations: special pattern, child mode, entertainment mode, global function pattern, other patterns, the state that different mode of operations not only embodies surrounding environment also embodies the function that system under this pattern can provide.Global function pattern: all functions of robot can use.Special pattern: in manager nobody time, robot must be in heightened security posture, authentication mode must adopt highest level, and must use recognition of face, Application on Voiceprint Recognition and fingerprint recognition start up system under this rank, Cai Nengshi robot enters the duty of global function pattern.Because recognition of face, Application on Voiceprint Recognition and fingerprint recognition have unique identity, can guarantee it is specific operating personnel.Child mode: under this pattern, robot specifically can exchange with children, provides the function of applicable children, can enter global function pattern in this mode by the mode of password.Specific function can only be started under this rank.Entertainment mode: global function pattern can be entered by modes such as speech recognition, touch and passwords in this mode, allow system enter Full Featured duty.Other mode of operations: user can building work pattern voluntarily, the function that under each mode of operation of self-defining, equipment provides, and enters Full Featured authentication mode.
Different working modes has different authorities, means different safe classes.Special pattern has most high safety grade (one-level safe class), and child mode is in secondary safety grade, and entertainment mode is in three level security grades, and global function pattern is in minimum safe grade.User according to other mode of operation of scene setting, and can correspond to corresponding safe class.The mode of amendment system operating mode: 1) system is directly set by interactive mode and enters certain specific mode of operation 2) do not carry out mutual duration by setting is man-machine, enter corresponding level of security after reaching specific duration to enter lower security grade from high safety grade and have to pass through authentication, enter high safety grade by voice, screen key complete operation from lower security grade.If be now in special pattern, prepare to enter entertainment mode, then first must carry out authentication, remove current safe class, enter global function pattern, and then enter entertainment mode.The principle of the second switch mode: if be now in entertainment mode, as unmanned mutual in exceeded specific duration, automatically enter special pattern, enter into the highest level of security.But lower security rank can not be entered from high level of security in this mode.Authentication mode includes, such as: 1) special pattern: the mode unlocking this pattern: recognition of face, Application on Voiceprint Recognition and fingerprint recognition.Special pattern is the highest level of security pattern, so the mode that the mode unlocking this pattern needs is the strictest, so adopt recognition of face, Application on Voiceprint Recognition, fingerprint recognition.A) recognition of face: equipment takes user's face information, by face characteristic information stored in system by camera.When carrying out authentication, system is by camera collection face characteristic information, and compared by face alignment algorithm and the face information deposited in system, after reaching matching threshold, authentication is passed through, otherwise failed authentication.B) Application on Voiceprint Recognition: operator extracts the voiceprint of operator, by vocal print characteristic information stored in system by vocal print checkout equipment.When operator carries out authentication, system by microphone voice checkout equipment, extracts voiceprint, and in the system of depositing in can in voiceprint compare, after reaching matching threshold, authentication is passed through, otherwise failed authentication.C) fingerprint recognition: system is furnished with fingerprint identification device, the prior typing finger print information of operator, when operator needs to carry out authentication, system by the finger print information collected by fingerprint identification device and system stored in finger print information compare, after reaching matching threshold, authentication is passed through, otherwise failed authentication.D) when the authentication of reality, consider that single authentication mode can be subject to extraneous interference, and then affect discrimination, such as during recognition of face, if be subject to affecting discrimination and reducing of ambient light.So we introduce a kind of weight authentication arithmetic based on above three kinds of authentication modes.This Weight algorithm uses recognition of face, fingerprint recognition and Application on Voiceprint Recognition as 3 kinds of authentication modes.Often kind of authentication mode has oneself degree of confidence, the threshold value be namely identified by; There is the threshold value that identification is not passed through simultaneously.The threshold value implication of 3 kinds of authentication modes is described below one by one.Face recognition algorithms, when the characteristic information matching value of the identification object extracted reaches upper limit threshold Fmax, thinks that recognition of face is successful.When the characteristic information matching value of the identification object extracted is lower than lower threshold Fmin, think that recognition of face is failed.When the characteristic information matching value Fscore of the identification object extracted is between Fmax and Fmin, this value is