CN108108975A - A kind of payment authentication method and terminal based on electrocardiogram and vocal print - Google Patents
A kind of payment authentication method and terminal based on electrocardiogram and vocal print Download PDFInfo
- Publication number
- CN108108975A CN108108975A CN201711316880.3A CN201711316880A CN108108975A CN 108108975 A CN108108975 A CN 108108975A CN 201711316880 A CN201711316880 A CN 201711316880A CN 108108975 A CN108108975 A CN 108108975A
- Authority
- CN
- China
- Prior art keywords
- voiceprint
- information
- electrocardiogram
- vocal print
- payment authentication
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
- G06Q20/40145—Biometric identity checks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
- H04L9/3226—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials using a predetermined code, e.g. password, passphrase or PIN
- H04L9/3231—Biological data, e.g. fingerprint, voice or retina
Abstract
The present invention provides a kind of payment authentication method and terminal based on electrocardiogram and vocal print, method comprises the following steps:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint;If no, according to ECG information, voiceprint, reserved user's ECG information and user's voiceprint, relevant matches operation carried out, according to matching result, judges whether payment authentication succeeds.The present invention makes payment more safe and reliable according to can effectively judge ECG information and voiceprint whether user is in by stress state or sleep state;It can largely be reduced by the method that electrocardiogram is combined with Application on Voiceprint Recognition and usurp risk, strengthen transaction security and reliability.
Description
Technical field
The present invention relates to e-payment technical field more particularly to a kind of payment authentication methods based on electrocardiogram and vocal print
And terminal.
Background technology
With the continuous development of Internet technology, carrying out shopping online by intelligent mobile terminal has become people's life
In essential something, this also greatly facilitates people’s lives.Since shopping online is related to the sensitivity of user
Information, therefore shopping on the web and safer payment authentication mode is needed when being paid;Present payment authentication mode master
To be that payment authentication is carried out by fingerprint or face recognition, have the following disadvantages:Biological characteristic is easily stolen:Finger print information exists
Negotiator is easier to extract when contacting article, and non-living body information, and facial image information is exactly originally disclosed, is passed through
Video takes pictures and is all easy to steal;The biological characteristic stolen is easy for attacking, and uses the fingerprint stolen and facial information can be with
The technology for referring to mould and image synthesis by making respectively attacks payment devices, so as to achieve the purpose that steal brush.
The content of the invention
The technical problems to be solved by the invention are:The present invention provides a kind of payment authentication side based on electrocardiogram and vocal print
Method and terminal improve the security of payment authentication.
In order to solve the above technical problem, the present invention provides a kind of payment authentication method based on electrocardiogram and vocal print,
It is characterised in that it includes following steps:
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge vocal print
It whether there is the high fdrequency component for being more than default second threshold in information;
S2:If no, believed according to ECG information, voiceprint, reserved user's ECG information and user's vocal print
Breath carries out relevant matches operation, according to matching result, judges whether payment authentication succeeds.
The present invention also provides a kind of payment authentication terminal based on electrocardiogram and vocal print, including memory, processor and
The computer program that can be run on a memory and on a processor is stored, the processor is realized following when performing described program
Step:
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge vocal print
It whether there is the high fdrequency component for being more than default second threshold in information;
S2:If no, believed according to ECG information, voiceprint, reserved user's ECG information and user's vocal print
Breath carries out relevant matches operation, according to matching result, judges whether payment authentication succeeds.
Beneficial effects of the present invention are:
A kind of payment authentication method and terminal based on electrocardiogram and vocal print provided by the invention, by ECG information,
It can interpolate that the problem of user whether in sleep state, prevents user in a sleep state, and ECG information is stolen;And
According to voiceprint, judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint, can prevent from passing through spelling
The recorded message that connects carries out payment authentication (recorded message of splicing its there are high fdrequency components in splice sections), and at the same time according to the heart
Electrograph information and voiceprint carry out payment authentication, and the security of payment is improved by way of double authentication, and above-mentioned
ECG information is living body characteristics, and attacker is not easy to camouflage electrocardiogram and payment authentication is attacked, and makes payment safer.
Description of the drawings
Fig. 1 is to be shown according to a kind of key step of payment authentication method based on electrocardiogram and vocal print of the embodiment of the present invention
It is intended to;
Fig. 2 is the structural representation according to a kind of payment authentication terminal based on electrocardiogram and vocal print of the embodiment of the present invention
Figure;
Label declaration:
1st, memory;2nd, processor.
Specific embodiment
For the technology contents that the present invention will be described in detail, the objects and the effects, below in conjunction with embodiment and coordinate attached
Figure is explained in detail.
Fig. 1 is refer to, the present invention provides a kind of payment authentication method based on electrocardiogram and vocal print, feature exists
In comprising the following steps:
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge vocal print
It whether there is the high fdrequency component for being more than default second threshold in information;
S2:If no, believed according to ECG information, voiceprint, reserved user's ECG information and user's vocal print
Breath carries out relevant matches operation, according to matching result, judges whether payment authentication succeeds.
As can be seen from the above description, a kind of payment authentication method based on electrocardiogram and vocal print provided by the invention, passes through the heart
Electrograph information, can interpolate that whether user is in sleep state, prevent user in a sleep state, what ECG information was stolen
Problem;And according to voiceprint, judge to whether there is the high fdrequency component for being more than default second threshold, Neng Goufang in voiceprint
Payment authentication (recorded message of splicing its there are high fdrequency components in splice sections) is only carried out by the recorded message of splicing, and it is same
When according to ECG information and voiceprint, carry out payment authentication, the security of payment improved by way of double authentication,
And above-mentioned ECG information is living body characteristics, attacker is not easy to camouflage electrocardiogram and payment authentication is attacked, and makes payment
It is safer.
