WO2016058540A1 - 身份验证方法、装置和存储介质 - Google Patents

身份验证方法、装置和存储介质 Download PDF

Info

Publication number
WO2016058540A1
WO2016058540A1 PCT/CN2015/091975 CN2015091975W WO2016058540A1 WO 2016058540 A1 WO2016058540 A1 WO 2016058540A1 CN 2015091975 W CN2015091975 W CN 2015091975W WO 2016058540 A1 WO2016058540 A1 WO 2016058540A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
verification
sample
lip
image
Prior art date
Application number
PCT/CN2015/091975
Other languages
English (en)
French (fr)
Inventor
陈敏浩
涂杰
王玉叶
陈辉
陈荣
郑兴
Original Assignee
腾讯科技（深圳）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Priority claimed from CN201410545593.XA external-priority patent/CN104967593B/zh
Priority claimed from CN201410647776.2A external-priority patent/CN104966086B/zh
Priority claimed from CN201410779600.2A external-priority patent/CN104965724A/zh
Application filed by 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Publication of WO2016058540A1 publication Critical patent/WO2016058540A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/08—Network architectures or network communication protocols for network security for authentication of entities
    • H04L63/0861—Network architectures or network communication protocols for network security for authentication of entities using biometrical features, e.g. fingerprint, retina-scan
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/60—Context-dependent security
    • H04W12/65—Environment-dependent, e.g. using captured environmental data