used for carrying out Weight algorithm process with other authentication values.Algorithm for recognizing fingerprint, when the characteristic information matching value of the identification object extracted reaches upper limit threshold Pmax, thinks that fingerprint recognition is successful.When the characteristic information matching value of the identification object extracted is lower than lower threshold Pmin, think that fingerprint recognition is failed.When the characteristic information matching value Pscore of the identification object extracted is between Pmax and Pmin, this value is used for carrying out Weight algorithm process with other authentication values.Voiceprint recognition algorithm, when the characteristic information matching value of the identification object extracted reaches upper limit threshold Smax, thinks that Application on Voiceprint Recognition is successful.When the characteristic information matching value of the vocal print object extracted is lower than lower threshold Smin, think that Application on Voiceprint Recognition is failed.When the characteristic information matching value Sscore of the identification object extracted is between Smax and Smin, this value is used for carrying out Weight algorithm process with other authentication values.2) child mode: when being in child mode, the authentication rank of needs reduces, and the mode of authentication adopts pin mode authentication.Simultaneously for the consideration to this specific crowd of children, need to provide corresponding function.A) code authentication: user can arrange the authentication password under this pattern, is arranged successfully, and the authentication password under child mode is saved in system by system.When user needs manipulation robot, eject Password Input prompting frame by point touching screen, after user inputs password, system is compared with the authentication password under entertainment mode, and after consistent, authentication is successful.The form of authentication password comprises numerical ciphers, sliding unlocks, but is not limited to forenamed mode.3) entertainment mode: when being in entertainment mode, the authentication mode of employing is touch, gesture, voice mode.A) authentication is touched: by touching the touch sensor of head.When after touch hot nose, sensor can send interrupting information to system, and system judges current mode of operation after receiving and touching interrupting information, finds to be in child mode, namely by authentication, opens user interactions.B) gesture authentication: user can be unlocked by specific gesture.User can arrange the authentication gesture under entertainment mode, arranges successfully, and the authentication gesture under entertainment mode is saved in system by system.When user needs manipulation robot, by skidding off specific gesture, compare with the authentication gesture under entertainment mode after systems axiol-ogy to gesture, authentication success after consistent.C) voice authentication: user self-defined authentication can wake word up, after user says and wakes word up, the speech conversion of user can be word by sound identification module, compares with the word that wakes up arranged in system, after comparison is consistent, authentication success.D) the relatively special pattern of the authentication mode under this pattern is easy to identify, safe class reduces, and does not consider the mode that multiple authentication mode combines.User can add voluntarily, modification pattern, add, delete certain mode of operation under function.
Conveniently understand explanation, according to the order of fingerprint recognition, recognition of face, Application on Voiceprint Recognition, to define Pp, Pf, Ps be respectively weight (being referred to as PA above) Pmax, Smax, Fmax is upper limit threshold (being referred to as Amax above), Pmin, Smin, Fmin are lower threshold (being referred to as Amin above), and Pscore, Fscore, Sscore are characteristic matching value (being referred to as Ascore above); △ P, △ F, △ S are respectively weights, are referred to as △ A above.△P=(Pscore-Pmin)/(Pmax-Pmin)；△F=(Fscore-Fmin)/(Fmax-Fmin)；△S=(Sscore-Smin)/(Smax-Smin)。
Embodiment 1 is with reference to shown in Fig. 1, to combined right-discriminating method wherein a kind of embodiment sketch, comprise the following steps: step 1-1-1: choose any one recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore; Step 1-2-1: choose any another kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore; Step 2-1-1: define weight respectively to each recognition method in above step, obtains the weights △ A relevant to its direct proportion according to characteristic of correspondence information matches value Ascore simultaneously respectively; Step 3-1-1: compute weighted to the recognition result of the different recognition method in above step, when its summed result is greater than preset value, then authentication is passed through.When using, the matching value weight of any 2 kinds of algorithm authentications is as follows:
1) fingerprint recognition (Pptof=Pp:(Pp+Pf) %): recognition of face (Pftop=Pf:(Pp+Pf) %)
2) fingerprint recognition (Pptos=Pp:(Pp+Ps) %): Application on Voiceprint Recognition (Pstop=Ps:(Pp+Ps) %)
3) recognition of face (Pftos=Pf:(Pf+Ps) %): Application on Voiceprint Recognition (Pstof=Ps:(Pf+Ps) %)
This weighted value is calculated ratio juris and relative merits according to three kinds and is carried out analysis and draw, fingerprint recognition is less by external interference, and recognition of face easily affects by ambient light, and Application on Voiceprint Recognition is easily subject to the impact of condition etc. of outside noise, people itself.