Further, further included before the S1:
S01:Show the corresponding specified word information of phonetic entry required during payment verification;
S02:While gathering ECG information, voiceprint is gathered;
S03:Judge whether the corresponding word of voiceprint word corresponding with specified word information is consistent;
S04:If consistent, step S1 is performed, otherwise payment authentication fails.
As can be seen from the above description, the voiceprint of acquisition carries out phonetic entry for user according to the specified word information of display
Acquired voiceprint, and user and the above-mentioned text information specified is unaware of in advance, by the above method, improve payment and test
The security of card.
Further, further included between the S02 and S03:
Noise reduction and filtering process are carried out respectively to the ECG information and voiceprint.
As can be seen from the above description, the accuracy of data processing can be improved by the above method.
Further, show that the corresponding specified word information of phonetic entry required during payment verification is specific in the S01
For:
The random generation required specified word information of phonetic entry, shows the specified word information.
As can be seen from the above description, by the above method, the voice inputted when can prevent criminal from being paid to user into
Row recording, ensure that the security of payment.
Further, if ECG information or voiceprint acquisition fail in preset time, the new finger of random display
Determine action message, and resurvey ECG information and voiceprint.
As can be seen from the above description, by the above method, payment authentication information can be prevented to be stolen, further improves branch
The security paid.
Further, further included before the S1:
It obtains more parts to be in by the first voiceprint of stress state, is calculated the first of every a first voiceprint
Characteristic parameter;
All fisrt feature parameters are fitted, are obtained by the first mathematical modulo between stress state and fisrt feature parameter
Type;
Obtain more parts in Normal Conscious state normal ECG information, more parts in sleep state sleep electrocardiogram letter
Breath is in more parts by the stress ECG information of stress state;
Be calculated every a normal ECG information second feature parameter, the per a sleep ECG information the 3rd
Characteristic parameter and the fourth feature parameter per a stress ECG information;
By the corresponding all second feature parameters of Normal Conscious state, the corresponding all third feature parameters of sleep state and
It is fitted respectively by the corresponding all fourth feature parameters of stress state, obtains corresponding mathematical model under different conditions.
As can be seen from the above description, the first number of source voiceprint under by stress state can be established by the above method
The corresponding different models of ECG information under model and different conditions are learned, so as to the shape residing for follow-up energy accurate judgement user
State improves the security of payment.
Further, the S1 is specially:
According to the ECG information and voiceprint acquired, Characteristics of electrocardiogram parameter is calculated respectively and vocal print is special
Levy parameter;
According to the vocal print feature parameter and the first mathematical model, judge whether user is in by stress state, Yi Jigen
According to corresponding mathematical model under Characteristics of electrocardiogram parameter and different conditions, judge whether user is in by stress state or sleep shape
State and judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint.
As can be seen from the above description, by the above method, it can judge whether use is in by stress state, sleep exactly
State, with prevent user coerced or sleep state under, user information is usurped by criminal, user is made to bring massive losses
The problem of.
Further, the S2 is specially:
If no, the vocal print feature parameter and Characteristics of electrocardiogram parameter are encrypted, and is sent to server, so that server
Vocal print feature parameter characteristic parameter corresponding with reserved user's voiceprint is subjected to significance analysis and electrocardiogram is special
It levies parameter characteristic parameter corresponding with reserved user's ECG information and carries out significance analysis, obtain significance analysis result;
According to the significance analysis as a result, judging whether payment authentication passes through.
As can be seen from the above description, transmission is encrypted in vocal print feature parameter and Characteristics of electrocardiogram parameter in transmission process,
The problem of user data being prevented to be stolen, user is made to cause certain economic loss;Pass through significance analysis, Neng Gouzhun simultaneously
Really judge whether vocal print feature parameter characteristic parameter corresponding with reserved user's voiceprint matches, and can accurate judgement
Whether Characteristics of electrocardiogram parameter characteristic parameter corresponding with reserved user's ECG information matches;The side of above-mentioned double verification
Formula improves the security of payment authentication.
Further, " judge to whether there is the high fdrequency component more than predetermined threshold value in voiceprint " specially:
According to the voiceprint, high fdrequency component set is calculated, judges to whether there is in the high fdrequency component set
More than the high fdrequency component of predetermined threshold value.
As can be seen from the above description, by the above method, it can accurately calculate whether voiceprint includes high fdrequency component, with
It prevents from being used for payment verification by the recording of synthesis, improves the security of payment.
Further, further included before the S1:
Authentication judgement is carried out to default transaction terminal and server, if failed authentication, payment authentication failure terminates to hand over
Easily;
If authenticate the current location information for successfully, obtaining the transaction terminal;
The current location information is encrypted, obtains position encryption information;
The position encryption information is sent to server so that server the position encryption information is stored in it is default
Security log information in.
As can be seen from the above description, information the location of during store transaction in process of exchange, makes transaction location that can trace.