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to an identity verification method, apparatus, and storage medium.
  • the prior art proposes many authentication schemes, such as a password-based authentication scheme, a fingerprint-based authentication scheme, or a mobile phone binding-based authentication scheme, and the like.
  • a password-based authentication scheme such as a password-based authentication scheme, a fingerprint-based authentication scheme, or a mobile phone binding-based authentication scheme, and the like.
  • it is generally required to save the password or fingerprint input by the user as sample data in advance, and then obtain the password or fingerprint input by the user in the authentication phase, and input the password or fingerprint of the user. Matches with the password or fingerprint sample data entered in advance. If it matches, it determines that the user's identity is legal and the verification is passed. Otherwise, if it does not match, the user's identity is invalid and the verification fails.
  • the security of the existing solution is not high. For example, once the password or fingerprint data is stolen by an illegal intruder, the illegal intruder can easily steal the information data in the account, and the information security cannot be guaranteed. The effect is not good.
  • the embodiment of the invention provides an identity verification method, device and storage medium to solve the problem that information security cannot be guaranteed in the prior art.
  • the authentication method provided by the embodiment of the present invention includes: the verification device provides the verification prompt information to the verification object; the verification device obtains the identity verification information, and the identity verification information is generated by the verification object according to the verification prompt information; the verification device sets the identity verification information and the verification prompt The information samples of the information are compared; when the matching degree between the identity verification information and the information sample of the verification prompt information satisfies a preset value, the verification object passes the verification.
  • the identity verification apparatus includes: one or more processors and a storage medium storing operation instructions, when executing an operation instruction in the storage medium, the processor executes: providing verification prompt information to the verification object; The authentication information is generated by the verification object according to the verification prompt information; the identity verification information is compared with the information sample of the verification prompt information; and the matching degree between the identity verification information and the information sample of the verification prompt information meets a preset value , the verification object is verified
  • the non-transitory computer readable storage medium provided by the embodiment of the present invention has computer executable instructions stored thereon, and when the executable instructions are run in the computer, executing: providing verification prompt information to the verification object; The authentication information is generated by the verification object according to the verification prompt information; the identity verification information is compared with the information sample of the verification prompt information; and the matching degree between the identity verification information and the information sample of the verification prompt information meets a preset value , the verification object is verified.
  • FIG. 1 is a schematic flowchart of an identity verification method according to an embodiment of the present invention.
  • FIG. 2 is a schematic flowchart of an identity verification method according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of an acquisition scenario of identity verification information according to an embodiment of the present disclosure
  • FIG. 4 is a network framework diagram of an identity verification method according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a scenario of an identity verification method according to an embodiment of the present disclosure.
  • FIG. 6 is a schematic flowchart of an identity verification method according to an embodiment of the present disclosure.
  • FIG. 7 is a schematic flowchart of an identity verification method according to an embodiment of the present disclosure.
  • FIG. 8 is a schematic structural diagram of an identity verification apparatus according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of an identity verification apparatus according to an embodiment of the present invention.
  • FIG. 10 is a schematic flowchart of an identity verification method according to an embodiment of the present disclosure.
  • FIG. 11 is a schematic flowchart of a method for setting reference audio data and reference text data according to an embodiment of the present invention
  • FIG. 12 is a schematic flowchart of an identity verification method according to an embodiment of the present disclosure.
  • FIG. 13 is a schematic diagram of an interface involved in an identity verification method according to an embodiment of the present disclosure.
  • FIG. 14 is a schematic diagram of another interface involved in an identity verification method according to an embodiment of the present disclosure.
  • 15 is a schematic diagram of model points involved in an identity verification method according to an embodiment of the present invention.
  • 16 is a flowchart of a method for calculating a probability involved in an identity verification method according to an embodiment of the present invention
  • FIG. 17 is a schematic diagram of matching calculation involved in an identity verification method according to an embodiment of the present invention.
  • FIG. 18 is a schematic flowchart of an identity verification method according to an embodiment of the present invention.
  • FIG. 19 is a schematic flowchart of an identity verification method according to an embodiment of the present invention.
  • FIG. 20 is a schematic flowchart of an identity verification method according to an embodiment of the present disclosure.
  • FIG. 21 is a schematic structural diagram of an identity verification apparatus according to an embodiment of the present disclosure.
  • FIG. 22 is a schematic structural diagram of an identity verification apparatus according to an embodiment of the present invention.
  • FIG. 23 is a schematic structural diagram of an identity verification apparatus according to an embodiment of the present invention.
  • FIG. 24 is a schematic structural diagram of an identity verification apparatus according to an embodiment of the present invention.
  • FIG. 25 is a schematic structural diagram of an identity verification apparatus according to an embodiment of the present invention.
  • FIG. 1 is a schematic flowchart of an identity verification method according to an embodiment of the present invention.
  • the method may be applied to a device such as a terminal or a server.
  • the terminal can be a device such as a mobile phone, a tablet computer or a personal computer.
  • the method may include steps 101 to 103.
  • step 101 the verification device provides verification prompt information to the verification object.
  • step 102 the verification device obtains the identity verification information, and the identity verification information is verified by the verification object according to the verification prompt. Information generation.
  • step 103 the verification device compares the identity verification information with the information sample of the verification prompt information; when the matching degree between the identity verification information and the information sample of the verification prompt information satisfies a preset value, the verification object passes the verification.
  • the identity verification device may be a device such as a terminal or a server.
  • the terminal may be a device such as a mobile phone, a tablet computer or a personal computer.
  • An authentication method includes: acquiring a data processing request; acquiring identity verification information according to the data processing request, where the identity verification information includes image information, sound information, and associated information of the image information and the sound information; Whether the authentication information matches the preset information sample; if so, the data processing request is passed; if not, the data processing request is rejected.
  • the specific process of the identity verification method may include steps 201 to 203.
  • step 201 a data processing request is obtained.
  • the data processing request sent by the terminal may be specifically received.
  • the identity verification device is a terminal, specifically, a user-triggered data processing request may be received, and the like.
  • the data processing request may specifically be a virtual resource transfer request, such as a transfer request, a payment request, or a repayment request, or the like, or the data processing request may also make other file processing requests or service requests, etc., the data processing.
  • the request may also carry information such as the identity of the user, and details are not described herein.
  • the identity verification information is acquired according to the received data processing request, wherein the identity verification information includes information such as image information, sound information, and associated information of the image information and the sound information.
  • association information of the image information and the sound information refers to a positional relationship and/or a temporal relationship between the image in the image information and the sound in the sound information, for example, between a certain frame image and a sound at a certain moment. Time interval, and so on.
  • the way to obtain the authentication information is also different.
  • the identity verification device is a server, at this time, the identity information acquisition request may be sent to the terminal according to the data processing request, and then the terminal verification information returned by the terminal according to the identity information acquisition request is received.
  • the identity verification device is a terminal, at this time, the identity verification information input by the user may be specifically received according to the data processing request.
  • the authentication information may be carried by multiple information formats, for example, may be video information, or may be independent image information and voice information, that is, the step “acquiring authentication information according to the received data processing request” may include The following two ways.
  • the identity information acquisition request may be sent to the terminal according to the data processing request, and then the receiving terminal returns the corresponding video information according to the identity information acquisition request, and extracts corresponding information from the video information.
  • Image information, sound information, and associated information of the image information and sound information may be sent to the terminal according to the data processing request, and then the receiving terminal returns the corresponding video information according to the identity information acquisition request, and extracts corresponding information from the video information.
  • the identity verification device is a terminal
  • the video information input by the user may be received according to the data processing request, and corresponding image information, sound information, and the image information are extracted from the video information.
  • the associated information of the sound information is a terminal
  • the image information and the sound information are respectively acquired according to the data processing request, and the association information of the image information and the sound information is determined according to the image information and the sound information.
  • the identity information acquisition request may be sent to the terminal according to the data processing request, and then the image information and the sound information returned by the terminal according to the identity information acquisition request are received, according to the image information.
  • sound information determines association information of the image information and the sound information.
  • the image information and the sound information respectively input by the user may be received according to the data processing request, and the association between the image information and the sound information is determined according to the image information and the sound information. information.
  • the user may input the text directly, or display the text for the user to read aloud, wherein the displayed text is the corresponding text of the voice content in the preset information sample, that is, the step “acquires the sound”.
  • the information “specifically can include:
  • step 203 it is determined whether the identity verification information matches the preset information sample, and if it matches, the data processing request is passed; otherwise, if not, the data processing request is rejected.
  • the information sample may include an image feature sample, a voiceprint feature sample, and an associated information sample of the image information and the sound information, and the step “determining whether the identity verification information matches the preset information sample” may be specifically as follows:
  • Extracting image features from the image information, and extracting voiceprint features from the sound information respectively determining the image features, voiceprint features and the associated information and preset image feature samples, voiceprint feature samples, and associated information Whether the sample matches (ie, determining whether the image feature matches the preset image feature sample, determining whether the voiceprint feature matches the preset voiceprint feature sample, and determining whether the association information matches the preset associated information sample); If the number of matched items exceeds a preset threshold, determining that the authentication information matches a preset information sample; if the number of matched items does not exceed a preset threshold, determining the identity verification information and the preset information sample Mismatch.
  • the preset threshold may be set according to the requirements of the actual application, and the threshold is greater than or equal to 2.
  • the threshold is set to 2
  • voiceprint features and voiceprint feature samples match
  • image features and image feature samples match
  • associated information and associations “Information sample matching”, or "the voiceprint feature and the voiceprint feature sample match
  • the associated information matches the associated information sample
  • the threshold is set to be greater than or equal to 2, if "the image feature and the image feature sample match, and the voiceprint feature and the voiceprint feature sample match", or “the image feature and the image feature sample match, and the association
  • the information matches the associated information sample, or, "the voiceprint feature and the voiceprint feature sample match, and the associated information matches the associated information sample", or "the image feature and the image feature sample match, and the voiceprint feature and the voiceprint feature If the sample matches, and the associated information matches the associated information sample, it is determined that the authentication information matches the preset information sample; otherwise, the identity verification is determined. The information does not match the preset information sample.
  • the threshold may be set to 3, that is, all the matches are determined, and the identity verification information is determined to match the preset information sample. Otherwise, if any one of the information does not match, the identity verification information is determined.
  • the preset information samples do not match, and so on, and will not be described here.
  • the voice content may also be verified, that is, the information sample may further include a voice content sample.
  • the step “determines whether the identity verification information and the preset information sample are Matching can include:
  • the pattern of the feature, the sample of the voice content, and the sample of the associated information match (ie, determine whether the image feature matches the preset image feature sample, determine whether the voiceprint feature matches the preset voiceprint feature sample, the voice content and the preset voice Whether the content samples match, and determining whether the association information matches the preset association information sample; if the number of matched items exceeds a preset threshold, determining that the identity verification information matches the preset information sample; if the matching If the number of items does not exceed the preset threshold, it is determined that the authentication information does not match the preset information sample.
  • the preset threshold may be set according to the requirements of the actual application, and the threshold is greater than or equal to 2, for example, may be set to 2, 3, or 4, or may be set to a certain range, such as greater than or equal to 2, greater than or equal to 3, and the like. See the previous description and will not repeat them here.
  • the information sample may be set according to the requirements of the actual application or the preference of the user, that is, before obtaining the data processing request, the identity verification may further include acquiring and saving the information sample.
  • information samples can be obtained and saved in the following two ways.
  • Obtaining a video information sample of the user extracting corresponding image information samples, sound information samples, and associated information samples of the image information and the sound information from the video information samples, extracting image feature samples from the image information samples, and extracting
  • the voiceprint feature sample is extracted from the sound information sample, and the image feature sample, the voiceprint feature sample, and the associated information sample of the image information and the sound information are saved.
  • the voice content sample may also be extracted from the sound information sample and the voice content sample may be saved.
  • the voice content sample may also be extracted from the sound information sample and the voice content sample may be saved.
  • the sound information sample may be directly input by the user, or the text may be displayed for the user to read aloud, that is, the step of “acquiring the sound information sample” may include:
  • the terminal can take the action of the user “snap finger” and display the words “open sesame opening”, record the sound information when the user reads the four words, and then provide the information to the server, by the server. Extracting the image features, voiceprint features, voice content, and the relationship between the image information and the sound information from the information, for example, the association relationship may be the shooting time of the action of "sounding finger” and the user reading "the sesame opening door” The time interval of four words, and so on, and then the extracted information is saved as a sample of information as an authenticated password, such as a user's payment password, a transfer password, or a password for other services, and the like.
  • an authenticated password such as a user's payment password, a transfer password, or a password for other services, and the like.
  • the data processing may be performed according to the data processing request, for example, operations such as transferring, paying, or opening a folder, and details are not described herein.
  • the data processing request is acquired, and the identity verification information is obtained according to the data processing request, wherein the identity verification information includes image information, sound information, and associated information of the image information and the sound information; Whether the authentication information matches the preset information sample, if it matches, the data processing request is passed, otherwise, if it does not match, the data processing request is rejected; since in this scheme, the image information and the sound information may be The combination is used as the authentication information, so that the authentication information can include not only the image information and the sound information, but also the associated information of the image information and the sound information. Therefore, only a single password or fingerprint is used in the prior art. In terms of information, the complexity of the verification information is increased, and the difficulty of stealing the identity verification information by the illegal intruder is improved. Therefore, compared with the prior art, the effect of the identity verification can be enhanced, and the security of the information can be improved.
  • An embodiment of the present invention provides another method for authenticating an identity.
  • the identity verification device is specifically configured as a server.
  • the communication system to which the authentication method is applied may include a server and a terminal.
  • the terminal is configured to send a data processing request to the server, and after receiving the identity information obtaining request sent by the server, obtain the identity verification information, and send the identity verification information to the server.
  • the identity verification information includes information such as image information, sound information, and associated information of the image information and the sound information.
  • association information of the image information and the sound information refers to a positional relationship and/or a temporal relationship between the image in the image information and the sound in the sound information, for example, between a certain frame image and a sound at a certain moment. Time interval, and so on.
  • a server configured to receive a data processing request sent by the terminal, send an identity information obtaining request to the terminal according to the data processing request, and then the receiving terminal determines the identity verification information and the preset information according to the identity verification information returned by the identity information obtaining request. Whether the samples match; if so, the request is processed by the data; if not, the data processing request is rejected.
  • the preset information sample may be stored in the server, or may be retained in other devices, and the information in the information sample may be stored in the same entity, or may be separately stored in different entities, for example, Referring to FIG. 5, the server may include a mobile payment server, an image feature server, a voiceprint feature server, a voice content server, etc., and the image feature samples may be saved in the image feature server, and the image feature server verifies whether the image features are Image feature sample matching; the voiceprint feature sample can be saved in the voiceprint feature server, and the voiceprint feature server can verify whether the voiceprint feature matches the voiceprint feature sample; the voice content sample can be saved in the voice a content server, and the voice content server verifies whether the voice content matches the voice content sample; saves the associated information sample of the image information and the sound information in the mobile payment server, and the mobile payment server verifies whether the associated information matches the associated information sample , etc., are not listed here.
  • the description will be made by taking an example in which these information samples are all stored in
  • the payment request is described as a data processing request.
  • an identity authentication method provided by this implementation may include steps 601 to 608.
  • step 601 the terminal sends a payment request to the server, for example, may instruct to transfer a certain amount of money into an account.
  • the payment request may carry the identity of the user.
  • the identity of the user may specifically include the user's bank account number, email account number, user name, and/or user nickname.
  • step 602 after receiving the payment request, the server sends an identity information acquisition request to the terminal according to the payment request.
  • step 603 after receiving the identity information acquisition request sent by the server, the terminal acquires the identity verification information of the user according to the identity information acquisition request, and sends the identity verification information to the server.
  • the authentication information includes information such as image information, sound information, and associated information of the image information and the sound information.
  • association information of the image information and the sound information refers to a positional relationship and/or a temporal relationship between the image in the image information and the sound in the sound information, for example, between a certain frame image and a sound at a certain moment. Time interval, and so on.
  • the prompt information may be generated according to the identity information obtaining request, where the prompt information indicates that the user captures a video, or separately captures a series of pictures and records a sound, and receives a video recorded by the user according to the indication of the prompt information, or Images and information are provided to the server as authentication information.
  • the server extracts information such as image information, voice information, and associated information of the image information and the sound information from the identity verification information.
  • the received verification information is a piece of video information
  • the corresponding image information, the sound information, and the associated information of the image information and the sound information are extracted from the video information, such as correspondingly in a certain frame image. There are sounds, and so on.
  • association information of the image information and the sound information may be determined according to the image information and the sound information, such as determining a time interval when the image information and the sound information are acquired. For example, if the user records a certain photo and then records a sound at intervals of 1 second, it can be determined that the time interval is 1 s, and so on, and details are not described herein again.
  • step 605 the server extracts image features from the image information, respectively, and extracts voiceprint features from the sound information.
  • the information sample may further include a voice content sample
  • the voice content may also be extracted from the sound information at this time.
  • step 606 the server acquires corresponding preset information samples according to the identity of the user, and determines the map separately. Whether the feature, the voiceprint feature, and the associated information match the information sample, if the number of matched items exceeds a preset threshold, determining that the identity verification information matches the preset information sample, indicating that the verification is passed, and then performing step 607 If the number of matched items exceeds the preset threshold, it is determined that the authentication information does not match the preset information sample, indicating that the verification fails, and then step 608 is performed.
  • the information sample may include an image feature sample, a voiceprint feature sample, and a related information sample, and may further include a voice content sample, which may be specifically as follows:
  • Determining whether the image feature, the voiceprint feature, and the associated information match the preset image feature sample, the voiceprint feature sample, and the associated information sample that is, determining whether the image feature matches the preset image feature sample, and determining the voiceprint Whether the feature matches the preset voiceprint feature sample, and determines whether the association information matches the preset association information sample.
  • the voice content needs to be verified, then the voice content and the pre-determination need to be determined at this time.
  • step 607 Whether the matched voice content samples match, if the number of matching items in the above items exceeds the preset threshold, it indicates that the authentication information matches the preset information sample, and then step 607 is performed, if the number of matched items does not exceed The preset threshold indicates that the authentication information does not match the preset information sample, and then step 608 is performed.
  • the preset threshold may be set according to the requirements of the actual application, and the threshold is greater than or equal to 2. For details, refer to the previous embodiment, and details are not described herein again.
  • the information sample may be set according to the requirements of the actual application or the user's preference.
  • For the specific setting method refer to the embodiment shown in FIG. 2, and details are not described herein again.
  • step 607 the server passes the verification of the payment request, and at this time, the response to the verification can be returned to the terminal.
  • the server can perform the payment processing according to the payment request, for example, the money of a certain amount can be transferred to an account, and the like, and the processing result can be notified to the terminal, and details are not described herein again.
  • step 608 the server rejects the verification of the payment request, and at this point, a response to the verification failure may be returned to the terminal.
  • the identity verification device is specifically a terminal
  • the data processing request is specifically a terminal application startup request as an example.
  • an identity authentication method provided in this embodiment may include steps 701 to 707.
  • the terminal receives a user-initiated terminal application activation request, for example, may instruct to open a terminal application, such as an album or a personal document, and the like.
  • step 702 after receiving the terminal application startup request, the terminal generates and displays prompt information according to the terminal application startup request, where the prompt information indicates that the user captures a video, or separately takes a series of pictures and records a sound, and Receiving a video, or a picture and information recorded by the user according to the indication of the prompt information, and obtaining identity verification information.
  • the authentication information includes information such as image information, sound information, and associated information of the image information and the sound information.
  • association information of the image information and the sound information refers to a positional relationship and/or a temporal relationship between the image in the image information and the sound in the sound information, for example, between a certain frame image and a sound at a certain moment. Time interval, and so on.
  • step 703 the terminal extracts information such as image information, sound information, and associated information of the image information and the sound information from the identity verification information.
  • step 704 the terminal extracts image features from the image information, respectively, and extracts voiceprint features from the sound information.
  • the information sample may further include a voice content sample
  • the voice content may also be extracted from the sound information at this time.