Following formula is in two kinds of recognition method, preferred scheme, a) as employing fingerprint recognition and recognition of face: T=(Pscore-Pmin)/(Pmax-Pmin)+((Fscore-Fmin)/(Fmax-Fmin)) * (Pftop/Pptof); If T>1, authentication is passed through.B) as employing fingerprint recognition and Application on Voiceprint Recognition: T=(Pscore-Pmin)/(Pmax-Pmin)+((Sscore-Smin)/(Smax-Smin)) * (Pstop/Pptos); If T>1, authentication is passed through.C) as employing recognition of face and Application on Voiceprint Recognition: T=(Fscore-Fmin)/(Fmax-Fmin)+((Sscore-Smin)/(Smax-Smin)) * (Pstof/Pftos); If T>1, authentication is passed through.For a) abbreviation be, T*Pp=(Pscore-Pmin)/(Pmax-Pmin) * Pp+ (Fscore-Fmin)/(Fmax-Fmin) * Pf, when a) formula, namely adopt fingerprint recognition and recognition of face, the weight of fingerprint recognition is greater than the weight of recognition of face, so when carrying out combined type authentication by these two kinds of recognition method, corresponding value is multiplied by with regard to needing the weight obtained, then a result (as shown in formula on the right of above formula) is obtained by weighting algorithm, when its result is greater than preset value, just by authorizing, in equation, this preset value is set to T*Pp, just passable, T can be arbitrary constant, be exactly be according to carrying out authentication (the value Pp due to weight corresponding to fingerprint recognition is greater than the value of weight Pf corresponding to recognition of face) with Pp like this, so when the result that weighting obtains is greater than T*Pp, pass through with regard to authentication.
Embodiment 2 is with reference to shown in Fig. 2, it is compared with the difference of embodiment 1, in step 1-1-1 and step 1-2-1, add step and also comprise step 1-1-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-1-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-2-1; Step 1-1-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-1-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-2-1, step 1-1-2 and step 1-1-3 order can be exchanged.
Embodiment 3 is with reference to shown in Fig. 3, it is compared with the difference of embodiment 1, step 1-3-1 is added in step 1-2-1 and step 2-1-1, choose the last a kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore.Matching value weight when enabling 3 kinds of algorithm authentications is as follows: T=(Pscore-Pmin)/(Pmax-Pmin)+((Fscore-Fmin)/(Fmax-Fmin)) * (Pftop/Pptof)+((Sscore-Smin)/(Smax-Smin)) * (Pstop/Pptos).If T>1, authentication is passed through.The weights obtained in definition single authentication are respectively △ P, △ F, △ S, i.e. the mark of single identification acquisition.
Embodiment 4 is with reference to shown in Fig. 4, it is compared with the difference of embodiment 3, step 1-2-2 is added: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax in step 1-2-1 and step 1-3-1, when the characteristic information matching value Ascore obtained in step 1-2-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-3-1; Step 1-2-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-2-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-3-1, step 1-2-2 and step 1-2-3 order can be exchanged.