Refer to Fig. 2, the present invention provides a kind of payment authentication terminal based on electrocardiogram and vocal print, including memory 1,
Processor 2 and the computer program that can be run on memory 1 and on processor 2 is stored in, the processor 2 performs the journey
Following steps are realized during sequence:
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge vocal print
It whether there is the high fdrequency component for being more than default second threshold in information;
S2:If no, believed according to ECG information, voiceprint, reserved user's ECG information and user's vocal print
Breath carries out relevant matches operation, according to matching result, judges whether payment authentication succeeds.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, the S1 are further included before:
S01:Show the corresponding specified word information of phonetic entry required during payment verification;
S02:While gathering ECG information, voiceprint is gathered;
S03:Judge whether the corresponding word of voiceprint word corresponding with specified word information is consistent;
S04:If consistent, step S1 is performed, otherwise payment authentication fails.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, which is characterized in that the S02
It is further included between S03:
Noise reduction and filtering process are carried out respectively to the ECG information and voiceprint.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print shows that payment is tested in the S01
The corresponding specified word information of required phonetic entry is specially during card:
The random generation required specified word information of phonetic entry, shows the specified word information.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, if the electrocardio in preset time
Figure information or voiceprint acquisition failure, the then new required movement information of random display, and resurvey ECG information harmony
Line information.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, the S1 are further included before:
It obtains more parts to be in by the first voiceprint of stress state, is calculated the first of every a first voiceprint
Characteristic parameter;
All fisrt feature parameters are fitted, are obtained by the first mathematical modulo between stress state and fisrt feature parameter
Type;
Obtain more parts in Normal Conscious state normal ECG information, more parts in sleep state sleep electrocardiogram letter
Breath is in more parts by the stress ECG information of stress state;
Be calculated every a normal ECG information second feature parameter, the per a sleep ECG information the 3rd
Characteristic parameter and the fourth feature parameter per a stress ECG information;
By the corresponding all second feature parameters of Normal Conscious state, the corresponding all third feature parameters of sleep state and
It is fitted respectively by the corresponding all fourth feature parameters of stress state, obtains corresponding mathematical model under different conditions.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, the S1 are specially:
According to the ECG information and voiceprint acquired, Characteristics of electrocardiogram parameter is calculated respectively and vocal print is special
Levy parameter;
According to the vocal print feature parameter and the first mathematical model, judge whether user is in by stress state, Yi Jigen
According to corresponding mathematical model under Characteristics of electrocardiogram parameter and different conditions, judge whether user is in by stress state or sleep shape
State and judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, the S2 are specially:
If no, the vocal print feature parameter and Characteristics of electrocardiogram parameter are encrypted, and is sent to server, so that server
Vocal print feature parameter characteristic parameter corresponding with reserved user's voiceprint is subjected to significance analysis and electrocardiogram is special
It levies parameter characteristic parameter corresponding with reserved user's ECG information and carries out significance analysis, obtain significance analysis result;
According to the significance analysis as a result, judging whether payment authentication passes through.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, " judge in voiceprint whether
In the presence of the high fdrequency component more than predetermined threshold value " be specially:
According to the voiceprint, high fdrequency component set is calculated, judges to whether there is in the high fdrequency component set
More than the high fdrequency component of predetermined threshold value.
Further, a kind of payment authentication terminal based on electrocardiogram and vocal print, the S1 are further included before:
Authentication judgement is carried out to default transaction terminal and server, if failed authentication, payment authentication failure terminates to hand over
Easily;
If authenticate the current location information for successfully, obtaining the transaction terminal;
The current location information is encrypted, obtains position encryption information;
The position encryption information is sent to server so that server the position encryption information is stored in it is default
Security log information in.
Fig. 1 is refer to, the embodiment of the present invention one is:
The present invention provides a kind of payment authentication methods based on electrocardiogram and vocal print, comprise the following steps:
S0:The random generation required specified word information of phonetic entry, shows the specified word information;Gather electrocardio
While figure information, voiceprint is gathered;After noise reduction and filtering process being carried out to the voiceprint and ECG information respectively,
Judge whether the corresponding word of voiceprint word corresponding with specified word information is consistent, it is no if unanimously, performing step S1
Then payment authentication fails;
Wherein, if voiceprint or ECG information acquisition failure in preset time, re-execute step S0;
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge vocal print
It whether there is the high fdrequency component for being more than default second threshold in information;
S2:If no, believed according to ECG information, voiceprint, reserved user's ECG information and user's vocal print
Breath carries out relevant matches operation, according to matching result, judges whether payment authentication succeeds.
The embodiment of the present invention two is:
The present invention provides a kind of payment authentication methods based on electrocardiogram and vocal print, comprise the following steps:
It obtains more parts and is in the fisrt feature ginseng that every a voiceprint by the voiceprint of stress state, is calculated
Number;All fisrt feature parameters are fitted, are obtained by the first mathematical model between stress state and fisrt feature parameter;
Wherein " all fisrt feature parameters of fitting are obtained by the first number between stress state and fisrt feature parameter
Learn model " be specially:
It reads per the corresponding fisrt feature parameter of a sample (voiceprint), passes through deep learning convolutional neural networks
Method is fitted all fisrt feature parameters, that is, defines neutral net, collect initial data, is classification based training, correction, defeated
Go out as a result, obtaining by the first mathematical model between stress state and fisrt feature parameter;
Obtain more parts in Normal Conscious state normal ECG information, more parts in sleep state sleep electrocardiogram letter
Breath is in more parts by the stress ECG information of stress state;The second feature of every a normal ECG information is calculated
Parameter, the third feature parameter per a sleep ECG information and the fourth feature parameter per a stress ECG information;
By the corresponding all second feature parameters of Normal Conscious state, the corresponding all third feature parameters of sleep state and by stress shape
The corresponding all fourth feature parameters of state are fitted respectively, obtain corresponding mathematical model under different conditions;
Wherein, by fitting, corresponding mathematical model under different conditions is obtained, used method is again by depth
The method for practising convolutional neural networks is respectively defined corresponding all characteristic parameters under different conditions neutral net, collects original
Beginning data, classification based training, correction, output are as a result, can obtain corresponding mathematical model under different conditions.