  • step 705 the terminal acquires corresponding preset information samples, and determines whether the image feature, the voiceprint feature, and the associated information match the information sample respectively. If the number of matching items in the items exceeds a preset threshold, then It is determined that the authentication information matches the preset information sample, and then step 706 is executed. If the number of matched items does not exceed the preset threshold, it is determined that the identity verification information does not match the preset information sample, and then step 707 is performed.
  • the information sample may include an image feature sample, a voiceprint feature sample, and a related information sample, and may further include a voice content sample, which may be specifically as follows:
  • Determining whether the image feature, the voiceprint feature, and the associated information match the preset image feature sample, the voiceprint feature sample, and the associated information sample that is, determining whether the image feature matches the preset image feature sample, and determining the voiceprint Whether the feature matches the preset voiceprint feature sample, and determines whether the association information matches the preset association information sample.
  • the voice content needs to be verified, then the voice content and the pre-determination need to be determined at this time.
  • step 706 Whether the matched voice content samples match, if the number of matching items in the above items exceeds the preset threshold, it indicates that the authentication information matches the preset information sample, and then step 706 is performed, if the matching item does not exceed the pre-predetermined Setting the threshold indicates that the authentication information does not match the preset information sample, and then step 707 is performed.
  • the preset threshold may be set according to the requirements of the actual application, and the threshold is greater than or equal to 2. For details, refer to the previous embodiment, and details are not described herein again.
  • the information sample may be set according to the requirements of the actual application or the user's preference.
  • For the specific setting method refer to the embodiment shown in FIG. 6 , and details are not described herein again.
  • step 706 the terminal initiates a request through the terminal, and at this time, the notification of the verification pass can be displayed to the user.
  • the terminal can start a corresponding terminal application according to the terminal application launch request, such as opening an album or a personal document, and the like.
  • step 707 the terminal rejects the terminal application initiation request, and at this time, a notification of the verification failure may be displayed to the user.
  • An embodiment of the present invention further provides an identity verification apparatus.
  • the identity verification apparatus includes a request acquisition unit 801, an information acquisition unit 802, and a verification unit 803, as follows:
  • the request obtaining unit 801 is configured to acquire a data processing request.
  • the request obtaining unit 801 may specifically receive a data processing request sent by the terminal.
  • the request acquisition unit 801 may specifically receive a data processing request triggered by the user, and the like.
  • the data processing request may specifically be a virtual resource transfer request, such as a transfer request, a payment request, or a repayment request, or the like, or the data processing request may also make other file processing requests or service requests, etc., the data processing.
  • the request may also carry information such as the identity of the user, and details are not described herein.
  • the information obtaining unit 802 is configured to acquire identity verification information according to the data processing request, where the identity verification information includes information such as image information, sound information, and associated information of the image information and the sound information.
  • association information of the image information and the sound information refers to a positional relationship and/or a temporal relationship between the image in the image information and the sound in the sound information, for example, between a certain frame image and a sound at a certain moment. Time interval, and so on.
  • the verification unit 803 is configured to determine whether the identity verification information matches the preset information sample, and if yes, pass the data processing request, and if not, reject the data processing request.
  • the way to obtain the authentication information is different, for example, as follows:
  • the information acquisition unit 802 may specifically send an identity information acquisition request to the terminal according to the data processing request, and then receive the identity verification information returned by the terminal according to the identity information acquisition request.
  • the information acquisition unit 802 may specifically receive the identity verification information input by the user according to the data processing request.
  • the identity verification information may be carried in a plurality of information formats, for example, the video information, or the independent image information and the voice information, that is, the information obtaining unit 802 may obtain the identity verification information in two ways, specifically See method embodiment. I will not repeat them here.
  • the user may input the text directly, or display the text for the user to read aloud.
  • the displayed text is the corresponding text of the voice content in the preset information sample, that is, the operation “acquires the sound”.
  • the information can be:
  • the information acquiring unit 802 may be configured to: obtain the voice content in the information sample according to the data processing request, display the voice content in the form of a text, and prompt the user to read the text, and record the voice when the user reads the text. Get sound information.
  • the information sample may include image feature samples, voiceprint feature samples, and associated information samples of image information and sound information, etc., where:
  • the verification unit 803 is specifically configured to: separately extract image features from the image information, and extract voiceprint features from the sound information, respectively determine the image features, voiceprint features, and the associated information and preset image feature samples. Whether the voiceprint feature sample and the associated information sample match (ie, determining whether the image feature matches the preset image feature sample, determining whether the voiceprint feature matches the preset voiceprint feature sample, and determining the associated information and the pre-determination If the number of matching items exceeds a preset threshold, determining that the authentication information matches the preset information sample; if the number of matching items does not exceed the preset threshold, determining the The authentication information does not match the preset information sample.
  • the voice content may also be verified, that is, the information sample may further include a voice content sample, then:
  • the verification unit 803 is specifically configured to separately extract image features from the image information, and extract voiceprint features and voice content from the sound information, and determine whether the image features, voiceprint features, voice content, and the associated information are respectively determined.
  • Pre-set image feature samples, voiceprint feature samples, voice content samples, and associated information sample matching ie, determining whether the image features match preset image feature samples, and determining whether the voiceprint features match the preset voiceprint feature samples Whether the voice content matches the preset voice content sample, and determining the associated information and the preset associated information sample is If the number of matching items exceeds the preset threshold, determining that the authentication information matches the preset information sample; if the number of matching items does not exceed the preset threshold, determining the authentication information and The preset information samples do not match.
  • the preset threshold may be set according to the requirements of the actual application, and the threshold is greater than or equal to 2. For details, refer to the previous embodiment, and details are not described herein again.
  • the information sample may be set according to the requirements of the actual application or the user's preference.
  • the identity verification apparatus may further include a sample obtaining unit 804, as follows:
  • the sample obtaining unit 804 is configured to acquire and save an information sample.
  • the information sample may be obtained and saved in another manner.
  • the method embodiment which is not described here.
  • the voice content sample may also be extracted from the sound information sample and the voice content sample may be saved.
  • the user may input the text directly, or display the text for the user to read aloud, that is:
  • the sample obtaining unit 804 may be specifically configured to generate characters according to a preset policy, display the characters, and prompt the user to read the characters, and record the sounds when the user reads the characters to obtain a sound information sample.
  • the sample acquisition unit 804 can capture the action of the user “snap finger” and display the words “open sesame opening”, when the user reads the four words.
  • the sound information is then provided to the server, and the server extracts image features, voiceprint features, voice content, and association relationship between the image information and the sound information from the information.
  • the association relationship may be “sounding finger” "The shooting time of this action is the time interval between when the user reads the words "Sesame Opens", and so on, and then the extracted information is saved as a sample of information, as an authentication password, for example, as a user. Pay passwords, transfer passwords, or passwords for other businesses, and more.
  • the data processing may be performed according to the data processing request, for example, performing operations such as transferring, paying, or opening a folder, that is, the identity verification device may further include data.
  • the processing unit is as follows:
  • the data processing unit is configured to perform data processing according to the data processing request after the data processing request by the user.
  • the foregoing various units may be implemented as a separate entity, and may be implemented in any combination, and may be implemented as the same entity or a plurality of entities.
  • the foregoing various units refer to the foregoing embodiments, and details are not described herein again.
  • the identity verification device may be a terminal or a server, and the terminal may be a device such as a mobile phone, a tablet computer or a PC.
  • the image information and the sound information can be combined as the authentication information, so that the identity verification information can include not only the image information and the sound information, but also the associated information of the image information and the sound information, and therefore, Compared with the prior art, only a single password or a fingerprint is used, the complexity of the verification information is increased, and the difficulty of stealing the authentication information by the illegal intruder is improved. Therefore, compared with the prior art, It can enhance the effectiveness of authentication and improve the security of information.
  • the embodiment of the present invention further provides an identity verification system, which may include any identity verification device provided by an embodiment of the present invention.
  • the identity verification device may be a terminal or a server, and the identity verification is performed below.
  • the device is a terminal and a server respectively, and is briefly described.
  • the communication system may include a terminal in addition to the server, as follows:
  • the terminal is configured to send a data processing request to the server, and after receiving the identity information obtaining request sent by the server, obtain the identity verification information, and send the identity verification information to the server.
  • the identity verification information includes information such as image information, sound information, and associated information of the image information and the sound information.
  • association information of the image information and the sound information refers to a positional relationship and/or a temporal relationship between the image in the image information and the sound in the sound information, for example, between a certain frame image and a sound at a certain moment. Time interval, and so on.
  • a server configured to receive a data processing request sent by the terminal, send an identity information obtaining request to the terminal according to the data processing request, and then the receiving terminal determines the identity verification information and the preset information according to the identity verification information returned by the identity information obtaining request. Whether the samples match; if so, the request is processed by the data; if not, the data processing request is rejected.
  • the preset information samples may be stored in the server, or may be retained in other devices, and the information in the information samples may be stored in the same entity or in different entities.
  • the communication system can include at least a terminal as follows:
  • the terminal is configured to receive a data processing request triggered by the user, send an identity information obtaining request to the terminal according to the data processing request, and then the receiving terminal determines the identity verification information and the preset information according to the identity verification information returned by the identity information obtaining request. Whether the samples match; if so, the request is processed by the data; if not, the data processing request is rejected.
  • the communication system can include any of the identity verification devices provided by the embodiments of the present invention
  • the beneficial effects that can be implemented by any of the identity verification devices provided by the embodiments of the present invention can also be implemented.
  • the embodiment of the invention provides a method for identity verification, and the execution body of the method may be a terminal or a server.
  • the terminal may be a mobile phone, a tablet computer, a computer, or the like.
  • subsequent data processing such as virtual resource transfer, and switching of working status, such as lock screen status and screen unlock status, can be performed.
  • the operation mode is switched from the lock screen state to the screen unlock state as an example, and the terminal is used as a mobile phone as an example for detailed description of the solution.
  • the situation of other terminals is similar, and the embodiment is not described in detail.
  • the identity verification method provided by the present invention includes steps 1001 to 1003.
  • step 1001 the first audio data collected by the audio collection device is acquired.
  • the reference audio data and the reference text data may be preset and stored, and the setting process may be that the user inputs a voice to the audio collection device (such as a microphone), and then the terminal obtains The audio data of the speech is taken, and the audio data is converted into text data as the reference audio data and the reference text data.
  • the audio collection device such as a microphone
  • the user can input the voice to the audio collection device at this time, and the content of the voice can be the same as the content of the input voice when setting the reference audio data by itself, for example, for saying "open sesame door” to the microphone.
  • the audio collection device collects the audio data input by the user (ie, the first audio data), and the terminal can acquire the audio data.
  • the lock screen interface may not be displayed, the screen is closed, and audio data is acquired.
  • the terminal may generate an unlock request after acquiring the audio data input by the user, and carry the audio data in the unlock request, and may also carry the identifier of the terminal, such as IMEI, to send the unlock request to the server.
  • the server may obtain the audio data collected by the audio collection device as described above from the unlock request.
  • step 1002 text conversion is performed on the first audio data to obtain corresponding first text data.
  • the terminal or the server may pre-store an audio text conversion database, and store audio features corresponding to each character. Based on the audio features of the first audio data, the corresponding characters may be found in the audio text conversion database, and the characters are pressed. The sequences are combined so that the audio data can be converted into the first text data.
  • step 1003 if the first text data is the same as the pre-stored reference text data, and the first audio data and the pre-stored reference audio data satisfy a preset proximity condition, the first working state is switched to the second work. status.
  • the first working state and the second working state may be any set working state, which is not limited in this embodiment.
  • the terminal may compare the converted text data with the stored reference text data to determine whether they are the same, and at the same time, The audio data is compared with the stored reference audio data to determine whether the preset proximity condition is met.
  • the proximity condition may be arbitrarily set according to requirements.
  • the proximity condition may be the frequency of two audio data. The curve satisfies a certain degree of approximation, for example, the sum of the absolute values of the difference values of the two frequency curves at each point is less than a preset threshold.
  • the working state switching is triggered, otherwise, the working state switching is not triggered. In this way, the operation such as unlocking by voice triggering can effectively improve the security of the terminal with respect to the manner of the sliding operation.
  • the server may pre-store the reference text data and the reference audio data corresponding to each terminal, and specifically establish the correspondence between the identifier of the terminal and the reference text data and the reference audio data.
  • the server may obtain the identifier of the terminal from the unlock request, and then search for the reference text data and the reference audio data corresponding to the identifier of the terminal in the corresponding relationship, and then perform the processing of the comparison judgment.
  • the server may send a switching notification to the terminal, so that the terminal switches the first working state to the second working state.
  • step 1003 may include step 10031 to Step 10032.
  • step 10031 if the first reference text data is identical to the first text data in the correspondence between the pre-stored reference text data and the reference audio data, according to the correspondence between the reference text data and the reference audio data, Determining first reference audio data corresponding to the first reference text data.
  • the user can set multiple sets of reference text data and reference audio data, and the terminal can store the reference text data and the reference audio data. Correspondence relationship.
  • the terminal converts the acquired audio data (ie, the first audio data) into text data (ie, the first text data)
  • the terminal may search for the text data in a correspondence relationship between the reference text data and the reference audio data, if a certain reference exists.
  • the text data ie, the first reference audio data
  • the reference audio data corresponding to the reference text data ie, the first reference audio data
  • step 10032 if the first audio data and the first reference audio data satisfy a preset proximity condition, the first working state is switched to the second working state.
  • the terminal may compare the first audio data with the first reference audio data in a similar manner as described above to determine whether to trigger the working state switching.
  • the process of switching the first working state to the second working state may have various situations. Several feasible situations are given below: Case 1, switching the lock screen state to the screen Unlock status; Case 2, switch the application lock status to the application unlock status.
  • the above step 1001 may trigger execution when the application is launched.
  • the embodiment of the present invention further provides a method for setting the reference audio data and the reference text data.
  • the processing flow may be as shown in FIG. 11 and includes steps 1101 to 1103.
  • step 1101 a setting instruction is received to acquire second audio data collected by the audio collection device.
  • the user can operate the terminal to open the audio lock setting page, in which the option of inputting voice can be set.
  • the terminal receives the setting instruction, can activate the audio collection device of the terminal, and start the audio data.
  • the user can input voice to the audio collection device, such as “open sesame”.
  • the audio collection device collects the audio data input by the user (ie, the second audio data), and the terminal can obtain the voice. Audio data.
  • step 1102 text conversion is performed on the second audio data to obtain corresponding second text data.
  • the process is similar to the step 1102, and the corresponding content can be referred to, and the description is not repeated here.
  • the following processing may be performed: displaying the second text data.
  • step 1103 the second audio data and the second text data are stored as reference audio data and reference text data, respectively.
  • the user may confirm the second text data, and correspondingly, the processing of step 1103 may be as follows: if the confirmation instruction is received, the second audio data and the second The text data is stored as reference audio data and reference text data, respectively.
  • the confirmation box may be displayed, and the second text data is displayed in the confirmation box, such as “Open Sesame”, and the confirmation button and the re-enter button are displayed.
  • the confirmation button can be clicked.
  • the terminal receives the confirmation command, and the terminal can store the second audio data and the second text data as the reference audio, respectively. Data and benchmark text data. If the user finds that the displayed second text data does not match the voice input by himself, the user can click the re-enter button, and the trigger will be triggered to re-execute step 1101.
  • the first audio data collected by the audio collection device is acquired, and the first audio data is text-converted to obtain corresponding first text data, if the first text data is the same as the pre-stored reference text data, and The first audio data and the pre-stored reference audio data satisfy a preset approximation condition, and then the first working state is switched to the second working state, so that the working state switching can be triggered without multiple click operations on the screen.
  • the efficiency of the work state switching can be improved.
  • This embodiment will explain in detail the process of voice recognition and the process of voiceprint analysis in the above embodiment.
  • An audio text conversion database may be pre-established in the terminal or the server, and an audio feature (such as frequency, etc.) corresponding to each character may be stored in the audio text conversion database.
  • the voice data input by the user ie, the first audio data
  • the voice data may be divided into a plurality of syllables according to the change of the audio characteristics of the voice data, for example, the user inputs the voice of “opening the door of sesame”.
  • Data can be divided into audio data of "Zhi", "Hemp”, “Open”, and "Door”.
  • the audio data may be matched with the audio features in the audio text conversion database, and if the matched audio features are found, the characters corresponding to the audio features may be further obtained, so that The characters corresponding to the audio data of each syllable can be obtained.
  • the audio features are matched in the audio text conversion database, respectively Find the characters of "Zhi", “Hemp”, “Open”, “Door”.
  • the characters may be sorted and combined according to the order of the corresponding audio data to obtain final text data (ie, the first text data), for example, based on the above-mentioned "Zhi”, “Ma”, “Open”, “Gate” "The character can get the "sesame open” string.
  • the terminal or the server acquires the voice data input by the user (ie, the first audio data) and the pre-stored reference audio data, and divides the voice data and the reference audio data into a plurality of syllables according to their respective audio characteristics, respectively. For example, if the user inputs the voice data of "Open Sesame”, it can be divided into audio data of "Zhi”, “Ma”, “Open”, “Gate”, and the division of the reference audio data is similar. Then, compare whether the number of syllables they divide is the same. If they are different, the speech data fails to match the reference audio data.
  • the speech data can be compared with the corresponding syllables in the reference audio data according to the order of each syllable. .
  • the time points of the preset number (such as 3, 15 or 30, etc.) can be uniformly selected over the duration of the two syllables, and the frequencies of all corresponding time points in the two syllables are calculated. The absolute value of the difference, and the sum of the absolute values of these frequency differences is added as the error degree.
  • the error degree may be compared with a preset error degree threshold. If the error degree is greater than the error degree threshold, the matching failure may be determined. Otherwise, the matching may be determined successfully, if the voice data and the syllable of the reference audio data are If both of the matches are successful, it can be determined that the voice data and the reference audio data satisfy the preset approximation condition.
  • the first syllable in the voice data is the syllable a
  • the first syllable in the reference audio data is the syllable b, which is uniformly selected in the duration of the syllable a and the syllable b for three time points, and the first one of the syllables a is calculated.
  • the absolute value of the frequency difference between the time point and the first time point in the syllable b, the absolute value of the frequency difference between the second time point in the syllable a and the second time point in the syllable b, the third time point in the syllable a The absolute value of the frequency difference at the third time point in the syllable b is added, and the absolute values of the three frequency differences are added as the error degree, and then the subsequent judgment is made based on the error degree.
  • the double judgment is performed through the voice recognition and the voiceprint analysis, and the terminal is unlocked and the like, and the security of the terminal can be effectively improved.
  • FIG. 12 is a flowchart of a method for an identity verification method according to an embodiment of the present invention. This embodiment is exemplified by applying the identity verification method to a terminal. The method can include steps 1201 through 1206.
  • step 1201 lip language prompt information is provided to the verification object.
  • the terminal generates a lip language prompt message.
  • a lip language prompt information library exists in the terminal.
  • the terminal randomly extracts a lip language prompt information from the lip language prompt information base as the lip language prompt information generated by the terminal.
  • the server generates a lip prompt message.
  • a lip prompt information library in the server.
  • the server randomly extracts a lip prompt information from the lip prompt information base as the lip prompt information generated by the server.
  • the server sends the lip prompt information to the terminal.
  • the terminal receives the lip prompt information.
  • the lip prompt information may be randomly generated prompt information; or the lip prompt information is different prompt information in the adjacent verification process.
  • the terminal provides the lip prompt information to the verification object.
  • the lip language prompt information may be provided to the verification object by any one of the following four methods.
  • the terminal displays the lip prompt information in a text form
  • FIG. 13 shows an interface diagram of the lip-speech information in the form of text displayed by the terminal.
  • a lip-speech information 21 in the form of text is displayed on the display interface 20 of the terminal, and the lip-speech information 21 prompts the verified object to read "I want to unlock".
  • the terminal displays the lip prompt information in the form of an image
  • FIG. 14 shows an interface diagram of the lip-speech information in the form of a terminal display image.
  • a lip-speech information 22 in the form of an image is displayed on the display interface 20 of the terminal, and the lip-speech information 22 prompts the object to be verified to read the name "watermelon" of the fruit in the picture.
  • the terminal displays the lip language prompt information in the form of a video
  • a lip-speech message in the form of a video is displayed on the display interface of the terminal, and the lip-speech information prompts the verification object to answer the answer to the question posed in the video.
  • the terminal plays the lip-speech information in the form of audio.
  • the terminal plays the lip-speech information in an audio form, and the lip-speech information prompts the verification object to read the lip-speech information consistent with the audio.
  • step 1202 at least one frame of the image of the verification object is acquired.
  • the terminal collects at least one frame of the image of the verification object.
  • the terminal may acquire at least one frame of the image of the verification object by taking a photo.
  • the terminal can also collect at least one frame of the verification object by capturing a video.
  • step 1203 when a lip change is included in at least one frame of image, a lip image sequence in at least one frame of image is extracted, the lip image sequence being used to characterize a lip change of the verification subject.
  • the terminal detects whether the lip change matches the lip prompt information.
  • the terminal can determine whether there is a face in at least one frame of the collected image.
  • the position of the lip in the face can be determined by the difference between the color of the lip and the skin color.
  • the method of the model point can be used to determine whether the verification object in the image has a lip change.
  • the model point refers to a corner point on the curve of the lip edge.
  • the upper edge of the lip may include 5 model points 222
  • the lower edge of the lip may also include 5 Model point 222.
  • the terminal can determine the position of each point of the lip by using a model point method to form a state of the lip image at a certain moment. In chronological order, the terminal extracts the lip image in each frame of image to form a sequence of lip images. The sequence of lip images is used to characterize lip changes in the subject being validated.
  • step 1204 a probability value of the string matching of the lip image sequence and the lip hint information is calculated.
  • the terminal calculates a probability value of the string matching of the lip image sequence and the lip prompt information.
  • Each lip prompt message corresponds to a character string, and the terminal calculates a probability value of the matching between the extracted lip image sequence and the lip word corresponding to the lip hint information.