Embodiment 5 is with reference to shown in Fig. 5, its comparatively embodiment 4 difference be, in step 1-3-1 and step 2-1-1, add step also comprise step 1-3-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-3-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-1-1;
Step 1-3-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-3-3 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 2-1-1, step 1-3-2 and step 1-3-3 order can be exchanged.
The above is only the preferred embodiment of the present invention, protection scope of the present invention be not only confined to above-described embodiment, and all technical schemes belonged under thinking of the present invention all belong to protection scope of the present invention.It should be pointed out that for those skilled in the art, some improvements and modifications without departing from the principles of the present invention, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (9)
1. a combined right-discriminating method, is characterized in that:
And comprise the following steps:
Step 1-1-1: choose any one recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore;
Step 1-2-1: choose any another kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore;
Step 2-1-1: define weight respectively to each recognition method in above step, obtains the weights △ A relevant to its direct proportion according to characteristic of correspondence information matches value Ascore simultaneously respectively;
Step 3-1-1: compute weighted to the recognition result of the different recognition method in above step, when its summed result is greater than preset value, then authentication is passed through.
2. a kind of combined right-discriminating method as claimed in claim 1, is characterized in that: described preset value is proportional to the value of weight maximum in selected recognition method.
3. a kind of combined right-discriminating method as claimed in claim 1, is characterized in that:
Also comprise step 1-1-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-1-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-2-1;
Step 1-1-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-1-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-2-1, step 1-1-2 and step 1-1-3 order can be exchanged.
4. a kind of combined right-discriminating method as claimed in claim 3, is characterized in that: in step 2-1-1, △ A=(Ascore-Amin)/(Amax-Amin).
5. a kind of combined right-discriminating method as claimed in claim 1, is characterized in that: in step 2-1-1, and the weight ratio of fingerprint recognition, recognition of face and Application on Voiceprint Recognition is respectively 4:3.5:2.5.
6. as a kind of combined right-discriminating method in claim 1-5 as described in any one, it is characterized in that: also comprise step 1-3-1: choose the last a kind of recognition method in fingerprint recognition, recognition of face and Application on Voiceprint Recognition and gather its identifying information, the information that comparison collection obtains draws characteristic information matching value Ascore.
7. a kind of combined right-discriminating method as described in claim 6, is characterized in that:
Also comprise step 1-2-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-2-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-3-1;
Step 1-2-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-2-1 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 1-3-1, step 1-2-2 and step 1-2-3 order can be exchanged.
8. a kind of combined right-discriminating method as described in claim 7, it is characterized in that: also comprise step 1-3-2: in the recognition method selected by each, also correspondence is provided with a upper limit threshold Amax, when the characteristic information matching value Ascore obtained in step 1-3-1 is greater than the upper limit threshold Amax of its correspondence, then authentication is passed through, and no longer enters step 1-1-1;
Step 1-3-3, in recognition method selected by each, also correspondence is provided with a lower limit threshold value A min, when the characteristic information matching value Ascore obtained in step 1-3-3 is less than the lower threshold Amin of its correspondence, then authentication is not passed through, no longer enter step 2-1-1, step 1-3-2 and step 1-3-3 order can be exchanged.