In payment, authentication judgement first is carried out to default transaction terminal and server, if failed authentication, payment authentication
Failure, closes the trade;
If authenticate the current location information for successfully, obtaining the transaction terminal;The current location information is encrypted, is obtained
Position encryption information;The position encryption information is sent to server, so that server preserves the position encryption information
In default security log information;
It is random to generate the required specified word information of phonetic entry after authenticating successfully;Gather the same of ECG information
When, gather voiceprint;After carrying out noise reduction and filtering process respectively to the voiceprint and ECG information, judge that vocal print is believed
Whether breath word corresponding with specified word information is consistent, if inconsistent, payment authentication failure;Otherwise, following steps are performed:
According to the ECG information and voiceprint acquired, Characteristics of electrocardiogram parameter is calculated respectively and vocal print is special
Levy parameter;
According to the vocal print feature parameter and the first mathematical model, judge whether user is in by stress state, Yi Jigen
According to corresponding mathematical model under Characteristics of electrocardiogram parameter and different conditions, judge whether user is in by stress state or sleep shape
State and judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint;
If no, the vocal print feature parameter and Characteristics of electrocardiogram parameter are encrypted, and is sent to server, so that service
Vocal print feature parameter characteristic parameter corresponding with reserved user's voiceprint is carried out significance analysis by device by F methods of inspection,
And Characteristics of electrocardiogram parameter characteristic parameter corresponding with reserved user's ECG information is subjected to conspicuousness by F methods of inspection
Analysis, obtains significance analysis result;Significance analysis result includes vocal print feature parameter significance analysis result and electrocardiogram
Feature significance analysis result if wherein any one conspicuousness deficiency, judges payment authentication not by the way that only two significantly
Property analysis result conspicuousness is apparent, and payment authentication passes through.
Fig. 2 is refer to, the embodiment of the present invention three is:
The present invention provides a kind of payment authentication terminals based on electrocardiogram and vocal print, including memory, processor and deposit
The computer program that can be run on a memory and on a processor is stored up, the processor realizes following step when performing described program
Suddenly:
S0:The random generation required specified word information of phonetic entry, shows the specified word information;Gather electrocardio
While figure information, voiceprint is gathered;After noise reduction and filtering process being carried out to the voiceprint and ECG information respectively,
Judge whether the corresponding word of voiceprint word corresponding with specified word information is consistent, it is no if unanimously, performing step S1
Then payment authentication fails;
Wherein, if voiceprint or ECG information acquisition failure in preset time, re-execute step S0;
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge vocal print
It whether there is the high fdrequency component for being more than default second threshold in information;
S2:If no, believed according to ECG information, voiceprint, reserved user's ECG information and user's vocal print
Breath carries out relevant matches operation, according to matching result, judges whether payment authentication succeeds.
The embodiment of the present invention four is:
The present invention provides a kind of payment authentication terminals based on electrocardiogram and vocal print, including memory, processor and deposit
The computer program that can be run on a memory and on a processor is stored up, the processor realizes following step when performing described program
Suddenly:
It obtains more parts and is in the fisrt feature ginseng that every a voiceprint by the voiceprint of stress state, is calculated
Number;All fisrt feature parameters are fitted, are obtained by the first mathematical model between stress state and fisrt feature parameter;
Wherein " all fisrt feature parameters of fitting are obtained by the first number between stress state and fisrt feature parameter
Learn model " be specially:
It reads per the corresponding fisrt feature parameter of a sample (voiceprint), passes through deep learning convolutional neural networks
Method is fitted all fisrt feature parameters, that is, defines neutral net, collect initial data, is classification based training, correction, defeated
Go out as a result, obtaining by the first mathematical model between stress state and fisrt feature parameter;
Obtain more parts in Normal Conscious state normal ECG information, more parts in sleep state sleep electrocardiogram letter
Breath is in more parts by the stress ECG information of stress state;The second feature of every a normal ECG information is calculated
Parameter, the third feature parameter per a sleep ECG information and the fourth feature parameter per a stress ECG information;
By the corresponding all second feature parameters of Normal Conscious state, the corresponding all third feature parameters of sleep state and by stress shape
The corresponding all fourth feature parameters of state are fitted respectively, obtain corresponding mathematical model under different conditions;
Wherein, by fitting, corresponding mathematical model under different conditions is obtained, used method is again by depth
The method for practising convolutional neural networks is respectively defined corresponding all characteristic parameters under different conditions neutral net, collects original
Beginning data, classification based training, correction, output are as a result, can obtain corresponding mathematical model under different conditions.