  • This step 1204 may include, but is not limited to, steps 1204a through 1204c, please refer to FIG.
  • step 1204a according to the character arrangement in the character string corresponding to the lip prompt information, the prediction model sequence corresponding to the character string is selected in the preset hidden Markov model, and the hidden Markov model includes n Predictive model, each predictive model corresponds to a character, n ⁇ 1.
  • the terminal inputs a character string corresponding to the lip prompt information into a hidden Markov model, and the hidden Markov model includes a prediction model corresponding to different characters.
  • the string corresponding to the lip prompt information is "215" and contains three characters.
  • the terminal selects a prediction model corresponding to the character "2", a prediction model corresponding to the character "1”, and a prediction model corresponding to the character "5" according to the character string "215" in the hidden Markov model to obtain a prediction model. sequence.
  • step 1204b the degree of matching of the lip image sequence with the predicted model sequence is identified.
  • the terminal identifies the degree of matching of the lip image sequence with the predicted model sequence.
  • the matching degree 1 can be calculated to be 90%; after the lip image 25 is input into the prediction model 26 of the character "1", It can be calculated that the matching degree 2 is 95%; after the lip image 27 is input into the prediction model 28 of the character "5", the matching degree 3 can be calculated to be 98%.
  • step 1204c a probability value of the lip image sequence matching the character string is calculated based on the matching degree.
  • the terminal calculates a probability value of the lip image sequence matching the character string according to the matching degree.
  • step 1205 it is detected whether the probability value is greater than a preset threshold.
  • the preset threshold is 80%, 84%>80%, so the probability value is greater than the preset threshold, that is, the lip change matches the lip prompt information.
  • step 1206 if the lip change matches the lip prompt information, it is determined that the verification object passes the verification.
  • the terminal determines that the verification object passes the identity verification; if the lip change does not match the lip prompt information, the terminal determines that the verification object does not pass the authentication.
  • the identity verification method provides lip-alert information to the verified object, acquires at least one frame image of the verification object, and detects lip changes when at least one frame image includes a lip change. Whether the information is matched with the lip prompt information; if the lip change matches the lip prompt information, it is determined that the verification object is authenticated. It solves the problem that if an illegal user puts a photo of a legitimate user in front of his face, he or she can still pass the identity authentication, resulting in a security risk.
  • the lip prompt information can be randomly generated prompt information, it has the same security level as the verification code verification mechanism.
  • the identity verification method provided in this embodiment further extracts a lip image sequence in at least one frame image, wherein the lip image sequence is used to represent a lip change of the verification object; and the lip image sequence is calculated to correspond to the lip prompt information.
  • the method may further include:
  • a prediction model also known as a hidden Markov model, is pre-established, which includes a prediction model of the lip variation corresponding to each character.
  • the method may include, but is not limited to, sub-steps a) through c).
  • step a at least one frame image of the lip change corresponding to the different characters fed back by the identified object is acquired, n ⁇ 1.
  • the terminal can perform statistics by obtaining a large number of samples.
  • step b a sequence of lip images corresponding to each character in at least one frame of image is extracted.
  • the terminal For each sample, the terminal extracts the lip images in each frame of image in chronological order to form a sequence of lip images.
  • the sequence of lip images is used to characterize lip changes in the subject.
  • step c) a prediction model of the lip change corresponding to each character is statistically obtained according to the lip image sequence corresponding to each character.
  • FIG. 18 is a flowchart of a method for identifying a living body according to still another embodiment of the present invention. This embodiment is exemplified by applying the living body identification method to a server. The method includes steps 1802 through 1808.
  • step 1802 at least one frame of image transmitted by the terminal is received.
  • step 1804 the lip prompt information provided by the terminal to the authenticated object is obtained.
  • step 1806 when a lip change is included in at least one frame of image, it is detected whether the lip change matches the lip prompt information.
  • step 1808 the detection result is fed back to the terminal.
  • the identity verification method receives the at least one frame image sent by the terminal, acquires the lip prompt information provided by the terminal to the verification object, and detects the lip when at least one frame image includes a lip change. Whether the change of the part and the lip-speech information match; the feedback result is fed back to the terminal; and if the illegal user puts the photo of the legitimate user in front of the face, the user can still pass the identity authentication, resulting in a security risk.
  • FIG. 19 a flow chart of a method for identifying a living body according to still another embodiment of the present invention is shown.
  • the example is exemplified by applying the living body identification method to the terminal and the server.
  • the method includes steps 1901 to 1909.
  • step 1901 the server generates lip prompt information.
  • the server generates a lip prompt message.
  • the lip prompt information may be randomly generated prompt information.
  • the server randomly extracts a lip prompt information from the lip prompt information base as the lip prompt information generated by the server.
  • the lip prompt information can be text, picture or video, and each lip prompt message corresponds to a string.
  • step 1902 the server sends the lip prompt information to the terminal.
  • the server sends the lip prompt information to the terminal.
  • step 1903 the terminal provides lip language prompt information to the authenticated object.
  • the terminal provides lip language prompt information to the verification object.
  • the lip language prompt information may be provided to the verification object by any one of the following four methods.
  • the terminal displays the lip prompt information in a text form
  • a lip-speech message in the form of text is displayed on the display interface of the terminal, and the lip-speech information prompts the verification object to read the text in the text.
  • the terminal displays the lip prompt information in the form of an image
  • a lip-speech information in the form of an image is displayed on the display interface of the terminal, and the lip-speech information prompts the verification object to read the name of the object in the image.
  • the terminal displays the lip language prompt information in the form of a video
  • a lip-speech message in the form of a video is displayed on the display interface of the terminal, and the lip-speech information prompts the verification object to answer the answer to the question posed in the video.
  • the terminal plays the lip-speech information in the form of audio.
  • the terminal plays the lip-speech information in an audio form, and the lip-speech information prompts the verification object to read the lip-speech information consistent with the audio.
  • the terminal acquires at least one frame of the image of the verification object.
  • the terminal collects at least one frame of the image of the verification object.
  • the terminal may acquire at least one frame of the image of the verification object by taking a photo.
  • the terminal can acquire at least one frame of the verification object by capturing a video.
  • step 1905 the terminal transmits the at least one frame image to the server.
  • step 1906 the server receives the at least one frame image transmitted by the terminal.
  • step 1907 when a lip change is included in at least one frame of image, the server detects whether the lip change and the lip prompt information match.
  • the face recognition technology it can be determined whether there is a face in at least one frame of the image.
  • the position of the lip in the face can be determined by the difference between the color of the lip and the skin color.
  • the method of the model point can be used to determine whether the verification object in the image has a lip change.
  • This step 1907 can include, but is not limited to, the following sub-steps:
  • Step one the server extracts a lip image sequence in at least one frame of the image, the lip image sequence is used to represent a lip change of the verification object;
  • the server can determine whether there is a face in at least one frame of the collected image by the face recognition technology.
  • the position of the lip in the face can be determined by the difference between the color of the lip and the skin color.
  • the method of the model point can be used to determine whether the verification object in the image has a lip change.
  • the model point is the corner point on the curve of the edge of the lip.
  • the server can use the model point method to determine the position of each point of the lip, forming a state of the lip image at a certain moment.
  • the terminal extracts the lip image in each frame of image to form a sequence of lip images.
  • the sequence of lip images is used to characterize lip changes in the subject.
  • Step 2 The server calculates a probability value of the string matching of the lip image sequence and the lip prompt information
  • the lip prompt information corresponds to a character string
  • the server calculates a probability value of the match according to the extracted lip image sequence and the string corresponding to the lip prompt information.
  • the second step may include but is not limited to the following sub-steps:
  • the server selects a prediction model sequence corresponding to the character string in a preset hidden Markov model according to the character arrangement in the character string corresponding to the lip prompt information, and the hidden Markov model includes n prediction models. , each prediction model corresponds to one character, n ⁇ 1;
  • the string is "WORD” and contains four characters.
  • the server selects a prediction model corresponding to each of the four characters “W”, “O”, “R”, and "D”.
  • the server identifies the matching degree of the lip image sequence with the predicted model sequence
  • the matching degree of the prediction model corresponding to the "W” character in the server identification lip image sequence is 90%, and the matching degree of the prediction model corresponding to the "O” character in the server identification lip image sequence is 95%, server identification
  • the matching degree of the prediction model corresponding to the "R” character in the lip image sequence is 98%; the matching degree of the prediction model corresponding to the "R” character in the server recognition lip image sequence is 90%.
  • the server calculates a probability value of the lip image sequence matching the string according to the matching degree
  • the server calculates the lip image sequence to match the string according to the matching degree.
  • Step 3 The server detects whether the probability value is greater than a preset threshold.
  • the preset threshold is 80%, 75% ⁇ 80%, so the probability value is less than the preset threshold, that is, the lip change does not match the lip prompt information.
  • step 1908 the server feeds back the detection result to the terminal.
  • the detection result may be that the lip change matches the lip prompt information, or the lip change does not match the lip prompt information.
  • Step 1909 if the lip change matches the lip prompt information, the terminal determines that the verification object passes the identity verification.
  • the terminal determines that the verification object passes the identity verification; if the lip change does not match the lip prompt information, the terminal determines that the verification object is authenticated.
  • the identity verification method provides a lip language prompt information to the verification object, and acquires at least one frame image of the verification object.
  • the lip change is detected. Lip prompt Whether the information matches; if the lip change matches the lip prompt information, it is determined that the verification object passes the authentication; if the illegal user puts the photo of the legitimate user in front of the face, the identity verification can still be passed, resulting in a security risk. problem.
  • the identity verification method provided in this embodiment further extracts a lip image sequence in at least one frame image, wherein the lip image sequence is used to represent a lip change of the verification object; and the lip image sequence is calculated to correspond to the lip prompt information.
  • FIG. 20 is a flowchart of a method for an identity verification method according to an embodiment of the present invention. This embodiment is exemplified by the application of the identity verification method to the terminal and the server. The method includes steps 2001 through 2010.
  • step 2001 the terminal generates lip language prompt information.
  • the terminal generates a lip prompt message.
  • the lip prompt information may be randomly generated prompt information.
  • a lip language prompt information library exists in the terminal.
  • the terminal randomly extracts a lip language prompt information from the lip language prompt information base as the lip language prompt information generated by the terminal.
  • the lip prompt information can be text, picture or video, and each lip prompt message corresponds to a string.
  • step 2002 the terminal provides lip language prompt information to the authenticated object.
  • the terminal provides lip language prompt information to the verification object.
  • the lip-speech information can be provided to the authenticated object in any of the four manners. For details, refer to the embodiment shown in FIG. 19, and no further description is made.
  • step 2003 the terminal sends the lip prompt information to the server.
  • the lip prompt information is information generated by the terminal
  • the lip prompt information is sent to the server.
  • step 2004 the server receives the lip prompt information sent by the terminal.
  • step 2005 the terminal collects at least one frame of the image of the verification object.
  • the terminal collects at least one frame of the image of the verification object.
  • the terminal can collect at least one frame of the image of the verification object through the front camera.
  • step 2006 the terminal transmits the at least one frame of image to the server.
  • step 2007, the server receives the at least one frame image sent by the terminal.
  • the terminal may simultaneously send the lip prompt information and the at least one frame image to the server.
  • step 2008 when a lip change is included in at least one frame of image, the server detects whether the lip change matches the lip prompt information.
  • the face recognition technology it can be determined whether there is a face in at least one frame of the image.
  • the position of the lip in the face can be determined by the difference between the color of the lip and the skin color.
  • the method of the model point can be used to determine whether the verification object in the image has a lip change.
  • This step 2008 may include, but is not limited to, sub-steps 2008A through 2008C as follows.
  • step 2008A the server extracts a sequence of lip images in at least one frame of the image, the sequence of lip images being used to characterize lip changes in the verification object.
  • the server extracts a sequence of lip images in at least one frame of the image, the sequence of lip images being used to characterize changes in the lips of the verification subject.
  • the server uses the model point method to determine the position of each point of the lip, forming a state of the lip image at a certain moment. In chronological order, the server extracts the lip image in each frame of image to form a sequence of lip images.
  • the sequence of lip images is used to characterize lip changes in the subject.
  • step 2008B the server calculates a probability value that the lip image sequence matches the string corresponding to the lip hint information.
  • the lip prompt information corresponds to a character string
  • the server calculates a probability value of the match according to the extracted lip image sequence and the string corresponding to the lip prompt information.
  • This step 2008B may include, but is not limited to, the following sub-steps:
  • Step 508B-1 the server selects a prediction model sequence corresponding to the character string in a preset hidden Markov model according to the character arrangement in the character string corresponding to the lip prompt information, and the hidden Markov model includes n Prediction models, each prediction model corresponding to one character, n ⁇ 1;
  • the string is "WORD” and contains four characters.
  • the server selects a prediction model corresponding to each of the four characters “W”, “O”, “R”, and "D”.
  • Step 508B-2 the server identifies the matching degree of the lip image sequence and the prediction model sequence
  • the matching degree of the prediction model corresponding to the "W” character in the server identification lip image sequence is 90%, and the matching degree of the prediction model corresponding to the "O” character in the server identification lip image sequence is 95%, server identification
  • the matching degree of the prediction model corresponding to the "R” character in the lip image sequence is 98%; the matching degree of the prediction model corresponding to the "R” character in the server recognition lip image sequence is 90%.
  • Step 508B-3 the server calculates a probability value that the lip image sequence matches the character string according to the matching degree
  • the server calculates the lip image sequence to match the string according to the matching degree.
  • step 2008C the server detects whether the probability value is greater than a preset threshold.
  • the preset threshold is 80%, 75% ⁇ 80%, so the probability value is less than the preset threshold, that is, the lip change does not match the lip prompt information.
  • step 2009 the server feeds back the detection result to the terminal.
  • the detection result may be that the lip change matches the lip prompt information, or the lip change does not match the lip prompt information.
  • step 2010 if the lip change matches the lip prompt information, the terminal determines that the verification object passes the identity verification. .
  • the terminal determines that the verification object passes the identity verification; if the lip change does not match the lip prompt information, the terminal determines that the verification object does not pass the authentication.
  • the identity verification method provides a lip language prompt information to the verification object, and acquires at least one frame image of the verification object.
  • the lip change is detected. Whether the lip prompt information matches; if the lip change matches the lip prompt information, it is determined that the verification object passes the authentication; if the illegal user puts the photo of the legitimate user in front of the face, the identity verification can still be caused. The problem of security risks.
  • the identity verification method provided in this embodiment further extracts a lip image sequence in at least one frame image, wherein the lip image sequence is used to represent a lip change of the verification object; and the lip image sequence is calculated to correspond to the lip prompt information.
  • FIG. 21 is a structural block diagram of an identity verification apparatus according to an embodiment of the present invention.
  • This embodiment is exemplified by the application of the identity verification device to the terminal.
  • the authentication device can be implemented as all or part of the terminal by software, hardware or a combination of the two, the device comprising:
  • the information providing module 2120 is configured to provide lip language prompt information to the verification object
  • the image acquisition module 2140 is configured to collect at least one frame image of the verification object
  • the first detecting module 2160 detects whether the lip change and the lip prompt information match when the lip change is included in the at least one frame image
  • the determining module 2180 is configured to determine that the verification object passes the identity verification if the lip change matches the lip prompt information.
  • the identity verification apparatus provides lip-alert information to the verification object, acquires at least one frame image of the verification object, and detects a lip change when at least one frame image includes a lip change. Whether the lip prompt information matches; if the lip change matches the lip prompt information, it is determined that the verification object passes the authentication; if the illegal user puts the photo of the legitimate user in front of the face, the identity verification can still be caused. The problem of security risks.
  • FIG. 22 a block diagram showing the structure of an identity verification apparatus according to another embodiment of the present invention is shown.
  • This embodiment is exemplified by the application of the identity verification device to the terminal.
  • the authentication device can be implemented as all or part of the terminal by software, hardware or a combination of the two, the device comprising:
  • the information providing module 2220 is configured to provide lip language prompt information to the verification object
  • the image acquisition module 2240 is configured to collect at least one frame image of the verification object
  • the first detecting module 2260 detects whether the lip change matches the lip prompt information when the lip change is included in the at least one frame image;
  • the determining module 2280 is configured to determine that the verification object passes the identity verification if the lip change matches the lip prompt information.
  • the first detecting module 2260 includes:
  • a first extracting unit 2262 configured to extract a lip image sequence in the at least one frame image, where the lip image sequence is used to represent a lip change of the verification object;
  • a first calculating unit 2264 configured to calculate a probability value that the lip image sequence matches a character string corresponding to the lip prompt information
  • the third detecting unit 2266 is configured to detect whether the probability value is greater than a preset threshold.
  • the first calculating unit 2264 includes:
  • a first selection subunit 22642 configured to select, according to a character arrangement in a character string corresponding to the lip prompt information, a prediction model sequence corresponding to the character string in a preset hidden Markov model,
  • the hidden Markov model includes n prediction models, each prediction model corresponding to one character, n ⁇ 1;
  • a first identifying subunit 22644 configured to identify a matching degree of the lip image sequence and the predicted model sequence
  • the third calculating sub-unit 22646 is configured to calculate, according to the matching degree, a probability value that the lip image sequence matches the character string.
  • the first detecting module 2260 includes:
  • the first detecting unit 2267 is configured to: when the lip prompt information is information generated by the terminal, send the at least one frame image and the lip prompt information to a server, where the server is used in the at least one Detecting whether the lip change matches the lip prompt information when the lip change is included in the frame image;
  • a second detecting unit 2268 configured to send the at least one frame image to the server when the lip prompt information is information acquired from a server, where the server is configured to include in the at least one frame image When the lip changes, it is detected whether the lip change matches the lip prompt information.
  • the information providing module 2220 includes:
  • a first providing unit 2222 configured to display the lip language prompt information in a text form
  • a second providing unit 2224 configured to display the lip language prompt information in an image form
  • a third providing unit 2226 configured to display the lip prompt information in a video format
  • the fourth providing unit 2228 is configured to play the lip prompt information in an audio form.
  • the identity verification apparatus provides lip-alert information to the verification object, acquires at least one frame image of the verification object, and detects a lip change when at least one frame image includes a lip change. Whether the lip prompt information matches; if the lip change matches the lip prompt information, it is determined that the verification object passes the authentication; if the illegal user puts the photo of the legitimate user in front of the face, the identity verification can still be caused. The problem of security risks.
  • FIG. 23 is a structural block diagram of an identity verification apparatus according to an embodiment of the present invention.
  • This embodiment is exemplified by the application of the identity verification device to a server.
  • the authentication device can be implemented as all or part of the server by software, hardware or a combination of the two, the device comprising:
  • the image receiving module 2320 is configured to receive at least one frame image sent by the terminal;
  • the information obtaining module 2340 is configured to obtain lip language prompt information provided by the terminal to the verification object.
  • the second detecting module 2360 is configured to detect whether the lip change and the lip prompt information match when the lip change is included in the at least one frame image;
  • the result feedback module 2380 is configured to feed back the detection result to the terminal.
  • the identity verification apparatus receives the at least one frame image sent by the terminal, acquires the lip prompt information provided by the terminal to the verification object, and includes the lip change in the at least one frame image. Detecting whether the lip change matches the lip prompt information; feeding back the detection result to the terminal; and solving the problem that if the illegal user places the photo of the legitimate user in front of the face, the user can still pass the identity authentication, resulting in the occurrence of The problem of security risks.
  • FIG. 24 is a structural block diagram of an identity verification apparatus according to an embodiment of the present invention.
  • the authentication device is applied to the server for illustration.
  • the authentication device can be implemented as all or part of the server by software, hardware or a combination of the two, the device comprising:
  • the image receiving module 2420 is configured to receive at least one frame image sent by the terminal;
  • the information obtaining module 2440 is configured to obtain lip language prompt information provided by the terminal to the verification object.
  • a second detecting module 2460 configured to detect whether the lip change and the lip prompt information match when the lip change is included in the at least one frame image
  • the result feedback module 2480 is configured to feed back the detection result to the terminal.
  • the second detecting module 960 includes:
  • a second extracting unit 2462 configured to extract a lip image sequence in the at least one frame image, where the lip image sequence is used to represent a lip change of the verification object;
  • a second calculating unit 2464 configured to calculate a probability value that the lip image sequence matches a character string corresponding to the lip prompt information
  • the fourth detecting unit 2466 is configured to detect whether the probability value is greater than a preset threshold.
  • the second calculating unit 2464 includes:
  • a second selection subunit 24642 configured to select, according to a character arrangement in a character string corresponding to the lip prompt information, a prediction model sequence corresponding to the character string in a preset hidden Markov model,
  • the hidden Markov model includes n prediction models, each prediction model corresponding to one character, n ⁇ 1;
  • a second identification subunit 24644 configured to identify a matching degree of the lip image sequence and the prediction model sequence
  • the fourth calculating subunit 24646 is configured to calculate, according to the matching degree, a probability value that the lip image sequence matches the character string.
  • the information obtaining module 2440 includes:
  • the first obtaining unit 2442 is configured to receive, when the lip prompt information is information generated by the terminal, the lip prompt information sent by the terminal;
  • the second obtaining unit 2444 is configured to read the lip prompt information stored in the server when the lip prompt information is information generated by the server and sent to the terminal.
  • the identity verification apparatus receives the at least one frame image sent by the terminal, acquires the lip prompt information provided by the terminal verification object, and includes the lip change in the at least one frame image. Detecting whether the lip change matches the lip prompt information; feeding back the detection result to the terminal; and solving the problem that if the illegal user puts the photo of the legitimate user in front of the face, the user can still pass the identity authentication, resulting in security. Hidden problems.
  • An embodiment of the present invention further provides an identity verification apparatus.
  • the apparatus may include one or more processors 2501 and a storage medium 2502 storing operation instructions when operating an operation instruction in the storage medium 2502.
  • the processor 2501 performs the following steps:
  • the verification object passes the verification.
  • the processor 2501 obtains the authentication information, it performs:
  • Image information, sound information, and associated information of image information and sound information are extracted from the video information.
  • the processor 2501 compares the authentication information with the information sample of the verification prompt information, the processor 2501 performs:
  • the verification object passes the verification when the number of matching of the image feature, the voiceprint feature, and the associated information and the image sample, the voiceprint sample, and the associated information sample is greater than or equal to a preset threshold.
  • processor 2501 further performs:
  • the verification object passes the verification.
  • the processor 2501 obtains the authentication information, it performs:
  • the processor 2501 when comparing the identity verification information with the information sample of the verification prompt information, the processor 2501 performs:
  • the verification object passes the verification.
  • the processor 2501 obtains the authentication information, it performs:
  • the processor 2501 compares the authentication information with the information sample of the verification prompt information, the processor 2501 performs:
  • the lip change is compared to the baseline lip change of the verification prompt message.
  • the processor 2501 compares the lip change with the reference lip change of the verification prompt information, the processor 2501 performs:
  • the verification object passes the verification.