9. an intelligent interactive system, is characterized in that: include the combined right-discriminating method as described in claim 1-8 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510698472.3A CN105224850A (en) | 2015-10-24 | 2015-10-24 | Combined right-discriminating method and intelligent interactive system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510698472.3A CN105224850A (en) | 2015-10-24 | 2015-10-24 | Combined right-discriminating method and intelligent interactive system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105224850A true CN105224850A (en) | 2016-01-06 |
Family
ID=54993814
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510698472.3A Pending CN105224850A (en) | 2015-10-24 | 2015-10-24 | Combined right-discriminating method and intelligent interactive system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105224850A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881604A (en) * | 2017-04-19 | 2018-11-23 | 深圳市汇顶科技股份有限公司 | Light-intensity test method, apparatus and intelligent terminal |
| CN108959865A (en) * | 2017-05-25 | 2018-12-07 | 阿里巴巴集团控股有限公司 | A kind of verification method and device |
| CN109300475A (en) * | 2017-07-25 | 2019-02-01 | 中国电信股份有限公司 | Microphone array sound pick-up method and device |
| CN109815670A (en) * | 2019-01-17 | 2019-05-28 | Oppo广东移动通信有限公司 | Document handling method and Related product |
| CN110364180A (en) * | 2019-06-06 | 2019-10-22 | 北京容联易通信息技术有限公司 | A kind of examination system and method based on audio-video processing |
| CN110599639A (en) * | 2019-08-13 | 2019-12-20 | 深圳市天彦通信股份有限公司 | Identity verification method and related product |
| CN110782575A (en) * | 2019-09-24 | 2020-02-11 | 深圳市智微智能科技开发有限公司 | Storehouse management method and related product |
| CN110992155A (en) * | 2019-11-28 | 2020-04-10 | 珠海采筑电子商务有限公司 | Bidding and enclosing processing method and related product |
| CN110990815A (en) * | 2019-11-11 | 2020-04-10 | 珠海格力电器股份有限公司 | Identity verification method and device |
| CN111971647A (en) * | 2018-04-09 | 2020-11-20 | 麦克赛尔株式会社 | Speech recognition apparatus, cooperation system of speech recognition apparatus, and cooperation method of speech recognition apparatus |
| WO2021000814A1 (en) * | 2019-06-29 | 2021-01-07 | 华为技术有限公司 | Voice control method and related apparatus |
| CN112420035A (en) * | 2018-06-29 | 2021-02-26 | 华为技术有限公司 | Voice control method, wearable device and terminal |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332093A (en) * | 2011-09-19 | 2012-01-25 | 汉王科技股份有限公司 | Identity authentication method and device adopting palmprint and human face fusion recognition |
| CN104008381A (en) * | 2014-06-16 | 2014-08-27 | 浙江维尔科技股份有限公司 | Identity recognition method and device |
| US20150067827A1 (en) * | 2013-08-29 | 2015-03-05 | Pantech Co., Ltd. | Apparatus and method for setting a user-defined pattern for an application |
| CN104573462A (en) * | 2015-01-30 | 2015-04-29 | 北京得意音通技术有限责任公司 | Fingerprint and voiceprint dual-authentication method and fingerprint and voiceprint collector |
-
2015
- 2015-10-24 CN CN201510698472.3A patent/CN105224850A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332093A (en) * | 2011-09-19 | 2012-01-25 | 汉王科技股份有限公司 | Identity authentication method and device adopting palmprint and human face fusion recognition |
| US20150067827A1 (en) * | 2013-08-29 | 2015-03-05 | Pantech Co., Ltd. | Apparatus and method for setting a user-defined pattern for an application |
| CN104008381A (en) * | 2014-06-16 | 2014-08-27 | 浙江维尔科技股份有限公司 | Identity recognition method and device |
| CN104573462A (en) * | 2015-01-30 | 2015-04-29 | 北京得意音通技术有限责任公司 | Fingerprint and voiceprint dual-authentication method and fingerprint and voiceprint collector |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881604A (en) * | 2017-04-19 | 2018-11-23 | 深圳市汇顶科技股份有限公司 | Light-intensity test method, apparatus and intelligent terminal |
| CN108881604B (en) * | 2017-04-19 | 2021-07-20 | 深圳市汇顶科技股份有限公司 | Light intensity detection method and device and intelligent terminal |
| US10935421B2 (en) | 2017-04-19 | 2021-03-02 | Shenzhen GOODIX Technology Co., Ltd. | Light intensity detection method and apparatus, and intelligent terminal |