In payment, authentication judgement first is carried out to default transaction terminal and server, if failed authentication, payment authentication
Failure, closes the trade;
If authenticate the current location information for successfully, obtaining the transaction terminal;The current location information is encrypted, is obtained
Position encryption information;The position encryption information is sent to server, so that server preserves the position encryption information
In default security log information;
It is random to generate the required specified word information of phonetic entry after authenticating successfully;Gather the same of ECG information
When, gather voiceprint;After carrying out noise reduction and filtering process respectively to the voiceprint and ECG information, judge that vocal print is believed
Whether breath word corresponding with specified word information is consistent, if inconsistent, payment authentication failure;Otherwise, following steps are performed:
According to the ECG information and voiceprint acquired, Characteristics of electrocardiogram parameter is calculated respectively and vocal print is special
Levy parameter;
According to the vocal print feature parameter and the first mathematical model, judge whether user is in by stress state, Yi Jigen
According to corresponding mathematical model under Characteristics of electrocardiogram parameter and different conditions, judge whether user is in by stress state or sleep shape
State and judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint;
If no, the vocal print feature parameter and Characteristics of electrocardiogram parameter are encrypted, and is sent to server, so that service
Vocal print feature parameter characteristic parameter corresponding with reserved user's voiceprint is carried out significance analysis by device by F methods of inspection,
And Characteristics of electrocardiogram parameter characteristic parameter corresponding with reserved user's ECG information is subjected to conspicuousness by F methods of inspection
Analysis, obtains significance analysis result;Significance analysis result includes vocal print feature parameter significance analysis result and electrocardiogram
Feature significance analysis result if wherein any one conspicuousness deficiency, judges payment authentication not by the way that only two significantly
Property analysis result conspicuousness is apparent, and payment authentication passes through.
The embodiment of the present invention five is:
The present invention provides a kind of POS machine, and including MCU (micro-control module), microphone, ECG acquisition modules (adopt by electrocardiogram
Collection module) and liquid crystal display, the MCU be electrically connected respectively with microphone, ECG acquisition modules and liquid crystal display;
1) before dispatching from the factory, the software of POS is trained by substantial amounts of machine learning, including the vocal print letter under the normal mood of parts up to ten thousand
Breath is with the voiceprint coerced and up to ten thousand parts of Normal Conscious state electrocardiograms, sleep electrocardiogram and by the heart under stress state
Electrograph.Identification software is deep by being utilized to all training samples by the reading to every a training sample progress special parameter
The method of degree study convolutional neural networks is fitted to a calculation formula, and (key step has:It defines neutral net, collect original number
According to, classification based training, correction, output result), the relations of the results such as these parameters and mood, sleep is drawn, so that vocal print knowledge
There is identification information source with the software of electrocardiogram identification whether in stress and the ability slept.
2) POS is authenticated with transaction backstage:If failed authentication, then it represents that POS does not have trading privilege, closes the trade;If
It authenticating successfully, then it represents that POS has trading privilege, opens POS and the encryption on backstage of merchandising, if POS machine is built-in with wireless module,
Encryption uploads current base station position at this time, and the content of the security log as transaction backstage preserves.
3) POS machine inside MCU generates the text information of formulation at random, by liquid crystal display negotiator is prompted to be read using microphone
Go out corresponding text information, while gather the ECG information of the negotiator by ECG acquisition modules by liquid crystal display prompting user.
4) ECG information of ECG acquisition modules acquisition negotiator, meanwhile, microphone gathers the acoustic information of negotiator.
5) MCU pre-processes acoustic information, the processing etc. such as including noise reduction.
6) MCU checks sound legitimacy, including checking whether sound-content is consistent with prompting, checking for more than threshold
The high-frequency information of value checks whether and is not under the nervous unhealthy emotion feared, if check not by, be judged to invalid information,
Refused, closed the trade.
7) MCU calculates vocal print feature information, i.e. MFCC (Mel Frequency Cepstrum Coefficients, Mei Er
Frequency cepstral coefficient).
8) MCU is transmitted to transaction backstage after vocal print feature value is encrypted.
9) holder's voiceprint that the voiceprint characteristic value of upload and bank are reserved in transaction backstage carries out conspicuousness
Analysis, determines whether to merchandise according to analysis result:If conspicuousness is insufficient, shows that Application on Voiceprint Recognition fails, inform that POS terminates
Transaction;If conspicuousness is apparent, shows Application on Voiceprint Recognition success, inform that POS allows to merchandise.
10) MCU pre-processes ECG information, electrocardiographic wave, denoising including interception single cardiac cycle.For
External harvester, it is necessary to do decryption oprerations before pre-processing.
11) whether calculating ECG is in the nervous stress state or sleep state feared, if being then considered illegally to believe
Breath, is refused, is closed the trade;Otherwise continuous business.
12) characteristic value of wave amplitude, area and the average of MCU calculating ECGs P ripples, QRS complex and T ripples as electrocardiogram.
13) MCU is transmitted to transaction backstage after Characteristics of electrocardiogram value is encrypted.
14) holder's ECG information that the Characteristics of electrocardiogram value of upload and bank are reserved in transaction backstage carries out conspicuousness
Analysis:If conspicuousness is insufficient, then it represents that electrocardiogram recognition failures inform that POS closes the trade;If conspicuousness is apparent, then it represents that the heart
Electrograph identifies successfully, carries out backstage transaction, and transaction results are informed POS.
15) transaction results are prompted to negotiator by POS machine.