Abstract

本发明实施例提供了一种身份验证方法、装置和存储介质。其中，该方法包括：验证设备向验证对象提供验证提示信息；所述验证设备获取身份验证信息，所述身份验证信息由所述验证对象根据所述验证提示信息生成；所述验证设备将所述身份验证信息与所述验证提示信息的信息样本进行比较；当所述身份验证信息与所述验证提示信息的信息样本之间的匹配度满足预设值，则所述验证对象通过验证。

Description

身份验证方法、装置和存储介质
本申请要求于2014年10月15日提交中国专利局、申请号为201410545593.X、发明名称为“一种身份验证方法、装置和系统”，于2014年11月14日提交中国专利局、申请号为201410647776.2、发明名称为“活体鉴别方法及装置”，于2014年12月16日提交中国专利局、申请号为201410779600.2、发明名称为“一种切换工作状态的方法和装置”，的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及通信技术领域，具体涉及一种身份验证方法、装置和存储介质。
背景技术
随着互联网技术的发展，信息安全也越来越受到各方的关注。为了保证信息安全，现有技术提出了很多身份验证方案，比如基于密码的身份验证方案、基于指纹的身份验证方案或基于手机绑定的身份验证方案，等等。在现有的这些身份验证方案中，一般都需要事先保存用户输入的密码或指纹等作为样本数据，然后在身份验证阶段，获取用户输入的密码或指纹等，并将用户输入的密码或指纹等和事先录入的密码或指纹样本数据进行匹配，若能匹配，则确定用户的身份合法，验证通过，否则，如果不匹配则表示用户的身份不合法，验证不通过。
但是现有方案的安全性并不高，比如，一旦密码或指纹数据被非法侵入者窃取，则非法入侵者很容易就可以盗取其账户中的信息数据，信息安全性无法受到保障，身份验证的效果不佳。
发明内容
本发明实施例提供了一种身份验证方法、装置和存储介质，以解决现有技术中信息安全无法保证的问题。
本发明实施例所提供的身份验证方法包括：验证设备向验证对象提供验证提示信息；验证设备获取身份验证信息，身份验证信息由验证对象根据验证提示信息生成；验证设备将身份验证信息与验证提示信息的信息样本进行比较；当身份验证信息与验证提示信息的信息样本之间的匹配度满足预设值，则验证对象通过验证。
本发明实施例所提供的身份验证装置包括：一个或多个处理器和存储有操作指令的存储介质，当运行存储介质中的操作指令时，处理器执行：向验证对象提供验证提示信息；获取身份验证信息，身份验证信息由验证对象根据验证提示信息生成；将身份验证信息与验证提示信息的信息样本进行比较；当身份验证信息与验证提示信息的信息样本之间的匹配度满足预设值，则验证对象通过验证
本发明实施例所提供的一种非瞬时性的计算机可读存储介质，其上存储有计算机可执行指令，当计算机中运行这些可执行指令时，执行：向验证对象提供验证提示信息；获取 身份验证信息，身份验证信息由验证对象根据验证提示信息生成；将身份验证信息与验证提示信息的信息样本进行比较；当身份验证信息与验证提示信息的信息样本之间的匹配度满足预设值，则验证对象通过验证。
附图说明
图1为本发明实施例提供的一种身份验证方法流程示意图；
图2为本发明实施例提供的一种身份验证方法流程示意图；
图3为本发明实施例中身份验证信息的获取场景示意图；
图4为本发明实施例提供的身份验证方法的网络框架图；
图5为本发明实施例提供的身份验证方法的场景示意图；
图6为本发明实施例提供的一种身份验证方法流程示意图；
图7为本发明实施例提供的一种身份验证方法流程示意图；
图8为本发明实施例提供的一种身份验证装置的结构示意图；
图9为本发明实施例提供的一种身份验证装置的结构示意图；
图10为本发明实施例提供的一种身份验证方法流程示意图；
图11为本发明实施例提供的设置基准音频数据和基准文本数据的方法流程示意图；
图12为本发明实施例提供的一种身份验证方法流程示意图；
图13为本发明实施例提供的身份验证方法所涉及的一界面示意图；
图14为本发明实施例提供的身份验证方法所涉及的另一界面示意图；
图15为本发明实施例提供的身份验证方法所涉及的模型点的示意图；
图16为本发明实施例提供的身份验证方法所涉及的概率计算方法的方法流程图；
图17为本发明实施例提供的身份验证方法所涉及的匹配计算示意图；
图18为本发明实施例提供的身份验证方法流程示意图；
图19为本发明实施例提供的身份验证方法流程示意图；
图20为本发明实施例提供的身份验证方法流程示意图；
图21为本发明实施例提供的身份验证装置的结构示意图；
图22是本发明实施例提供的身份验证装置的结构示意图；
图23是本发明实施例提供的身份验证装置的结构示意图；
图24是本发明实施例提供的身份验证装置的结构示意图；以及
图25是本发明实施例提供的身份验证装置的结构示意图。
具体实施方式
图1为本发明实施例所提供的一种身份验证方法流程示意图，该方法可以应用于终端或服务器等设备。其中，终端可以为手机、平板电脑或个人计算机等设备。如图1所示，该方法可以包括步骤101至步骤103。
在步骤101中，验证设备向验证对象提供验证提示信息。
在步骤102中，验证设备获取身份验证信息，身份验证信息由验证对象根据验证提示 信息生成。
在步骤103中，验证设备将身份验证信息与验证提示信息的信息样本进行比较；当身份验证信息与验证提示信息的信息样本之间的匹配度满足预设值，则验证对象通过验证。
本实施例将从身份验证装置的角度进行描述，该身份验证装置具体可以为终端或服务器等设备，其中，终端具体可以为手机、平板电脑或个人计算机等设备。
一种身份验证方法，包括：获取数据处理请求；根据所述数据处理请求获取身份验证信息，所述身份验证信息包括图像信息、声音信息、以及所述图像信息和声音信息的关联信息；判断所述身份验证信息与预置的信息样本是否匹配；若是，则通过所述数据处理请求；若否，则拒绝所述数据处理请求。
如图2所示，该身份验证方法的具体流程可以包括步骤201至步骤203。
在步骤201中，获取数据处理请求。
例如，如果该身份验证装置为服务器，则具体可以接收终端发送的数据处理请求。
又例如，如果该身份验证装置为终端，则具体可以接收用户触发的数据处理请求，等等。
其中，该数据处理请求具体可以为虚拟资源转移请求，比如转账请求、付款请求或还款请求，等等，或者，该数据处理请求也可以使其他的文件处理请求或业务请求等，该数据处理请求中还可以携带用户的身份标识等信息，在此不再赘述。
在步骤202中，根据接收到的数据处理请求获取身份验证信息，其中，该身份验证信息包括图像信息、声音信息、以及该图像信息和声音信息的关联信息等信息。
其中，图像信息和声音信息的关联信息指的是图像信息中的图像与声音信息中的声音之间的位置关系和/或时间关系，比如，某一帧图像和某一时刻的声音之间的时间间隔，等等。
其中，根据执行主体的不同，获取该身份验证信息的方式也有所不同。如果该身份验证装置为服务器，则此时，具体可以根据该数据处理请求向终端发送身份信息获取请求，然后接收终端根据该身份信息获取请求返回的身份验证信息。如果该身份验证装置为终端，则此时，具体可以根据该数据处理请求接收用户输入的身份验证信息。
其中，该身份验证信息可以通过多种信息格式来承载，比如，可以是视频信息，也可以是独立的图像信息和声音信息，即步骤“根据接收到的数据处理请求获取身份验证信息”可以包括如下两种方式。
方式一
根据该数据处理请求获取视频信息，从该视频信息中提取相应的图像信息、声音信息、以及该图像信息和声音信息的关联信息。
例如，如果该身份验证装置为服务器，则此时，具体可以根据该数据处理请求向终端发送身份信息获取请求，然后接收终端根据该身份信息获取请求返回的视频信息，从该视频信息中提取相应的图像信息、声音信息、以及该图像信息和声音信息的关联信息。
又例如，如果该身份验证装置为终端，则此时，具体可以根据该数据处理请求接收用户输入的视频信息，从该视频信息中提取相应的图像信息、声音信息、以及该图像信息和 声音信息的关联信息。
方式二
根据该数据处理请求分别获取图像信息和声音信息，根据该图像信息和声音信息确定所述图像信息和声音信息的关联信息。
例如，如果该身份验证装置为服务器，则此时，具体可以根据该数据处理请求向终端发送身份信息获取请求，然后接收终端根据该身份信息获取请求返回的图像信息和声音信息，根据该图像信息和声音信息确定所述图像信息和声音信息的关联信息。
又例如，如果该身份验证装置为终端，则此时，具体可以根据该数据处理请求接收用户分别输入的图像信息和声音信息，根据该图像信息和声音信息确定所述图像信息和声音信息的关联信息。
需说明的是，在获取声音信息时，可以由用户直接输入，也可以显示文字供用户朗读，其中，该显示的文字为预置的信息样本中的语音内容相应的文字，即步骤“获取声音信息”具体可以包括：
根据该数据处理请求获取该信息样本中的语音内容，将该语音内容以文字的形式进行显示，并提示用户朗读该文字，收录用户朗读该文字时的声音，得到声音信息。
在步骤203中，判断该身份验证信息与预置的信息样本是否匹配，若匹配，则通过该数据处理请求；否则，若不匹配，则拒绝该数据处理请求。
其中，信息样本可以包括图像特征样本、声纹特征样本、以及图像信息和声音信息的关联信息样本等，则步骤“判断该身份验证信息与预置的信息样本是否匹配”具体可以如下：
分别从该图像信息中提取图像特征，以及从该声音信息中提取声纹特征，分别确定该图像特征、声纹特征和该关联信息与预置的图像特征样本、声纹特征样本、以及关联信息样本是否匹配(即确定该图像特征与预置的图像特征样本是否匹配，确定声纹特征与预置的声纹特征样本是否匹配，以及确定该关联信息与预置的关联信息样本是否匹配)；若其中匹配的项数超过预置阈值，则确定所述身份验证信息与预置的信息样本匹配；若其中匹配的项数未超过预置阈值，则确定该身份验证信息与预置的信息样本不匹配。
其中，该预置阈值可以根据实际应用的需求进行设置，该阈值大于等于2。
例如，以该阈值设置为2为例，则如果“图像特征和图像特征样本匹配，且声纹特征和声纹特征样本匹配”，或，“图像特征和图像特征样本匹配，且关联信息与关联信息样本匹配”，或，“声纹特征和声纹特征样本匹配，且关联信息与关联信息样本匹配”，则确定该身份验证信息与预置的信息样本匹配，否则，则确定该身份验证信息与预置的信息样本不匹配。
又例如，以该阈值设置为大于等于2为例，则如果“图像特征和图像特征样本匹配，且声纹特征和声纹特征样本匹配”，或，“图像特征和图像特征样本匹配，且关联信息与关联信息样本匹配”，或，“声纹特征和声纹特征样本匹配，且关联信息与关联信息样本匹配”，或，“图像特征和图像特征样本匹配，且声纹特征和声纹特征样本匹配，且关联信息与关联信息样本匹配”，则确定该身份验证信息与预置的信息样本匹配，否则，则确定该身份验证 信息与预置的信息样本不匹配。
又例如，还可以将该阈值设置为3，即要全部都匹配，才确定该身份验证信息与预置的信息样本匹配，否则，若其中有任意一项不匹配，则确定该身份验证信息与预置的信息样本不匹配，以此类推，在此不再赘述。
可选的，为了进一步提高信息的安全性，还可以对语音内容也进行验证，即该信息样本还可以包括语音内容样本，则此时，步骤“判断该身份验证信息与预置的信息样本是否匹配”可以包括：
分别从该图像信息中提取图像特征，以及从该声音信息中提取声纹特征和语音内容，分别确定该图像特征、声纹特征、语音内容和该关联信息是否与预置的图像特征样本、声纹特征样本、语音内容样本以及关联信息样本匹配(即确定该图像特征与预置的图像特征样本是否匹配，确定声纹特征与预置的声纹特征样本是否匹配，语音内容与预置的语音内容样本是否匹配，以及确定该关联信息与预置的关联信息样本是否匹配)；若其中匹配的项数超过预置阈值，则确定所述身份验证信息与预置的信息样本匹配；若其中匹配的项数未超过预置阈值，则确定所述身份验证信息与预置的信息样本不匹配。
其中，该预置阈值可以根据实际应用的需求进行设置，该阈值大于等于2，比如，可以设置为2、3或4，也可以设置一定的范围，比如大于等于2，大于等于3，等等，参见前面的描述，在此不再赘述。
其中，该信息样本可以根据实际应用的需求或用户的喜好进行设置，即在获取数据处理请求之前，该身份验证还可以包括获取并保存信息样本。
在本实施例中，可以通过如下两种方式获取并保存信息样本。
方式一
获取用户的视频信息样本，从该视频信息样本中提取相应的图像信息样本、声音信息样本、以及所述图像信息和声音信息的关联信息样本，从该图像信息样本中提取图像特征样本、以及从该声音信息样本中提取声纹特征样本，保存该图像特征样本、声纹特征样本、以及所述图像信息和声音信息的关联信息样本。
当然，如果该信息样本还包括语音内容，则还可以从声音信息样本中提取语音内容样本，并保存该语音内容样本。
方式二
分别获取用户的图像信息样本和声音信息样本，根据该图像信息样本和声音信息样本确定该图像信息样本和声音信息样本的关联关系，得到关联信息样本，从该图像信息样本中提取图像特征样本、以及从该声音信息样本中提取声纹特征样本，保存该图像特征样本、声纹特征样本、以及所述图像信息和声音信息的关联信息样本。
当然，如果该信息样本还包括语音内容，则还可以从声音信息样本中提取语音内容样本，并保存该语音内容样本。
其中，在获取声音信息样本时，可以由用户直接输入，也可以显示文字供用户朗读，即步骤“获取声音信息样本”可以包括：
按照预置策略生成文字，显示这些文字，并提示用户朗读这些文字，收录用户朗读这 些文字时的声音，得到声音信息样本。
例如，参见图3，终端可以拍摄用户“打响指”的动作，并显示“芝麻开门”这四个字，录制用户朗读这四个字时的声音信息，然后将这些信息提供给服务器，由服务器从这些信息中提取出图像特征、声纹特征、语音内容、以及图像信息和声音信息的关联关系，比如，该关联关系可以是“打响指”这个动作的拍摄时间与用户朗读“芝麻开门”这四个字时的时间间隔，等等，然后将提取到的这些信息作为信息样本进行保存，以作为身份验证的密码，比如可以作为用户的支付密码、转账密码或其他业务的密码，等等。
还需说明的是，在通过用户的数据处理请求后，后续还可以根据该数据处理请求进行数据处理，比如，进行转账、付款或打开文件夹等等操作，在此不再赘述。
由上可知，本实施例采用获取数据处理请求，根据该数据处理请求获取身份验证信息，其中，该身份验证信息包括图像信息、声音信息、以及所述图像信息和声音信息的关联信息；判断该身份验证信息与预置的信息样本是否匹配，若匹配，则通过所述数据处理请求，否则，如果不匹配，则拒绝所述数据处理请求；由于在该方案中，可以将图像信息和声音信息结合起来作为身份验证信息，使得该身份验证信息不仅可以包括图像信息和声音信息，而且还可以包括该图像信息和声音信息的关联信息，因此，相对于现有技术中只采用单一的密码或指纹等信息而言，增加了验证信息的复杂度，提高了非法入侵者盗取该身份验证信息的难度，所以，相对于现有技术而言，可以加强身份验证的效果，提高信息的安全性。
本发明实施例提供了另一种身份认证方法，在本实施例中，将以该身份验证装置具体为服务器为例进行说明。
参见图4，该身份验证方法所应用的通信系统可以包括服务器和终端。
终端，用于发送数据处理请求给服务器，以及在接收到服务器发送的身份信息获取请求后，获取身份验证信息，将该身份验证信息发送给服务器。
其中，该身份验证信息包括图像信息、声音信息、以及该图像信息和声音信息的关联信息等信息。
其中，图像信息和声音信息的关联信息指的是图像信息中的图像与声音信息中的声音之间的位置关系和/或时间关系，比如，某一帧图像和某一时刻的声音之间的时间间隔，等等。
服务器，用于接收终端发送的数据处理请求，根据该数据处理请求向终端发送身份信息获取请求，然后接收终端根据该身份信息获取请求返回的身份验证信息，判断该身份验证信息与预置的信息样本是否匹配；若是，则通过该数据处理请求；若否，则拒绝该数据处理请求。
其中，该预置的信息样本可以保存在该服务器中，也可以保留在其他设备中，该信息样本中的各个信息可以保存在同一个实体中，也可以分别保存在不同的实体中，比如，参见图5，该服务器可以包括移动支付服务器、图像特征服务器、声纹特征服务器和语音内容服务器等，则可以将图像特征样本保存在图像特征服务器中，并由图像特征服务器来验证图像特征是否与图像特征样本匹配；可以将声纹特征样本保存在声纹特征服务器，并由声纹特征服务器来验证声纹特征是否与声纹特征样本匹配；可以将语音内容样本保存在语音 内容服务器，并由语音内容服务器来验证语音内容是否与语音内容样本匹配；将图像信息和声音信息的关联信息样本保存在移动支付服务器中，由移动支付服务器来验证关联信息是否与关联信息样本匹配，等等，在此不再列举。为了描述方便，在本实施例中，将以这些信息样本均保存在同一个服务器中为例进行说明。
基于该通信系统的结构，以下将对其身份验证方法的流程进行详细说明。为了描述方便，以支付请求作为数据处理请求进行说明。
如图6所示，本实施所提供的一种身份认证方法可以包括步骤601至步骤608。
在步骤601中，终端发送支付请求给服务器，比如，可以指示将某个金额的钱转入某个账户。
其中，该支付请求可以携带用户的身份标识，比如该用户的身份标识具体可以包括用户的银行账号、邮箱账号、用户名和/或用户昵称等标识。
在步骤602中，服务器接收到该支付请求后，根据该支付请求向终端发送身份信息获取请求。
在步骤603中，终端接收到服务器发送的身份信息获取请求后，根据该身份信息获取请求获取用户的身份验证信息，并将该身份验证信息发送给服务器。
其中，身份验证信息包括图像信息、声音信息、以及该图像信息和声音信息的关联信息等信息。
其中，图像信息和声音信息的关联信息指的是图像信息中的图像与声音信息中的声音之间的位置关系和/或时间关系，比如，某一帧图像和某一时刻的声音之间的时间间隔，等等。
例如，具体可以根据该身份信息获取请求生成提示信息，其中，该提示信息指示用户拍摄一段视频，或分别拍摄一系列图片以及录制一段声音，并接收用户根据该提示信息的指示录制的视频、或者图片和信息作为身份验证信息提供给服务器。
在步骤604中，服务器接收到该身份验证信息后，从该身份验证信息中提取出图像信息、声音信息、以及该图像信息和声音信息的关联信息等信息。
例如，如果接收到的是验证信息是一段视频信息，则从该视频信息中提取相应的图像信息、声音信息、以及该图像信息和声音信息的关联信息，比如在某一帧图像时相应地存在有声音，等等。
又例如，如果接收到的身份验证信息是图像信息和声音信息，则可以根据该图像信息和声音信息确定该图像信息和声音信息的关联信息，比如确定获取该图像信息和声音信息时的时间间隔，例如，如果用户是拍摄某张相片后，间隔1秒再录一段声音的话，则此时可以确定该时间间隔为1s，等等，在此不再赘述。
在步骤605中，服务器分别从该图像信息中提取图像特征，以及从该声音信息中提取声纹特征。
可选，如果该信息样本还可以包括语音内容样本，则此时还可以从该声音信息中提取语音内容。
在步骤606中，服务器根据用户的身份标识获取相应的预置的信息样本，分别确定该图 像特征、声纹特征和该关联信息是否与该信息样本匹配，若其中匹配的项数超过预置阈值，则确定该身份验证信息与预置的信息样本匹配，表示验证通过，于是执行步骤607，若其中匹配的项数超过预置阈值，则确定该身份验证信息与预置的信息样本不匹配，表示验证不通过，于是执行步骤608。
其中，该信息样本可以包括图像特征样本、声纹特征样本以及关联信息样本，还可以包括语音内容样本，具体可以如下：
分别确定该图像特征、声纹特征和该关联信息是否与预置的图像特征样本、声纹特征样本以及关联信息样本匹配，即确定该图像特征与预置的图像特征样本是否匹配，确定声纹特征与预置的声纹特征样本是否匹配，以及确定该关联信息与预置的关联信息样本是否匹配，当然，如果还需对语音内容进行验证的话，则此时还需确定该语音内容与预置的语音内容样本是否匹配，若上述几项中匹配的项数超过预置阈值，则表示所述身份验证信息与预置的信息样本匹配，于是执行步骤607，若其中匹配的项数未超过预置阈值，则表示该身份验证信息与预置的信息样本不匹配，于是执行步骤608。
其中，该预置阈值可以根据实际应用的需求进行设置，该阈值大于等于2，具体可参见前面的实施例，在此不再赘述。
其中，该信息样本可以根据实际应用的需求或用户的喜好进行设置，具体的设置方法可参见图2所示实施例，在此不再赘述。
在步骤607中，服务器通过该支付请求的验证，此时可以返回验证通过的响应给终端。
此后，服务器可以根据该支付请求进行支付处理，比如可以将某个金额的钱转入某个账户，等等，后续还可以将处理结果通知给终端，在此不再赘述。
在步骤608中，服务器拒绝该支付请求的验证，此时可以返回验证失败的响应给终端。
在本实施例中，将以该身份验证装置具体为终端，且数据处理请求具体为终端应用启动请求为例进行说明。
如图7所示，本实施例所提供的一种身份认证方法可以包括步骤701至步骤707。
在步骤701中，终端接收到用户触发的终端应用启动请求，比如，可以指示打开某个终端应用，比如相册或个人文档，等等。
在步骤702中，终端接收到该终端应用启动请求后，根据该终端应用启动请求生成并显示提示信息，其中，该提示信息指示用户拍摄一段视频，或分别拍摄一系列图片以及录制一段声音，并接收用户根据该提示信息的指示录制的视频、或者图片和信息，得到身份验证信息。
其中，身份验证信息包括图像信息、声音信息、以及该图像信息和声音信息的关联信息等信息。
其中，图像信息和声音信息的关联信息指的是图像信息中的图像与声音信息中的声音之间的位置关系和/或时间关系，比如，某一帧图像和某一时刻的声音之间的时间间隔，等等。
在步骤703中，终端从该身份验证信息中提取出图像信息、声音信息、以及该图像信息和声音信息的关联信息等信息。
在步骤704中，终端分别从该图像信息中提取图像特征，以及从该声音信息中提取声纹特征。
可选，如果该信息样本还可以包括语音内容样本，则此时还可以从该声音信息中提取语音内容。
在步骤705中，终端获取相应的预置的信息样本，分别确定该图像特征、声纹特征和该关联信息是否与该信息样本匹配，若这几项中匹配的项数超过预置阈值，则确定该身份验证信息与预置的信息样本匹配，于是执行步骤706，若其中匹配的项数未超过预置阈值，则确定该身份验证信息与预置的信息样本不匹配，于是执行步骤707。
其中，该信息样本可以包括图像特征样本、声纹特征样本以及关联信息样本，还可以包括语音内容样本，具体可以如下：
分别确定该图像特征、声纹特征和该关联信息是否与预置的图像特征样本、声纹特征样本以及关联信息样本匹配，即确定该图像特征与预置的图像特征样本是否匹配，确定声纹特征与预置的声纹特征样本是否匹配，以及确定该关联信息与预置的关联信息样本是否匹配，当然，如果还需对语音内容进行验证的话，则此时还需确定该语音内容与预置的语音内容样本是否匹配，若上述几项中匹配的项数超过预置阈值，则表示所述身份验证信息与预置的信息样本匹配，于是执行步骤706，若其中匹配的项目未超过预置阈值，则表示该身份验证信息与预置的信息样本不匹配，于是执行步骤707。
其中，该预置阈值可以根据实际应用的需求进行设置，该阈值大于等于2，具体可参见前面的实施例，在此不再赘述。
其中，该信息样本可以根据实际应用的需求或用户的喜好进行设置，具体的设置方法可参见图6所示实施例，在此不再赘述。