| CN108959865A (en) * | 2017-05-25 | 2018-12-07 | 阿里巴巴集团控股有限公司 | A kind of verification method and device |
| CN109300475A (en) * | 2017-07-25 | 2019-02-01 | 中国电信股份有限公司 | Microphone array sound pick-up method and device |
| CN111971647A (en) * | 2018-04-09 | 2020-11-20 | 麦克赛尔株式会社 | Speech recognition apparatus, cooperation system of speech recognition apparatus, and cooperation method of speech recognition apparatus |
| CN112420035A (en) * | 2018-06-29 | 2021-02-26 | 华为技术有限公司 | Voice control method, wearable device and terminal |
| CN109815670A (en) * | 2019-01-17 | 2019-05-28 | Oppo广东移动通信有限公司 | Document handling method and Related product |
| CN110364180A (en) * | 2019-06-06 | 2019-10-22 | 北京容联易通信息技术有限公司 | A kind of examination system and method based on audio-video processing |
| CN110364180B (en) * | 2019-06-06 | 2021-10-22 | 北京容联易通信息技术有限公司 | Examination system and method based on audio and video processing |
| WO2021000814A1 (en) * | 2019-06-29 | 2021-01-07 | 华为技术有限公司 | Voice control method and related apparatus |
| CN110599639B (en) * | 2019-08-13 | 2021-05-07 | 深圳市天彦通信股份有限公司 | Identity verification method and related product |
| CN110599639A (en) * | 2019-08-13 | 2019-12-20 | 深圳市天彦通信股份有限公司 | Identity verification method and related product |
| CN110782575A (en) * | 2019-09-24 | 2020-02-11 | 深圳市智微智能科技开发有限公司 | Storehouse management method and related product |
| CN110990815A (en) * | 2019-11-11 | 2020-04-10 | 珠海格力电器股份有限公司 | Identity verification method and device |
| CN110992155A (en) * | 2019-11-28 | 2020-04-10 | 珠海采筑电子商务有限公司 | Bidding and enclosing processing method and related product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105224850A (en) | Combined right-discriminating method and intelligent interactive system | |
| CN111133433B (en) | Automatic authentication for access control using face recognition | |
| Liu et al. | VibWrite: Towards finger-input authentication on ubiquitous surfaces via physical vibration | |
| US9985963B2 (en) | Method and system for authenticating liveness face, and computer program product thereof | |
| CN105068743B (en) | Based on the mobile terminal user identity authentication method for more referring to touch-control behavioural characteristic | |
| CN102332093B (en) | Identity authentication method and device adopting palmprint and human face fusion recognition | |
| CN106068512B (en) | Method and apparatus for verifying user on the mobile device | |
| CN107018121A (en) | The method and device of subscriber authentication | |
| CN104007929B (en) | Based on mobile terminal unlock method and the mobile terminal of gesture identification | |
| CN105760817A (en) | Method and device for recognizing, authenticating, unlocking and encrypting storage space by using human face | |
| KR20170046448A (en) | Method and device for complex authentication | |
| CN105005779A (en) | Face verification anti-counterfeit recognition method and system thereof based on interactive action | |
| CN104063642A (en) | Intelligent terminal safety authentication method and device based on pressure sensor | |
| CN110276189B (en) | User identity authentication method based on gait information | |
| WO2018145309A1 (en) | Method for controlling unmanned aerial vehicle, unmanned aerial vehicle, and remote control device | |
| CN106650392A (en) | VR headset device and unlock method | |
| CN103324947A (en) | Certification method and authentication method | |
| CN105354473A (en) | Iris identification authentication based device desktop unlocking method and apparatus | |
| CN104063643A (en) | Intelligent terminal security authentication method and device based on time interval feature | |
| CN105389497A (en) | Security verification method and system for operation interface of fingerprint recognition | |
| WO2019200572A1 (en) | Identity authentication method, identity authentication device, and electronic apparatus | |
| CN105469024A (en) | Wearable device and control method thereof | |
| Izuta et al. | Screen unlocking method using behavioral characteristics when taking mobile phone from pocket | |
| CN105653993A (en) | Password inputting method, apparatus and electronic device | |
| CN108255531A (en) | Electronic equipment and its visual identifying system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160106 |