In conclusion a kind of payment authentication method and terminal based on electrocardiogram and vocal print provided by the invention, pass through the heart
Electrograph information, can interpolate that whether user is in sleep state, prevent user in a sleep state, what ECG information was stolen
Problem;Judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint, can prevent the recording by splicing
Information carries out payment authentication (recorded message of splicing its there are high fdrequency components in splice sections);The present invention is by electrocardiogram simultaneously
Information and voiceprint can effectively judge whether user is in by stress state, make payment more safe and reliable;Pass through sound
The method that line is combined with electrocardiogram identification, which can be reduced largely, usurps risk, strengthens transaction security and reliability.It holds
The voiceprint of people is transmitted to server as sensitive information with ECG information by cipher mode, and unidirectional uplink is only allowed to pass
Defeated encrypted characteristic parameter, avoids sensitive information leakage.Above-mentioned ECG information is not easy to be stolen, even if ECG information
It is stolen, attacker is not easy to camouflage electrocardiogram and POS is attacked, and effectively improves the security of payment authentication, makes transaction
It is more safe and reliable.
The foregoing is merely the embodiment of the present invention, are not intended to limit the scope of the invention, every to utilize this hair
The equivalents that bright specification and accompanying drawing content are made directly or indirectly are used in other related technical areas, similarly
It is included within the scope of the present invention.
Claims (20)
1. a kind of payment authentication method based on electrocardiogram and vocal print, which is characterized in that comprise the following steps:
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge voiceprint
In whether there is the high fdrequency component for being more than default second threshold;
S2:If no, according to ECG information, voiceprint, reserved user's ECG information and user's voiceprint, into
Row relevant matches operate, and according to matching result, judge whether payment authentication succeeds.
A kind of 2. payment authentication method based on electrocardiogram and vocal print according to claim 1, which is characterized in that the S1
It further includes before:
S01:Show the corresponding specified word information of phonetic entry required during payment verification;
S02:While gathering ECG information, voiceprint is gathered;
S03:Judge whether the corresponding word of voiceprint word corresponding with specified word information is consistent;
S04:If consistent, step S1 is performed, otherwise payment authentication fails.
3. a kind of payment authentication method based on electrocardiogram and vocal print according to claim 2, which is characterized in that described
It is further included between S02 and S03:
Noise reduction and filtering process are carried out respectively to the ECG information and voiceprint.
4. a kind of payment authentication method based on electrocardiogram and vocal print according to claim 2, which is characterized in that described
Show that the corresponding specified word information of phonetic entry required during payment verification is specially in S01:
The random generation required specified word information of phonetic entry, shows the specified word information.
5. a kind of payment authentication method based on electrocardiogram and vocal print according to claim 4, which is characterized in that if pre-
If ECG information or voiceprint acquisition failure, the then new required movement information of random display, and resurvey the heart in the time
Electrograph information and voiceprint.
A kind of 6. payment authentication method based on electrocardiogram and vocal print according to claim 1, which is characterized in that the S1
It further includes before:
It obtains more parts and is in the fisrt feature that every a first voiceprint by the first voiceprint of stress state, is calculated
Parameter;
All fisrt feature parameters are fitted, are obtained by the first mathematical model between stress state and fisrt feature parameter;
Obtain more parts in Normal Conscious state normal ECG information, more parts in sleep state sleep ECG information and
More parts are in by the stress ECG information of stress state;
Second feature parameter, the third feature per a sleep ECG information that every a normal ECG information is calculated
Parameter and the fourth feature parameter per a stress ECG information;
By the corresponding all second feature parameters of Normal Conscious state, the corresponding all third feature parameters of sleep state and coerced
Compel the corresponding all fourth feature parameters of state to be fitted respectively, obtain corresponding mathematical model under different conditions.
A kind of 7. payment authentication method based on electrocardiogram and vocal print according to claim 6, which is characterized in that the S1
Specially:
According to the ECG information and voiceprint acquired, Characteristics of electrocardiogram parameter and vocal print feature ginseng are calculated respectively
Number;
According to the vocal print feature parameter and the first mathematical model, judge whether user is in by stress state and according to the heart
Corresponding mathematical model under electrograph characteristic parameter and different conditions, judges whether user is in by stress state or sleep state,
And judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint.
A kind of 8. payment authentication method based on electrocardiogram and vocal print according to claim 7, which is characterized in that the S2
Specially:
If no, the vocal print feature parameter and Characteristics of electrocardiogram parameter are encrypted, and is sent to server, so that server is by sound
Line characteristic parameter characteristic parameter corresponding with reserved user's voiceprint carries out significance analysis and joins Characteristics of electrocardiogram
Number characteristic parameter corresponding with reserved user's ECG information carries out significance analysis, obtains significance analysis result;According to
The significance analysis is as a result, judge whether payment authentication passes through.
9. a kind of payment authentication method based on electrocardiogram and vocal print according to claim 1 or 8, which is characterized in that " sentence
Whether there is the high fdrequency component more than predetermined threshold value in disconnected voiceprint " be specially:
According to the voiceprint, high fdrequency component set is calculated, judges in the high fdrequency component set with the presence or absence of being more than
The high fdrequency component of predetermined threshold value.
10. a kind of payment authentication method based on electrocardiogram and vocal print according to claim 1, which is characterized in that described
It is further included before S1:
Authentication judgement is carried out to default transaction terminal and server, if failed authentication, payment authentication failure is closed the trade;
If authenticate the current location information for successfully, obtaining the transaction terminal;
The current location information is encrypted, obtains position encryption information;
The position encryption information is sent to server, so that the position encryption information is stored in default peace by server
In full log information.