在步骤706中，终端通过该终端应用启动请求，此时可以显示验证通过的通知给用户。
此后，终端可以根据该终端应用启动请求启动相应的终端应用，比如打开相册或个人文档，等等。
在步骤707中，终端拒绝该终端应用启动请求，此时可以显示验证失败的通知给用户。
本发明实施例还提供一种身份验证装置，如图8所示，该身份验证装置包括请求获取单元801、信息获取单元802和验证单元803，如下：
请求获取单元801，用于获取数据处理请求。
例如，如果该身份验证装置为服务器，则请求获取单元801具体可以接收终端发送的数据处理请求。
又例如，如果该身份验证装置为终端，则请求获取单元801具体可以接收用户触发的数据处理请求，等等。
其中，该数据处理请求具体可以为虚拟资源转移请求，比如转账请求、付款请求或还款请求，等等，或者，该数据处理请求也可以使其他的文件处理请求或业务请求等，该数据处理请求中还可以携带用户的身份标识等信息，在此不再赘述。
信息获取单元802，用于根据该数据处理请求获取身份验证信息，其中，该身份验证信息包括图像信息、声音信息、以及所述图像信息和声音信息的关联信息等信息。
其中，图像信息和声音信息的关联信息指的是图像信息中的图像与声音信息中的声音之间的位置关系和/或时间关系，比如，某一帧图像和某一时刻的声音之间的时间间隔，等等。
验证单元803，用于判断所述身份验证信息与预置的信息样本是否匹配，若是，则通过所述数据处理请求，若否，则拒绝所述数据处理请求。
其中，根据执行主体的不同，获取该身份验证信息的方式也有所不同，例如，可以如下：
如果该身份验证装置为服务器，则此时，信息获取单元802具体可以根据该数据处理请求向终端发送身份信息获取请求，然后接收终端根据该身份信息获取请求返回的身份验证信息。
如果该身份验证装置为终端，则此时，信息获取单元802具体可以根据该数据处理请求接收用户输入的身份验证信息。
其中，该身份验证信息可以通过多种信息格式来承载，比如，可以是视频信息，也可以是独立的图像信息和声音信息，即信息获取单元802可通过两种方式获取身份验证信息，具体可参见方法实施例。此处不做赘述。
需说明的是，在获取声音信息时，可以由用户直接输入，也可以显示文字供用户朗读，其中，该显示的文字为预置的信息样本中的语音内容相应的文字，即操作“获取声音信息”具体可以为：
信息获取单元802，具体可以用于根据该数据处理请求获取该信息样本中的语音内容，将该语音内容以文字的形式进行显示，并提示用户朗读该文字，收录用户朗读该文字时的声音，得到声音信息。
其中，信息样本可以包括图像特征样本、声纹特征样本、以及图像信息和声音信息的关联信息样本等，则：
验证单元803，具体可以用于：分别从该图像信息中提取图像特征，以及从该声音信息中提取声纹特征，分别确定该图像特征、声纹特征和该关联信息与预置的图像特征样本、声纹特征样本、以及关联信息样本是否匹配(即确定该图像特征与预置的图像特征样本是否匹配，确定声纹特征与预置的声纹特征样本是否匹配，以及确定该关联信息与预置的关联信息样本是否匹配)；若其中匹配的项数超过预置阈值，则确定所述身份验证信息与预置的信息样本匹配；若其中匹配的项数未超过预置阈值，则确定该身份验证信息与预置的信息样本不匹配。
可选的，为了进一步提高信息的安全性，还可以对语音内容也进行验证，即该信息样本还可以包括语音内容样本，则此时：
验证单元803，具体可以用于分别从该图像信息中提取图像特征，以及从该声音信息中提取声纹特征和语音内容，分别确定该图像特征、声纹特征、语音内容和该关联信息是否与预置的图像特征样本、声纹特征样本、语音内容样本以及关联信息样本匹配(即确定该图像特征与预置的图像特征样本是否匹配，确定声纹特征与预置的声纹特征样本是否匹配，语音内容与预置的语音内容样本是否匹配，以及确定该关联信息与预置的关联信息样本是 否匹配)；若其中匹配的项数超过预置阈值，则确定所述身份验证信息与预置的信息样本匹配；若其中匹配的项数未超过预置阈值，则确定所述身份验证信息与预置的信息样本不匹配。
其中，该预置阈值可以根据实际应用的需求进行设置，该阈值大于等于2，具体可参见前面的实施例，在此不再赘述。
其中，该信息样本可以根据实际应用的需求或用户的喜好进行设置，即如图9所示，该身份验证装置还可以包括样本获取单元804，如下：
样本获取单元804，用于获取并保存信息样本。例如，具体可以采用另种方式获取并保存信息样本，具体可参见方法实施例，此处不做赘述。
当然，如果该信息样本还包括语音内容，则还可以从声音信息样本中提取语音内容样本，并保存该语音内容样本。
其中，在获取声音信息样本时，可以由用户直接输入，也可以显示文字供用户朗读，即：
样本获取单元804，具体可以用于按照预置策略生成文字，显示这些文字，并提示用户朗读这些文字，收录用户朗读这些文字时的声音，得到声音信息样本。
例如，以该身份验证装置具体为服务器为例，参见图3，样本获取单元804可以拍摄用户“打响指”的动作，并显示“芝麻开门”这四个字，录制用户朗读这四个字时的声音信息，然后将这些信息提供给服务器，由服务器从这些信息中提取出图像特征、声纹特征、语音内容、以及图像信息和声音信息的关联关系，比如，该关联关系可以是“打响指”这个动作的拍摄时间与用户朗读“芝麻开门”这四个字时的时间间隔，等等，然后将提取到的这些信息作为信息样本进行保存，以作为身份验证的密码，比如可以作为用户的支付密码、转账密码或其他业务的密码，等等。
还需说明的是，在通过用户的数据处理请求后，后续还可以根据该数据处理请求进行数据处理，比如，进行转账、付款或打开文件夹等等操作，即该身份验证装置还可以包括数据处理单元，如下：
数据处理单元，用于在通过用户的数据处理请求后，根据该数据处理请求进行数据处理。
具体实施时，以上各个单元可以作为独立的实体来实现，也可以进行任意组合，作为同一或若干个实体来实现，以上各个单元的具体实施可以参见前面的实施例，在此不再赘述。
该身份验证装置具体可以为终端，也可以为服务器，其中，该终端具体可以为手机、平板电脑或PC等设备。
由于在该方案中，可以将图像信息和声音信息结合起来作为身份验证信息，使得该身份验证信息不仅可以包括图像信息和声音信息，而且还可以包括该图像信息和声音信息的关联信息，因此，相对于现有技术中只采用单一的密码或指纹等信息而言，增加了验证信息的复杂度，提高了非法入侵者盗取该身份验证信息的难度，所以，相对于现有技术而言，可以加强身份验证的效果，提高信息的安全性。
相应的，本发明实施例还提供一种身份验证系统，可以包括本发明实施例提供的任一种身份验证装置，该身份验证装置即可以为终端，也可以为服务器，以下将以该身份验证装置分别为终端和服务器为例，进行简略说明。
(1)身份验证装置作为服务器时；
该通信系统除了可以包括服务器之外，还可以包括终端，如下：
终端，用于发送数据处理请求给服务器，以及在接收到服务器发送的身份信息获取请求后，获取身份验证信息，将该身份验证信息发送给服务器。
其中，该身份验证信息包括图像信息、声音信息、以及该图像信息和声音信息的关联信息等信息。
其中，图像信息和声音信息的关联信息指的是图像信息中的图像与声音信息中的声音之间的位置关系和/或时间关系，比如，某一帧图像和某一时刻的声音之间的时间间隔，等等。
服务器，用于接收终端发送的数据处理请求，根据该数据处理请求向终端发送身份信息获取请求，然后接收终端根据该身份信息获取请求返回的身份验证信息，判断该身份验证信息与预置的信息样本是否匹配；若是，则通过该数据处理请求；若否，则拒绝该数据处理请求。
其中，该预置的信息样本可以保存在该服务器中，也可以保留在其他设备中，该信息样本中的各个信息可以保存在同一个实体中，也可以分别保存在不同的实体中。
(2)身份验证装置作为终端时；
该通信系统至少可以包括终端，如下：
终端，用于接收用户触发的数据处理请求，根据该数据处理请求向终端发送身份信息获取请求，然后接收终端根据该身份信息获取请求返回的身份验证信息，判断该身份验证信息与预置的信息样本是否匹配；若是，则通过该数据处理请求；若否，则拒绝该数据处理请求。
以上各个设备的具体操作可参见前面的实施例，在此不再赘述。
由于该通信系统可以包括本发明实施例提供的任一种身份验证装置，因此同样可以实现本发明实施例提供的任一种身份验证装置所能实现的有益效果，详见前面的实施例，在此不再赘述。
本发明实施例提供了一种身份验证的方法，该方法的执行主体可以为终端或服务器等。其中，终端可以是手机、平板电脑、计算机等。通过该身份验证，可以进行后续数据处理，例如虚拟资源转移，还可以进行工作状态的切换等，如锁屏状态和屏幕解锁状态等。本实施例中，以工作状态由锁屏状态切换到屏幕解锁状态为例，并以终端为手机为例进行方案的详细说明，其他终端的情况与之类似，本实施例不再累述。
如图10所示，本发明所提供的身份验证方法包括步骤1001至步骤1003。
在步骤1001中，获取通过音频采集设备采集的第一音频数据。
在实施中，在执行屏幕解锁的流程之前，可以预先设置并存储基准音频数据和基准文本数据，设置过程可以是由用户对音频采集设备(如麦克风)输入一段语音，然后终端获 取此段语音的音频数据，并将音频数据转换为文本数据，作为基准音频数据和基准文本数据，此设置过程在后面内容中会由详细阐述。
对于执行主体为终端的情况，用户这时可以对着音频采集设备输入语音，语音的内容可以与自己设置基准音频数据时输入语音的内容相同，例如，用于对麦克风说“芝麻开门”。这时，音频采集设备则会采集到用户输入的音频数据(即第一音频数据)，终端则可以获取到该音频数据。当然，此过程中也可以不显示锁屏界面，保持屏幕关闭状态，进行音频数据获取。
对于执行主体为服务器的情况，终端在获取到用户输入的音频数据后，可以生成解锁请求，在解锁请求中携带该音频数据，还可以携带终端的标识，如IMEI，将解锁请求发送给服务器，服务器可以从解锁请求中获取上述通过音频采集设备采集的音频数据。
在步骤1002中，对第一音频数据进行文本转换，得到对应的第一文本数据。
在实施中，终端或服务器可以预先存储音频文本转换数据库，存储了每个字符对应的音频特征，基于第一音频数据的音频特征，可以在音频文本转换数据库查找到对应的字符，将这些字符按顺序组合在一起，从而可以将音频数据转换为第一文本数据。
在步骤1003中，如果第一文本数据与预先存储的基准文本数据相同，且第一音频数据与预先存储的基准音频数据满足预设的近似度条件，则将第一工作状态切换为第二工作状态。
其中，第一工作状态和第二工作状态可以是设置好的任意工作状态，本实施例不做限制。
在实施中，对于执行主体为终端的情况，终端在将获取的音频数据转换为文本数据后，可以将转换得到的文本数据与存储的基准文本数据进行比较，判断它们是否相同，同时，将该音频数据与存储的基准音频数据进行近似度比较，判定它们之间是否满足预设的近似度条件，此近似度条件可以根据需求任意设置，例如，该近似度条件可以是两个音频数据的频率曲线满足一定的近似度，如两个频率曲线在各个点的差值绝对值的总和小于预设阈值。如果第一文本数据与基准文本数据相同，且第一音频数据与基准音频数据满足近似度条件，则触发工作状态切换，否则，不触发工作状态切换。这样，通过语音触发进行解锁等操作，相对于滑动操作的方式，可以有效的提高终端的安全性。
对于执行主体为服务器的情况，服务器可以预先存储每个终端对应的基准文本数据和基准音频数据，具体可以建立终端的标识与基准文本数据、基准音频数据的对应关系。服务器可以从上述解锁请求中获取终端的标识，然后在上述对应关系中查找该终端的标识对应的基准文本数据、基准音频数据，再进行上述比较判断的处理。当确定触发工作状态切换时，服务器可以向终端发送切换通知，以使终端将第一工作状态切换为第二工作状态。
可选的，可以存储多组基准文本数据和基准音频数据，第一音频数据和第一文本数据只要能与其中一组匹配，就可以触发工作模式切换，相应的，步骤1003可以包括步骤10031至步骤10032。
在步骤10031中，如果在预先存储的基准文本数据与基准音频数据的对应关系中存在第一基准文本数据与第一文本数据相同，则根据基准文本数据与基准音频数据的对应关系， 确定第一基准文本数据对应的第一基准音频数据。
在实施中，以执行主体为终端为例(服务器的处理与之类似，不再累述)，用户可以设置多组基准文本数据和基准音频数据，终端则可以存储基准文本数据与基准音频数据的对应关系。终端在将获取的音频数据(即第一音频数据)转换为文本数据(即第一文本数据)后，可以在基准文本数据与基准音频数据的对应关系中，查找该文本数据，如果存在某基准文本数据(即第一基准音频数据)与该文本数据相同，则可以进一步查找该基准文本数据对应的基准音频数据(即第一基准音频数据)。
在步骤10032中，如果第一音频数据与所述第一基准音频数据满足预设的近似度条件，则将第一工作状态切换为第二工作状态。
在实施中，终端在确定第一基准音频数据后，可以按照上述类似的方式，将第一音频数据与第一基准音频数据进行近似度比较，以决定是否触发工作状态切换。
可选的，基于设置的不同，将第一工作状态切换为第二工作状态的处理可以存在多种多样的情况，以下给出了几种可行的情况：情况一，将锁屏状态切换为屏幕解锁状态；情况二，将应用程序锁定状态切换为应用程序解锁状态。
在实施中，对于情况二，上述步骤1001可以在启动应用程序的时候触发执行。
本发明实施例，还提供了设置基准音频数据和基准文本数据的方法，处理流程可以如图11所示，包括步骤1101至步骤1103。
在步骤1101中，接收设置指令，获取通过音频采集设备采集的第二音频数据。
在实施中，用户可以操作终端开启音频锁设置页面，在此界面中可以设置输入语音的选项，用户点击该选项，终端则会接收到设置指令，可以激活终端的音频采集设备，开始进行音频数据的采集，用户这时可以对着音频采集设备输入语音，如“芝麻开门”，这时，音频采集设备则会采集到用户输入的音频数据(即第二音频数据)，终端则可以获取到该音频数据。
在步骤1102中，对第二音频数据进行文本转换，得到对应的第二文本数据。
该处理过程与步骤1102类似，可以参考相应的内容，此处不再累述。
可选的，在文本转换得到第二文本数据后，可以进行以下处理：显示第二文本数据。
在步骤1103中，将第二音频数据和第二文本数据分别存储为基准音频数据和基准文本数据。
可选的，基于上述对第二文本数据的显示，可以让用户对第二文本数据进行确认，相应的，步骤1103的处理可以如下：如果接收到确认指令，则将第二音频数据和第二文本数据分别存储为基准音频数据和基准文本数据。
在实施中，终端获取第二文本数据后，可以显示确认框，在确认框中显示第二文本数据，如“芝麻开门”，并显示确认按键和重新输入按键。用户确认显示的第二文本数据确实是自己输入的语音匹配时，可以点击确认按键，这时，终端会接收到确认指令，进而终端可以将第二音频数据和第二文本数据分别存储为基准音频数据和基准文本数据。用户如果发现显示的第二文本数据与自己输入的语音不匹配时，可以点击重新输入按键，则会触发转至重新执行步骤1101。
本发明实施例中，获取通过音频采集设备采集的第一音频数据，对第一音频数据进行文本转换，得到对应的第一文本数据，如果第一文本数据与预先存储的基准文本数据相同，且第一音频数据与预先存储的基准音频数据满足预设的近似度条件，则将第一工作状态切换为第二工作状态，这样，无需在屏幕上进行多次点击操作，即可触发工作状态切换，从而，可以提高工作状态切换的效率。
本实施例将对上述实施例中语音识别的过程和声纹分析的过程进行详细说明。
终端或服务器中可以预先建立有音频文本转换数据库，在音频文本转换数据库可以存储每个字符对应的音频特征(如频率等)。当接收到用户输入的语音数据(即第一音频数据)的时候，可以根据语音数据的音频特征的变化，将该语音数据按字划分为多个音节，例如，用户输入“芝麻开门”的语音数据，可以将其划分为“芝”、“麻”、“开”、“门”的音频数据。然后，对于划分出的每个音频数据，可以将语音数据与音频文本转换数据库中的音频特征进行匹配判定，如果查找到相匹配的音频特征，则可以进一步获取该音频特征对应的字符，这样就可以得到每个音节的音频数据对应的字符，例如，对于上面的“芝”、“麻”、“开”、“门”的语音数据，在音频文本转换数据库中进行音频特征的匹配，分别可以查找到“芝”、“麻”、“开”、“门”的字符。再后，可以按照对应的音频数据的顺序，对各字符进行排序组合得到最终的文本数据(即上述第一文本数据)，例如，基于上述“芝”、“麻”、“开”、“门”的字符，可以得到“芝麻开门”字符串。
声纹分析的方式可以多种多样，以下给出了一种可行的处理方式。
终端或服务器获取用户输入的语音数据(即第一音频数据)与预先存储的基准音频数据，对于语音数据和基准音频数据，分别根据它们各自音频特征的变化，将它们按字划分为多个音节，例如，用户输入“芝麻开门”的语音数据，可以将其划分为“芝”、“麻”、“开”、“门”的音频数据，基准音频数据的划分也与之类似。然后，比较它们划分的音节数目是否相同，如果不同，则语音数据与基准音频数据匹配失败，如果相同，则可以继续按照各音节的顺序，将语音数据与基准音频数据中相对应的音节进行比较。在对任意两个音节进行具体比较时，可以在两个音节的时长上均匀选取预设数目(如3、15或30等)的时间点，计算两个音节中所有相对应的时间点的频率差的绝对值，再将这些频率差的绝对值相加的和作为误差度。在确定误差度之后，可以将误差度与预设的误差度阈值进行比较，如果误差度大于误差度阈值，则可以判定匹配失败，否则，可以判定匹配成功，如果语音数据与基准音频数据的音节两两匹配都成功，则可以判定语音数据与基准音频数据满足预设的近似度条件。
例如，语音数据中第一个音节为音节a，基准音频数据中第一个音节为音节b，分别在音节a和音节b的时长上均匀的选取3个时间点，计算音节a中第一个时间点和音节b中第一个时间点的频率差的绝对值，音节a中第二个时间点和音节b中第二个时间点的频率差的绝对值，音节a中第三个时间点和音节b中第三个时间点的频率差的绝对值，再将这三个频率差的绝对值相加作为误差度，然后基于误差度进行后续判断。
本发明实施例中，通过语音识别和声纹分析进行双重判断，触发终端解锁等操作，可以有效的提高终端的安全性。
请参考图12，其示出了本发明实施例提供的一种身份验证方法的方法流程图。本实施例以该身份验证方法应用于终端中来举例说明。该方法可以包括步骤1201至步骤1206。
在步骤1201中，向验证对象提供唇语提示信息。
首先，终端生成唇语提示信息。比如，在终端中存在一个唇语提示信息库，当需要进行身份验证时，终端从该唇语提示信息库中随机抽取一个唇语提示信息作为终端生成的唇语提示信息。
或者，服务器生成唇语提示信息。比如，在服务器中存在一个唇语提示信息库，当需要进行身份验证时，服务器从该唇语提示信息库中随机抽取一个唇语提示信息作为服务器生成的唇语提示信息。服务器将该唇语提示信息发送给终端。对应地，终端接收该唇语提示信息。
也即，该唇语提示信息可以是随机生成的提示信息；或，该唇语提示信息在相邻的验证过程中是不同的提示信息。
然后，终端向验证对象提供该唇语提示信息。本发明实施例中可以通过以下四种方式中的任意一种向验证对象提供唇语提示信息。
第一种可能的实现方式，终端显示文本形式的该唇语提示信息；
请参考图13，其示出了终端显示文本形式的该唇语提示信息的界面示意图。在终端的显示界面20上显示一个文本形式的唇语提示信息21，该唇语提示信息21提示被验证对象读出“我想要解锁”。
第二种可能的实现方式，终端显示图像形式的该唇语提示信息；
请参考图14，其示出了终端显示图像形式的该唇语提示信息的界面示意图。在终端的显示界面20上显示一个图像形式的唇语提示信息22，该唇语提示信息22提示被验证对象读出图片中的水果的名称“西瓜”。
第三种可能的实现方式，终端显示视频形式的该唇语提示信息；
在终端的显示界面上显示一个视频形式的唇语提示信息，该唇语提示信息提示验证对象回答出视频中提出的问题的答案。
第四种可能的实现方式，终端播放音频形式的该唇语提示信息。
终端播放一段音频形式的该唇语提示信息，该唇语提示信息提示验证对象读出与该音频中一致的该唇语提示信息。
在步骤1202中，采集验证对象的至少一帧图像。
终端采集验证对象的至少一帧图像。
当终端具有前置摄像头时，终端可以通过拍摄照片的方式采集到验证对象的至少一帧图像。终端还可以通过拍摄视频的方式采集到验证对象的至少一帧图像。
在步骤1203中，在至少一帧图像中包括唇部变化时，提取至少一帧图像中的唇部图像序列，该唇部图像序列用于表征验证对象的唇部变化。
在采集到的至少一帧图像中包括唇部变化时，终端检测唇部变化与唇语提示信息是否匹配。
终端通过人脸识别技术，可以确定采集到的至少一帧图像中是否存在人脸。当该至少 一帧图像中存在人脸时，可以通过唇部颜色与肤色的区别确定人脸中的唇部的位置。当获取到唇部的位置时，可以采用模型点的方法确定该图像中的验证对象是否存在唇部变化。
模型点是指位于唇部边缘曲线上的角点，请参考图15，一种可能的实施方式中，唇部的上边缘可以包括5个模型点222，唇部的下边缘也可以包括5个模型点222。
终端可以采用模型点的方法确定唇部各点的位置，构成某一时刻的唇部图像的一个状态。按照时间顺序，终端提取每一帧图像中的唇部图像，构成一个唇部图像序列。该唇部图像序列用于表征被验证对象的唇部变化。
在步骤1204中，计算唇部图像序列与唇语提示信息所对应的字符串匹配的概率值。
终端计算唇部图像序列与唇语提示信息所对应的字符串匹配的概率值。
每个唇语提示信息对应一个字符串，终端根据提取到的唇部图像序列和唇语提示信息对应的字符串计算两者匹配的概率值。
该步骤1204可以包括但不限于如步骤1204a至步骤1204c，请参考图16。
在步骤1204a中，根据与唇语提示信息对应的字符串中的字符排列，在预设的隐马尔科夫模型中选择与该字符串所对应的预测模型序列，隐马尔科夫模型包括n个预测模型，每个预测模型与一个字符对应，n≥1。
终端将与唇语提示信息对应的字符串输入隐马尔科夫模型，该隐马尔科夫模型包括与不同字符对应的预测模型。
比如，与唇语提示信息对应的字符串为“215”，包含三个字符。终端根据字符串“215”在隐马尔科夫模型中，选择与字符“2”对应的预测模型、与字符“1”对应的预测模型和与字符“5”对应的预测模型，得到一个预测模型序列。
在步骤1204b中，识别唇部图像序列与预测模型序列的匹配度。
终端识别唇部图像序列与预测模型序列的匹配度。
比如，如图17所示，终端将唇部图像23输入字符“2”的预测模型24后，可以计算得到匹配度1为90％；将唇部图像25输入字符“1”的预测模型26后，可以计算得到匹配度2为95％；将唇部图像27输入字符“5”的预测模型28后，可以计算得到匹配度3为98％。
在步骤1204c中，根据匹配度计算唇部图像序列与字符串匹配的概率值。
终端根据匹配度计算唇部图像序列与字符串匹配的概率值。
比如，唇部图像序列和与字符串中三个字符对应的预测模型的匹配度分别为90％、95％和98％，则终端根据匹配度计算唇部图像序列与该字符串匹配的概率值为90％*95％*98％＝84％。
在步骤1205中，检测概率值是否大于预设阈值。
终端检测概率值是否大于预设阈值。
比如，预设阈值为80％，84％>80％，故该概率值大于预设阈值，即唇部变化与该唇语提示信息匹配。
在步骤1206中，若唇部变化与唇语提示信息匹配，则确定验证对象通过验证。
若唇部变化与唇语提示信息匹配，则终端确定验证对象通过身份验证；若唇部变化与唇语提示信息不匹配，则终端确定验证对象未通过身份验证。
综上所述，本实施例提供的身份验证方法，通过向被验证对象提供唇语提示信息；采集验证对象的至少一帧图像；在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示信息是否匹配；若唇部变化与唇语提示信息匹配，则确定验证对象通过身份验证。解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
而且由于唇语提示信息可以是随机生成的提示信息，其具有与验证码验证机制同等的安全级别。
本实施例提供的身份验证方法，还通过提取至少一帧图像中的唇部图像序列，该唇部图像序列用于表征验证对象的唇部变化；计算唇部图像序列与唇语提示信息所对应的字符串匹配的概率值；检测概率值是否大于预设阈值；达到了更可靠检测唇部变化与唇语提示信息是否匹配的效果。
还需要说明的是，在步骤1204之前，该方法还可以包括：
预先建立预测模型，也即隐马尔科夫模型，该预测模型包括每个字符对应的唇部变化的预测模型。
该方法可以包括但不限于子步骤a)至步骤c)。
在步骤a)中，获取n个被鉴别对象反馈的不同字符对应的唇部变化的至少一帧图像，n≥1。
n个验证对象根据唇语提示信息反馈不同字符对应的唇部变化，终端获取对应该唇部变化的至少一帧图像。每个验证对象对应的至少一帧图像，可以称为一个样本。终端可以通过获取大量样本，进行统计。
在步骤b)中，提取至少一帧图像中每个字符对应的唇部图像序列。
对于每个样本，终端按照时间顺序，提取每一帧图像中的唇部图像，构成一个唇部图像序列。该唇部图像序列用于表征验证对象的唇部变化。
在步骤c)中，根据每个字符对应的唇部图像序列，统计得到每个字符对应的唇部变化的预测模型。
请参考图18，其示出了本发明又一个实施例提供的活体鉴别方法的方法流程图。本实施例以该活体鉴别方法应用于服务器中来举例说明。该方法，包括步骤1802至步骤1808。
在步骤1802中，接收终端发送的至少一帧图像。
在步骤1804中，获取终端向被鉴别对象提供的唇语提示信息。
在步骤1806中，在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示信息是否匹配。
在步骤1808中，向终端反馈检测结果。
综上所述，本实施例提供的身份验证方法，通过接收终端发送的至少一帧图像；获取终端向验证对象提供的唇语提示信息；在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示信息是否匹配；向终端反馈检测结果；解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
请参考图19，其示出了本发明还一个实施例提供的活体鉴别方法的方法流程图。本实 施例以该活体鉴别方法应用于终端和服务器中来举例说明。该方法，包括步骤1901至步骤1909。
在步骤1901中，服务器生成唇语提示信息。
服务器生成唇语提示信息。
该唇语提示信息可以是随机生成的提示信息。比如，在服务器中存在一个唇语提示信息库，当需要进行身份时，服务器从该唇语提示信息库中随机抽取一个唇语提示信息作为服务器生成的唇语提示信息。
唇语提示信息可以是文字、图片或者视频，每个唇语提示信息对应一个字符串。
在步骤1902中，服务器向终端发送该唇语提示信息。
在唇语提示信息是服务器生成的信息时，服务器向终端发送该唇语提示信息。
在步骤1903中，终端向被鉴别对象提供唇语提示信息。
终端向验证对象提供唇语提示信息。本发明实施例中可以通过以下四种方式中的任意一种向验证对象提供唇语提示信息。
第一种可能的实现方式，终端显示文本形式的该唇语提示信息；
在终端的显示界面上显示一个文本形式的唇语提示信息，该唇语提示信息提示验证对象读出该文本中的文字。
第二种可能的实现方式，终端显示图像形式的该唇语提示信息；
在终端的显示界面上显示一个图像形式的唇语提示信息，该唇语提示信息提示验证对象读出图片中的物体的名称。
第三种可能的实现方式，终端显示视频形式的该唇语提示信息；
在终端的显示界面上显示一个视频形式的唇语提示信息，该唇语提示信息提示验证对象回答出视频中提出的问题的答案。
第四种可能的实现方式，终端播放音频形式的该唇语提示信息。
终端播放一段音频形式的该唇语提示信息，该唇语提示信息提示验证对象读出与该音频中一致的该唇语提示信息。
在步骤1904，终端采集验证对象的至少一帧图像。
终端采集验证对象的至少一帧图像。当终端具有前置摄像头时，终端可以通过拍摄照片的方式采集到验证对象的至少一帧图像。当终端具有前置摄像头时，终端可以通过拍摄视频的方式采集到验证对象的至少一帧图像。
在步骤1905中，终端将该至少一帧图像发送给服务器。
在步骤1906中，服务器接收终端发送的该至少一帧图像。
在步骤1907中，在至少一帧图像中包括唇部变化时，服务器检测唇部变化与唇语提示信息是否匹配。
通过人脸识别技术，可以确定至少一帧图像中是否存在人脸。当该至少一帧图像中存在人脸时，可以通过唇部颜色与肤色的区别确定人脸中的唇部的位置。当获取到唇部的位置时，可以采用模型点的方法确定该图像中的验证对象是否存在唇部变化。
该步骤1907可以包括但不限于如下子步骤：
步骤一，服务器提取至少一帧图像中的唇部图像序列，该唇部图像序列用于表征验证对象的唇部变化；
服务器通过人脸识别技术，可以确定采集到的至少一帧图像中是否存在人脸。当该至少一帧图像中存在人脸时，可以通过唇部颜色与肤色的区别确定人脸中的唇部的位置。当获取到唇部的位置时，可以采用模型点的方法确定该图像中的验证对象是否存在唇部变化。
模型点是指位于唇部边缘曲线上的角点。服务器可以采用模型点的方法确定唇部各点的位置，构成某一时刻的唇部图像的一个状态。按照时间顺序，终端提取每一帧图像中的唇部图像，构成一个唇部图像序列。该唇部图像序列用于表征验证对象的唇部变化。
步骤二，服务器计算唇部图像序列与唇语提示信息所对应的字符串匹配的概率值；
唇语提示信息对应一个字符串，服务器根据提取到的唇部图像序列和唇语提示信息对应的字符串计算两者匹配的概率值。
该步骤二可以包括但不限于如下子步骤：
1，服务器根据与唇语提示信息对应的字符串中的字符排列，在预设的隐马尔科夫模型中选择与该字符串所对应的预测模型序列，隐马尔科夫模型包括n个预测模型，每个预测模型与一个字符对应，n≥1；
比如，字符串为“WORD”，包含四个字符。服务器选择与“W”、“O”、“R”、“D”四个字符分别对应的预测模型。
2，服务器识别唇部图像序列与预测模型序列的匹配度；
比如，服务器识别唇部图像序列中与“W”字符对应的预测模型的匹配度为90％，服务器识别唇部图像序列中与“O”字符对应的预测模型的匹配度为95％，服务器识别唇部图像序列中与“R”字符对应的预测模型的匹配度为98％；服务器识别唇部图像序列中与“R”字符对应的预测模型的匹配度为90％。
3，服务器根据匹配度计算唇部图像序列与字符串匹配的概率值；
比如，唇部图像序列和与字符串中四个字符对应的预测模型的匹配度分别为90％、95％、98％和90％，则服务器根据匹配度计算唇部图像序列与该字符串匹配的概率值为90％*95％*98％*90％＝75％。
步骤三，服务器检测概率值是否大于预设阈值。
比如，预设阈值为80％，75％<80％，故该概率值小于预设阈值，即唇部变化与该唇语提示信息不匹配。
步骤1908，服务器向终端反馈检测结果。
该检测结果可以是唇部变化与该唇语提示信息匹配，也可以是唇部变化与该唇语提示信息不匹配。
步骤1909，若唇部变化与唇语提示信息匹配，则终端确定验证对象通过身份验证。
若唇部变化与唇语提示信息匹配，则终端确定验证对象通过身份验证；若唇部变化与唇语提示信息不匹配，则终端确定验证对象为通过身份验证。
综上所述，本实施例提供的身份验证方法，通过向验证对象提供唇语提示信息；采集验证对象的至少一帧图像；在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示 信息是否匹配；若唇部变化与唇语提示信息匹配，则确定验证对象通过身份验证；解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
本实施例提供的身份验证方法，还通过提取至少一帧图像中的唇部图像序列，该唇部图像序列用于表征验证对象的唇部变化；计算唇部图像序列与唇语提示信息所对应的字符串匹配的概率值；检测概率值是否大于预设阈值；达到了更可靠地检测唇部变化与唇语提示信息是否匹配的效果。
请参考图20，其示出了本发明一个实施例提供的身份验证方法的方法流程图。本实施例以该身份验证别方法应用于终端和服务器中来举例说明。该方法包括步骤2001至步骤2010。
在步骤2001中，终端生成唇语提示信息。
终端生成唇语提示信息。
该唇语提示信息可以是随机生成的提示信息。比如，在终端中存在一个唇语提示信息库，当需要进行身份验证时，终端从该唇语提示信息库中随机抽取一个唇语提示信息作为终端生成的唇语提示信息。
唇语提示信息可以是文字、图片或者视频，每个唇语提示信息对应一个字符串。
在步骤2002，终端向被鉴别对象提供唇语提示信息。
终端向验证对象提供唇语提示信息。本实施例中可以通过四种方式中的任意一种向被鉴别对象提供唇语提示信息，此处可参见如图19所示实施例，再次不做赘述。
在步骤2003中，终端向服务器发送该唇语提示信息。
在唇语提示信息是终端生成的信息时，向服务器发送该唇语提示信息。
在步骤2004中，服务器接收终端发送的该唇语提示信息。
在步骤2005中，终端采集验证对象的至少一帧图像。
终端采集验证对象的至少一帧图像。当终端具有前置摄像头时，终端可以通过前置摄像头采集到验证对象的至少一帧图像。
在步骤2006中，终端将该至少一帧图像发送给服务器。
在步骤2007中，服务器接收终端发送的该至少一帧图像。
需要说明的是，在其它实施例中，终端可以将唇语提示信息和至少一帧图像同时发送给服务器。
在步骤2008，在至少一帧图像中包括唇部变化时，服务器检测唇部变化与唇语提示信息是否匹配。
通过人脸识别技术，可以确定至少一帧图像中是否存在人脸。当该至少一帧图像中存在人脸时，可以通过唇部颜色与肤色的区别确定人脸中的唇部的位置。当获取到唇部的位置时，可以采用模型点的方法确定该图像中的验证对象是否存在唇部变化。
该步骤2008可以包括但不限于如下子步骤2008A至步骤2008C。
在步骤2008A中，服务器提取至少一帧图像中的唇部图像序列，该唇部图像序列用于表征验证对象的唇部变化。
服务器提取至少一帧图像中的唇部图像序列，该唇部图像序列用于表征验证对象的唇部变化。服务器采用模型点的方法确定唇部各点的位置，构成某一时刻的唇部图像的一个状态。按照时间顺序，服务器提取每一帧图像中的唇部图像，构成一个唇部图像序列。该唇部图像序列用于表征验证对象的唇部变化。
在步骤2008B中，服务器计算唇部图像序列与唇语提示信息所对应的字符串匹配的概率值。
唇语提示信息对应一个字符串，服务器根据提取到的唇部图像序列和唇语提示信息对应的字符串计算两者匹配的概率值。
该步骤2008B可以包括但不限于如下子步骤：
步骤508B-1，服务器根据与唇语提示信息对应的字符串中的字符排列，在预设的隐马尔科夫模型中选择与该字符串所对应的预测模型序列，隐马尔科夫模型包括n个预测模型，每个预测模型与一个字符对应，n≥1；
比如，字符串为“WORD”，包含四个字符。服务器选择与“W”、“O”、“R”、“D”四个字符分别对应的预测模型。
步骤508B-2，服务器识别唇部图像序列与预测模型序列的匹配度；
比如，服务器识别唇部图像序列中与“W”字符对应的预测模型的匹配度为90％，服务器识别唇部图像序列中与“O”字符对应的预测模型的匹配度为95％，服务器识别唇部图像序列中与“R”字符对应的预测模型的匹配度为98％；服务器识别唇部图像序列中与“R”字符对应的预测模型的匹配度为90％。
步骤508B-3，服务器根据匹配度计算唇部图像序列与字符串匹配的概率值；
比如，唇部图像序列和与字符串中四个字符对应的预测模型的匹配度分别为90％、95％、98％和90％，则服务器根据匹配度计算唇部图像序列与该字符串匹配的概率值为90％*95％*98％*90％＝75％。
在步骤2008C中，服务器检测概率值是否大于预设阈值。
比如，预设阈值为80％，75％<80％，故该概率值小于预设阈值，即唇部变化与该唇语提示信息不匹配。
在步骤2009中，服务器向终端反馈检测结果。
该检测结果可以是唇部变化与该唇语提示信息匹配，也可以是唇部变化与该唇语提示信息不匹配。
在步骤2010，若唇部变化与唇语提示信息匹配，则终端确定验证对象通过身份验证。。
若唇部变化与唇语提示信息匹配，则终端确定验证对象通过身份验证；若唇部变化与唇语提示信息不匹配，则终端确定验证对象未通过身份验证。
综上所述，本实施例提供的身份验证方法，通过向验证对象提供唇语提示信息；采集验证对象的至少一帧图像；在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示信息是否匹配；若唇部变化与唇语提示信息匹配，则确定验证对象通过身份验证；解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
本实施例提供的身份验证方法，还通过提取至少一帧图像中的唇部图像序列，该唇部图像序列用于表征验证对象的唇部变化；计算唇部图像序列与唇语提示信息所对应的字符串匹配的概率值；检测概率值是否大于预设阈值；达到了更可靠地检测唇部变化与唇语提示信息是否匹配的效果。
请参考图21，其示出了本发明实施例提供的身份验证装置的结构方框图。本实施例以该身份验证装置应用于终端中来举例说明。该身份验证装置可以通过软件、硬件或者两者的结合实现成为终端的全部或者一部分，该装置，包括：
信息提供模块2120，用于向验证对象提供唇语提示信息；
图像采集模块2140，用于采集验证对象的至少一帧图像；
第一检测模块2160，在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示信息是否匹配；
确定模块2180，用于若唇部变化与唇语提示信息匹配，则确定验证对象通过身份验证。
综上所述，本实施例提供的身份验证装置，通过向验证对象提供唇语提示信息；采集验证对象的至少一帧图像；在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示信息是否匹配；若唇部变化与唇语提示信息匹配，则确定验证对象通过身份验证；解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
请参考图22，其示出了本发明另实施例提供的身份验证装置的结构方框图。本实施例以身份验证装置应用于终端中来举例说明。该身份验证装置可以通过软件、硬件或者两者的结合实现成为终端的全部或者一部分，该装置，包括：
信息提供模块2220，用于向验证对象提供唇语提示信息；
图像采集模块2240，用于采集验证对象的至少一帧图像；
第一检测模块2260，在至少一帧图像中包括唇部变化时，检测所述唇部变化与所述唇语提示信息是否匹配；
确定模块2280，用于若所述唇部变化与所述唇语提示信息匹配，则确定验证对象通过身份验证。
可选地，所述第一检测模块2260，包括：
第一提取单元2262，用于提取所述至少一帧图像中的唇部图像序列，所述唇部图像序列用于表征验证对象的唇部变化；
第一计算单元2264，用于计算所述唇部图像序列与所述唇语提示信息所对应的字符串匹配的概率值；
第三检测单元2266，用于检测所述概率值是否大于预设阈值。
可选地，所述第一计算单元2264，包括：
第一选择子单元22642，用于根据与所述唇语提示信息对应的字符串中的字符排列，在预设的隐马尔科夫模型中选择与所述字符串所对应的预测模型序列，所述隐马尔科夫模型包括n个预测模型，每个预测模型与一个字符对应，n≥1；
第一识别子单元22644，用于识别所述唇部图像序列与所述预测模型序列的匹配度；
第三计算子单元22646，用于根据所述匹配度计算所述唇部图像序列与所述字符串匹配的概率值。
可选地，所述第一检测模块2260，包括：
第一检测单元2267，用于在所述唇语提示信息是终端生成的信息时，将所述至少一帧图像和所述唇语提示信息发送给服务器，所述服务器用于在所述至少一帧图像中包括唇部变化时，检测所述唇部变化与所述唇语提示信息是否匹配；
或，
第二检测单元2268，用于在所述唇语提示信息是从服务器获取的信息时，将所述至少一帧图像发送给所述服务器，所述服务器用于在所述至少一帧图像中包括唇部变化时，检测所述唇部变化与所述唇语提示信息是否匹配。
可选地，所述信息提供模块2220，包括：
第一提供单元2222，用于显示文本形式的所述唇语提示信息；
和/或，
第二提供单元2224，用于显示图像形式的所述唇语提示信息；
和/或，
第三提供单元2226，用于显示视频形式的所述唇语提示信息；
和/或，
第四提供单元2228，用于播放音频形式的所述唇语提示信息。
综上所述，本实施例提供的身份验证装置，通过向验证对象提供唇语提示信息；采集验证对象的至少一帧图像；在至少一帧图像中包括唇部变化时，检测唇部变化与唇语提示信息是否匹配；若唇部变化与唇语提示信息匹配，则确定验证对象通过身份验证；解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
请参考图23，其示出了本发明实施例提供的身份验证装置的结构方框图。本实施例以该身份验证装置应用于服务器中来举例说明。该身份验证装置可以通过软件、硬件或者两者的结合实现成为服务器的全部或者一部分，该装置，包括：
图像接收模块2320，用于接收终端发送的至少一帧图像；
信息获取模块2340，用于获取所述终端向验证对象提供的唇语提示信息；
第二检测模块2360，用于在所述至少一帧图像中包括唇部变化时，检测所述唇部变化与所述唇语提示信息是否匹配；
结果反馈模块2380，用于向所述终端反馈检测结果。
综上所述，本实施例提供的身份验证装置，通过接收终端发送的至少一帧图像；获取所述终端向验证对象提供的唇语提示信息；在所述至少一帧图像中包括唇部变化时，检测所述唇部变化与所述唇语提示信息是否匹配；向所述终端反馈检测结果；解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
请参考图24，其示出了本发明实施例提供的身份验证装置的结构方框图。本实施例以 该身份验证装置应用于服务器中来举例说明。该身份验证装置可以通过软件、硬件或者两者的结合实现成为服务器的全部或者一部分，该装置，包括：
图像接收模块2420，用于接收终端发送的至少一帧图像；
信息获取模块2440，用于获取所述终端向验证对象提供的唇语提示信息；
第二检测模块2460，用于在所述至少一帧图像中包括唇部变化时，检测所述唇部变化与所述唇语提示信息是否匹配；
结果反馈模块2480，用于向所述终端反馈检测结果。
可选地，所述第二检测模块960，包括：
第二提取单元2462，用于提取所述至少一帧图像中的唇部图像序列，所述唇部图像序列用于表征所述验证对象的唇部变化；
第二计算单元2464，用于计算所述唇部图像序列与所述唇语提示信息所对应的字符串匹配的概率值；
第四检测单元2466，用于检测所述概率值是否大于预设阈值。
可选地，所述第二计算单元2464，包括：
第二选择子单元24642，用于根据与所述唇语提示信息对应的字符串中的字符排列，在预设的隐马尔科夫模型中选择与所述字符串所对应的预测模型序列，所述隐马尔科夫模型包括n个预测模型，每个预测模型与一个字符对应，n≥1；
第二识别子单元24644，用于识别所述唇部图像序列与所述预测模型序列的匹配度；
第四计算子单元24646，用于根据所述匹配度计算所述唇部图像序列与所述字符串匹配的概率值。
可选地，所述信息获取模块2440，包括：
第一获取单元2442，用于在所述唇语提示信息是所述终端生成的信息时，接收所述终端发送的所述唇语提示信息；
第二获取单元2444，用于在所述唇语提示信息是服务器生成并发送给所述终端的信息时，读取存储在所述服务器的所述唇语提示信息。
综上所述，本实施例提供的身份验证装置，通过接收终端发送的至少一帧图像；获取所述终端验证对象提供的唇语提示信息；在所述至少一帧图像中包括唇部变化时，检测所述唇部变化与所述唇语提示信息是否匹配；向所述终端反馈检测结果；解决了若非法用户将合法用户的照片放在自身面部前方，仍然能够通过身份认证，导致出现安全隐患的问题。
本发明实施例还提供了一种身份验证装置，如图25所示，该装置可以包括一个或多个处理器2501和存储有操作指令的存储介质2502，当运行存储介质2502中的操作指令时，处理器2501执行如下步骤：
向验证对象提供验证提示信息；
获取身份验证信息，身份验证信息由验证对象根据所述验证提示信息生成；
将身份验证信息与验证提示信息的信息样本进行比较；
当身份验证信息与验证提示信息的信息样本之间的匹配度满足预设值，则验证对象通过验证。
进一步的，处理器2501获取身份验证信息时，执行：
获取验证对象根据所述验证提示信息产生的视频信息；
从视频信息中提取图像信息、声音信息、以及图像信息和声音信息的关联信息。
进一步的，处理器2501将身份验证信息与验证提示信息的信息样本进行比较时，执行：
从图像信息中提取图像特征，从声音信息中提取声纹特征；
分别将图像特征、声纹特征以及关联信息与所述信息样本的图像样本、声纹样本以及关联信息样本进行比较；
图像特征、声纹特征以及关联信息与图像样本、声纹样本以及关联信息样本的匹配项数大于或等于预设阈值时，验证对象通过验证。
进一步的，处理器2501还执行：
从声音信息中提取语音内容；
将语音内容与信息样本的语音内容样本进行比较；
当图像特征、声纹特征、语音内容以及关联信息与所述图像样本、声纹样本、语音内容样本以及关联信息样本的匹配项数大于或等于预设阈值时，验证对象通过验证。
进一步的，处理器2501获取身份验证信息时，执行：
获取验证对象根据验证提示信息所产生的音频数据。
进一步的，处理器2501将所述身份验证信息与所述验证提示信息的信息样本进行比较时，执行：
将音频数据进行文本转换，得到文本数据；
将文本数据与信息样本的基准文本数据进行比较，将音频数据与信息样本的基准音频数据进行比较；
当文本数据与信息样本的基准文本数据相同，且所述音频数据与所述信息样本的基准音频数据满足预设的近似度条件时，验证对象通过验证。
进一步的，处理器2501获取身份验证信息时，执行：
获取验证对象的至少一帧图像，图像包括所述验证对象根据所述验证提示信息所产生的唇部变化。
进一步的，处理器2501将身份验证信息与验证提示信息的信息样本进行比较时，执行：
将唇部变化与验证提示信息的基准唇部变化进行比较。
进一步的，处理器2501将唇部变化与验证提示信息的基准唇部变化进行比较时，执行：
提取所述图像中的唇部图像序列，唇部图像序列用于表征验证对象的唇部变化；
计算唇部图像序列与验证提示信息的基准唇部变化所对应的字符串匹配的概率值；
当概率值大于预设阈值时，验证对象通过验证。