11. a kind of payment authentication terminal based on electrocardiogram and vocal print including memory, processor and stores on a memory simultaneously
The computer program that can be run on a processor, which is characterized in that the processor realizes following steps when performing described program:
S1:According to ECG information, judge user whether in sleep state;And according to voiceprint, judge voiceprint
In whether there is the high fdrequency component for being more than default second threshold;
S2:If no, according to ECG information, voiceprint, reserved user's ECG information and user's voiceprint, into
Row relevant matches operate, and according to matching result, judge whether payment authentication succeeds.
12. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 11, which is characterized in that described
It is further included before S1:
S01:Show the corresponding specified word information of phonetic entry required during payment verification;
S02:While gathering ECG information, voiceprint is gathered;
S03:Judge whether the corresponding word of voiceprint word corresponding with specified word information is consistent;
S04:If consistent, step S1 is performed, otherwise payment authentication fails.
13. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 12, which is characterized in that described
It is further included between S02 and S03:
Noise reduction and filtering process are carried out respectively to the ECG information and voiceprint.
14. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 12, which is characterized in that described
Show that the corresponding specified word information of phonetic entry required during payment verification is specially in S01:
The random generation required specified word information of phonetic entry, shows the specified word information.
15. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 14, which is characterized in that if
ECG information or voiceprint acquisition failure, the then new required movement information of random display, and resurveying in preset time
ECG information and voiceprint.
16. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 11, which is characterized in that described
It is further included before S1:
It obtains more parts and is in the fisrt feature that every a first voiceprint by the first voiceprint of stress state, is calculated
Parameter;
All fisrt feature parameters are fitted, are obtained by the first mathematical model between stress state and fisrt feature parameter;
Obtain more parts in Normal Conscious state normal ECG information, more parts in sleep state sleep ECG information and
More parts are in by the stress ECG information of stress state;
Second feature parameter, the third feature per a sleep ECG information that every a normal ECG information is calculated
Parameter and the fourth feature parameter per a stress ECG information;
By the corresponding all second feature parameters of Normal Conscious state, the corresponding all third feature parameters of sleep state and coerced
Compel the corresponding all fourth feature parameters of state to be fitted respectively, obtain corresponding mathematical model under different conditions.
17. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 16, which is characterized in that described
S1 is specially:
According to the ECG information and voiceprint acquired, Characteristics of electrocardiogram parameter and vocal print feature ginseng are calculated respectively
Number;
According to the vocal print feature parameter and the first mathematical model, judge whether user is in by stress state and according to the heart
Corresponding mathematical model under electrograph characteristic parameter and different conditions, judges whether user is in by stress state or sleep state,
And judge to whether there is the high fdrequency component for being more than default second threshold in voiceprint.
18. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 17, which is characterized in that described
S2 is specially:
If no, the vocal print feature parameter and Characteristics of electrocardiogram parameter are encrypted, and is sent to server, so that server is by sound
Line characteristic parameter characteristic parameter corresponding with reserved user's voiceprint carries out significance analysis and joins Characteristics of electrocardiogram
Number characteristic parameter corresponding with reserved user's ECG information carries out significance analysis, obtains significance analysis result;According to
The significance analysis is as a result, judge whether payment authentication passes through.
19. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 11 or 18, which is characterized in that
" judging to whether there is the high fdrequency component more than predetermined threshold value in voiceprint " is specially:
According to the voiceprint, high fdrequency component set is calculated, judges in the high fdrequency component set with the presence or absence of being more than
The high fdrequency component of predetermined threshold value.
20. a kind of payment authentication terminal based on electrocardiogram and vocal print according to claim 11, which is characterized in that described
It is further included before S1:
Authentication judgement is carried out to default transaction terminal and server, if failed authentication, payment authentication failure is closed the trade;
If authenticate the current location information for successfully, obtaining the transaction terminal;
The current location information is encrypted, obtains position encryption information;
The position encryption information is sent to server, so that the position encryption information is stored in default peace by server
In full log information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711316880.3A CN108108975A (en) | 2017-12-12 | 2017-12-12 | A kind of payment authentication method and terminal based on electrocardiogram and vocal print |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711316880.3A CN108108975A (en) | 2017-12-12 | 2017-12-12 | A kind of payment authentication method and terminal based on electrocardiogram and vocal print |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108108975A true CN108108975A (en) | 2018-06-01 |
Family
ID=62215600