Claims (20)

  1. 一种身份验证方法，包括：
    验证设备向验证对象提供验证提示信息；
    所述验证设备获取身份验证信息，所述身份验证信息由所述验证对象根据所述验证提示信息生成；
    所述验证设备将所述身份验证信息与所述验证提示信息的信息样本进行比较；
    当所述身份验证信息与所述验证提示信息的信息样本之间的匹配度满足预设值，则所述验证对象通过验证。
  2. 根据权利要求1所述的方法，所述验证设备获取身份验证信息包括：
    所述验证设备获取所述验证对象根据所述验证提示信息产生的视频信息；
    所述验证设备从所述视频信息中提取图像信息、声音信息、以及所述图像信息和声音信息的关联信息。
  3. 根据权利要求2所述的方法，所述验证设备将所述身份验证信息与所述验证提示信息的信息样本进行比较，包括：
    所述验证设备从所述图像信息中提取图像特征，从所述声音信息中提取声纹特征；
    所述验证对象分别将所述图像特征、声纹特征以及所述关联信息与所述信息样本的图像样本、声纹样本以及关联信息样本进行比较；
    当所述图像特征、声纹特征以及关联信息与所述图像样本、声纹样本以及关联信息样本的匹配项数大于或等于预设阈值时，所述验证对象通过验证。
  4. 根据权利要求3所述的方法，还包括：
    所述验证设备从所述声音信息中提取语音内容；
    所述验证设备将所述语音内容与所述信息样本的语音内容样本进行比较；
    当所述图像特征、声纹特征、语音内容以及关联信息与所述图像样本、声纹样本、语音内容样本以及关联信息样本的匹配项数大于或等于预设阈值时，所述验证对象通过验证。
  5. 根据权利要求1所述的方法，所述验证设备获取身份验证信息，包括：
    所述验证设备获取所述验证对象根据所述验证提示信息所产生的音频数据。
  6. 根据权利要求5所述的方法，所述验证设备将所述身份验证信息与所述验证提示信息的信息样本进行比较，包括：
    所述验证设备将所述音频数据进行文本转换，得到文本数据；
    所述验证设备将所述文本数据与所述信息样本的基准文本数据进行比较，将所述音频数据与所述信息样本的基准音频数据进行比较。
  7. 根据权利要求8所述的方法，当所述文本数据与所述信息样本的基准文本数据相同，且所述音频数据与所述信息样本的基准音频数据满足预设的近似度条件时，所述验证对象通过验证。
  8. 根据权利要求1所述的方法，所述验证设备获取身份验证信息，包括：
    所述验证设备获取所述验证对象的至少一帧图像，所述图像包括所述验证对象根据所述验证提示信息所产生的唇部变化。
  9. 根据权利要求8所述的方法，所述验证设备将所述身份验证信息与所述验证提示信息的信息样本进行比较，包括：
    所述验证设备将所述唇部变化与所述验证提示信息的基准唇部变化进行比较。
  10. 根据权利要求9所述的方法，所述验证设备将所述唇部变化与所述验证提示信息的基准唇部变化进行比较，包括：
    所述验证设备提取所述图像中的唇部图像序列，所述唇部图像序列用于表征所述验证对象的唇部变化；
    所述验证设备计算所述唇部图像序列与所述验证提示信息的基准唇部变化所对应的字符串匹配的概率值；
    当所述概率值大于预设阈值时，所述验证对象通过验证。
  11. 一种身份验证装置，包括一个或多个处理器和存储有操作指令的存储介质，当运行所述存储介质中的操作指令时，所述处理器执行如下步骤：
    向验证对象提供验证提示信息；
    获取身份验证信息，所述身份验证信息由所述验证对象根据所述验证提示信息生成；
    将所述身份验证信息与所述验证提示信息的信息样本进行比较；
    当所述身份验证信息与所述验证提示信息的信息样本之间的匹配度满足预设值，则所述验证对象通过验证。
  12. 根据权利要求11所述的装置，所述处理器获取身份验证信息时，执行：
    获取所述验证对象根据所述验证提示信息产生的视频信息；
    从所述视频信息中提取图像信息、声音信息、以及所述图像信息和声音信息的关联信息。
  13. 根据权利要求12所述的装置，所述处理器将所述身份验证信息与所述验证提示信息的信息样本进行比较时，执行：
    从所述图像信息中提取图像特征，从所述声音信息中提取声纹特征；
    分别将所述图像特征、声纹特征以及所述关联信息与所述信息样本的图像样本、声纹样本以及关联信息样本进行比较；
    所述图像特征、声纹特征以及关联信息与所述图像样本、声纹样本以及关联信息样本的匹配项数大于或等于预设阈值时，所述验证对象通过验证。
  14. 根据权利要求13所述的装置，所处处理器还执行：
    从所述声音信息中提取语音内容；
    将所述语音内容与所述信息样本的语音内容样本进行比较；
    当所述图像特征、声纹特征、语音内容以及关联信息与所述图像样本、声纹样本、语音内容样本以及关联信息样本的匹配项数大于或等于预设阈值时，所述验证对象通过验证。
  15. 根据权利要求11所述的装置，所述处理器获取身份验证信息时，执行：
    获取所述验证对象根据所述验证提示信息所产生的音频数据。
  16. 根据权利要求15所述的装置，所述处理器将所述身份验证信息与所述验证提示信息的信息样本进行比较时，执行：
    将所述音频数据进行文本转换，得到文本数据；
    将所述文本数据与所述信息样本的基准文本数据进行比较，将所述音频数据与所述信息样本的基准音频数据进行比较；
    当所述文本数据与所述信息样本的基准文本数据相同，且所述音频数据与所述信息样本的基准音频数据满足预设的近似度条件时，所述验证对象通过验证。
  17. 根据权利要求11所述的装置，所述处理器获取身份验证信息时，执行：
    获取所述验证对象的至少一帧图像，所述图像包括所述验证对象根据所述验证提示信息所产生的唇部变化。
  18. 根据权利要求17所述的装置，所述处理器将所述身份验证信息与所述验证提示信息的信息样本进行比较时，执行：
    将所述唇部变化与所述验证提示信息的基准唇部变化进行比较。
  19. 根据权利要求18所述的装置，所述处理器将所述唇部变化与所述验证提示信息的基准唇部变化进行比较时，执行：
    提取所述图像中的唇部图像序列，所述唇部图像序列用于表征所述验证对象的唇部变化；
    计算所述唇部图像序列与所述验证提示信息的基准唇部变化所对应的字符串匹配的概率值；
    当所述概率值大于预设阈值时，所述验证对象通过验证。
  20. 一种非瞬时性的计算机可读存储介质，其上存储有计算机可执行指令，当计算机中运行这些可执行指令时，执行如下步骤：
    向验证对象提供验证提示信息；
    获取身份验证信息，所述身份验证信息由所述验证对象根据所述验证提示信息生成；
    将所述身份验证信息与所述验证提示信息的信息样本进行比较；
    当所述身份验证信息与所述验证提示信息的信息样本之间的匹配度满足预设值，则所述验证对象通过验证。
PCT/CN2015/091975 2014-10-15 2015-10-15 身份验证方法、装置和存储介质 WO2016058540A1 (zh)

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
CN201410545593.X 2014-10-15
CN201410545593.XA CN104967593B (zh) 2014-10-15 2014-10-15 一种身份验证方法、装置和系统
CN201410647776.2A CN104966086B (zh) 2014-11-14 2014-11-14 活体鉴别方法及装置
CN201410647776.2 2014-11-14
CN201410779600.2A CN104965724A (zh) 2014-12-16 2014-12-16 一种切换工作状态的方法和装置
CN201410779600.2 2014-12-16

Publications (1)

Publication Number Publication Date
WO2016058540A1 true WO2016058540A1 (zh) 2016-04-21

Family

ID=55746142

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/091975 WO2016058540A1 (zh) 2014-10-15 2015-10-15 身份验证方法、装置和存储介质

Country Status (1)

Country Link
WO (1) WO2016058540A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111062236A (zh) * 2019-05-05 2020-04-24 杭州魔蝎数据科技有限公司 一种基于人工智能的数据授权方法和装置
CN112861100A (zh) * 2021-02-08 2021-05-28 北京百度网讯科技有限公司 身份验证方法、装置、设备以及存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060190264A1 (en) * 2005-02-22 2006-08-24 International Business Machines Corporation Verifying a user using speaker verification and a multimodal web-based interface
US20100161664A1 (en) * 2008-12-22 2010-06-24 General Instrument Corporation Method and System of Authenticating the Identity of a User of a Public Computer Terminal
CN103037368A (zh) * 2011-09-29 2013-04-10 中国移动通信集团四川有限公司 身份认证方法、装置及系统
CN103324918A (zh) * 2013-06-25 2013-09-25 浙江中烟工业有限责任公司 一种人脸识别与唇形识别相配合的身份认证方法
CN103634120A (zh) * 2013-12-18 2014-03-12 上海市数字证书认证中心有限公司 基于人脸识别的实名认证方法及系统
CN104966086A (zh) * 2014-11-14 2015-10-07 深圳市腾讯计算机系统有限公司 活体鉴别方法及装置
CN104965724A (zh) * 2014-12-16 2015-10-07 深圳市腾讯计算机系统有限公司 一种切换工作状态的方法和装置
CN104967593A (zh) * 2014-10-15 2015-10-07 腾讯科技（深圳）有限公司 一种身份验证方法、装置和系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060190264A1 (en) * 2005-02-22 2006-08-24 International Business Machines Corporation Verifying a user using speaker verification and a multimodal web-based interface
US20100161664A1 (en) * 2008-12-22 2010-06-24 General Instrument Corporation Method and System of Authenticating the Identity of a User of a Public Computer Terminal
CN103037368A (zh) * 2011-09-29 2013-04-10 中国移动通信集团四川有限公司 身份认证方法、装置及系统
CN103324918A (zh) * 2013-06-25 2013-09-25 浙江中烟工业有限责任公司 一种人脸识别与唇形识别相配合的身份认证方法
CN103634120A (zh) * 2013-12-18 2014-03-12 上海市数字证书认证中心有限公司 基于人脸识别的实名认证方法及系统
CN104967593A (zh) * 2014-10-15 2015-10-07 腾讯科技（深圳）有限公司 一种身份验证方法、装置和系统
CN104966086A (zh) * 2014-11-14 2015-10-07 深圳市腾讯计算机系统有限公司 活体鉴别方法及装置
CN104965724A (zh) * 2014-12-16 2015-10-07 深圳市腾讯计算机系统有限公司 一种切换工作状态的方法和装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111062236A (zh) * 2019-05-05 2020-04-24 杭州魔蝎数据科技有限公司 一种基于人工智能的数据授权方法和装置
CN112861100A (zh) * 2021-02-08 2021-05-28 北京百度网讯科技有限公司 身份验证方法、装置、设备以及存储介质
CN112861100B (zh) * 2021-02-08 2023-09-05 北京百度网讯科技有限公司 身份验证方法、装置、设备以及存储介质

Similar Documents

Publication Publication Date Title
US10135818B2 (en) User biological feature authentication method and system
JP6820062B2 (ja) アイデンティティ認証方法ならびに装置、端末及びサーバ
US10268910B1 (en) Authentication based on heartbeat detection and facial recognition in video data
US10540488B2 (en) Dynamic face and voice signature authentication for enhanced security
CN104834849B (zh) 基于声纹识别和人脸识别的双因素身份认证方法及系统
US10275672B2 (en) Method and apparatus for authenticating liveness face, and computer program product thereof
WO2017198014A1 (zh) 一种身份认证方法和装置
US9547760B2 (en) Method and system for authenticating user of a mobile device via hybrid biometics information
US9122913B2 (en) Method for logging a user in to a mobile device
KR102210775B1 (ko) 인적 상호 증명으로서 말하는 능력을 이용하는 기법
US11188628B2 (en) Biometric challenge-response authentication
TWI578181B (zh) 電子裝置、身份驗證系統及方法
CN109558106A (zh) 信息处理终端和方法、用于信息处理的系统及存储介质
WO2019090834A1 (zh) 一种基于声纹的快递柜取件方法及装置
JP2018533142A (ja) 電子署名セッションのオーサーシップを認証するためのシステム及び方法
US9792421B1 (en) Secure storage of fingerprint related elements
EP3001343B1 (en) System and method of enhanced identity recognition incorporating random actions
US20210320801A1 (en) Systems and methods for multi-factor verification of users using biometrics and cryptographic sequences
JPWO2017170384A1 (ja) 生体データ処理装置、生体データ処理システム、生体データ処理方法、生体データ処理プログラム、生体データ処理プログラムを記憶する記憶媒体
WO2023173686A1 (zh) 检测方法、装置、电子设备及存储介质
JP7428242B2 (ja) 認証装置、認証システム、認証方法および認証プログラム
CN106469269B (zh) 一种密码管理的方法、装置及终端
WO2016058540A1 (zh) 身份验证方法、装置和存储介质
JP2022117025A (ja) 本人確認方法、プログラム、及び情報システム
US20180349579A1 (en) Authentication by familiar media fragments

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15851283

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 30.08.2017)

122 Ep: pct application non-entry in european phase

Ref document number: 15851283

Country of ref document: EP

Kind code of ref document: A1