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711316880.3A Pending CN108108975A (en) | 2017-12-12 | 2017-12-12 | A kind of payment authentication method and terminal based on electrocardiogram and vocal print |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108108975A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112438731A (en) * | 2019-08-27 | 2021-03-05 | 深圳市理邦精密仪器股份有限公司 | Electrocardiosignal analysis method, electrocardiosignal analysis device and storage medium |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6219439B1 (en) * | 1998-07-09 | 2001-04-17 | Paul M. Burger | Biometric authentication system |
KR20100136354A (en) * | 2009-06-18 | 2010-12-28 | 주식회사 비즈모델라인 | System and method for settling mobile phone by multiple authentication mode otp authentication with biometrics and recording medium |
CN102402985A (en) * | 2010-09-14 | 2012-04-04 | 盛乐信息技术(上海)有限公司 | Voiceprint authentication system for improving voiceprint identification safety and method for realizing the same |
CN103136465A (en) * | 2013-03-06 | 2013-06-05 | 天津理工大学 | Method of using electrocardiosignals for identity recognition |
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 |
CN105488680A (en) * | 2015-11-27 | 2016-04-13 | 东莞酷派软件技术有限公司 | Payment method and device |
CN105611531A (en) * | 2015-10-30 | 2016-05-25 | 东莞酷派软件技术有限公司 | Terminal control method, terminal control device and terminal |
CN105678539A (en) * | 2016-01-11 | 2016-06-15 | 广东小天才科技有限公司 | NFC payment method for wearable device |
CN105718874A (en) * | 2016-01-18 | 2016-06-29 | 北京天诚盛业科技有限公司 | Method and device of in-vivo detection and authentication |
CN106157036A (en) * | 2015-04-14 | 2016-11-23 | 广州杰赛科技股份有限公司 | A kind of authentication device based on heart beating feature |
CN106446627A (en) * | 2016-08-31 | 2017-02-22 | 广州三星通信技术研究有限公司 | Method and device for carrying out identity authentication at electronic terminal |
CN106651380A (en) * | 2016-12-06 | 2017-05-10 | 深圳市金立通信设备有限公司 | Safety monitoring method and terminal |
-
2017
- 2017-12-12 CN CN201711316880.3A patent/CN108108975A/en active Pending
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6219439B1 (en) * | 1998-07-09 | 2001-04-17 | Paul M. Burger | Biometric authentication system |
KR20100136354A (en) * | 2009-06-18 | 2010-12-28 | 주식회사 비즈모델라인 | System and method for settling mobile phone by multiple authentication mode otp authentication with biometrics and recording medium |
CN102402985A (en) * | 2010-09-14 | 2012-04-04 | 盛乐信息技术(上海)有限公司 | Voiceprint authentication system for improving voiceprint identification safety and method for realizing the same |
CN103136465A (en) * | 2013-03-06 | 2013-06-05 | 天津理工大学 | Method of using electrocardiosignals for identity recognition |
CN106157036A (en) * | 2015-04-14 | 2016-11-23 | 广州杰赛科技股份有限公司 | A kind of authentication device based on heart beating feature |
CN105611531A (en) * | 2015-10-30 | 2016-05-25 | 东莞酷派软件技术有限公司 | Terminal control method, terminal control device and terminal |
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 |
CN105488680A (en) * | 2015-11-27 | 2016-04-13 | 东莞酷派软件技术有限公司 | Payment method and device |
CN105678539A (en) * | 2016-01-11 | 2016-06-15 | 广东小天才科技有限公司 | NFC payment method for wearable device |
CN105718874A (en) * | 2016-01-18 | 2016-06-29 | 北京天诚盛业科技有限公司 | Method and device of in-vivo detection and authentication |
CN106446627A (en) * | 2016-08-31 | 2017-02-22 | 广州三星通信技术研究有限公司 | Method and device for carrying out identity authentication at electronic terminal |
CN106651380A (en) * | 2016-12-06 | 2017-05-10 | 深圳市金立通信设备有限公司 | Safety monitoring method and terminal |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112438731A (en) * | 2019-08-27 | 2021-03-05 | 深圳市理邦精密仪器股份有限公司 | Electrocardiosignal analysis method, electrocardiosignal analysis device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108124488A (en) | A kind of payment authentication method and terminal based on face and vocal print | |
CN109660509A (en) | Login method, device, system and storage medium based on recognition of face | |
US20180300487A1 (en) | Framework for security strength and performance analysis of machine learning based biometric systems | |
WO2017071126A1 (en) | Touch-screen user key-press behavior pattern construction and analysis system and identity recognition method thereof | |
KR101019844B1 (en) | Method and apparatus for electro-biometric identity recognition | |
US20060136744A1 (en) | Method and apparatus for electro-biometric identity recognition | |
CN108804884A (en) | Identity authentication method, device and computer storage media | |
CN110459204A (en) | Audio recognition method, device, storage medium and electronic equipment | |
CN104331652A (en) | Dynamic cipher generation method for electronic equipment for fingerprint and voice recognition | |
EP2148295A1 (en) | Vein pattern management system, vein pattern registration device, vein pattern authentication device, vein pattern registration method, vein pattern authentication method, program, and vein data structure | |
JP5642210B2 (en) | Method and apparatus for electronic biometric identification recognition | |
US20080092232A1 (en) | Method for improving security of the security token and apparatus therefor | |
CN102413101A (en) | Voice-print authentication system having voice-print password voice prompting function and realization method thereof | |
Sun et al. | A 3‐D hand gesture signature based biometric authentication system for smartphones | |
CN106878344A (en) | A kind of biological characteristic authentication, register method and device | |
CN103699996A (en) | Payment authentication method based on human biological characteristics | |
CN109543635A (en) | Biopsy method, device, system, unlocking method, terminal and storage medium | |
CN105184571A (en) | Payment authentication system based on combination of hand veins and multiple fingerprints | |
CN109840494A (en) | Identity identifying method, device, computer program, storage medium and electronic equipment | |
Fallahi et al. | Brainnet: Improving brainwave-based biometric recognition with siamese networks | |
CN108108975A (en) | A kind of payment authentication method and terminal based on electrocardiogram and vocal print | |
Białas et al. | Multifactor authentication system using simplified EEG brain–computer interface | |
CN115906028A (en) | User identity verification method and device and self-service terminal | |
Sohankar et al. | Systematic analysis of liveness detection methods in biometrie security systems | |
CN109918883A (en) | A kind of auth method of the biocompatibility characteristics based on Freehandhand-drawing track |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180601 |
|
RJ01 | Rejection of invention patent application after publication |