WO2017198014A1 - 一种身份认证方法和装置 - Google Patents
一种身份认证方法和装置 Download PDFInfo
- Publication number
- WO2017198014A1 WO2017198014A1 PCT/CN2017/080196 CN2017080196W WO2017198014A1 WO 2017198014 A1 WO2017198014 A1 WO 2017198014A1 CN 2017080196 W CN2017080196 W CN 2017080196W WO 2017198014 A1 WO2017198014 A1 WO 2017198014A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- audio
- video stream
- target object
- feature
- voiceprint
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000012545 processing Methods 0.000 claims description 14
- 239000013598 vector Substances 0.000 description 11
- 238000001514 detection method Methods 0.000 description 8
- 230000001815 facial effect Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 206010001488 Aggression Diseases 0.000 description 1
- 230000016571 aggressive behavior Effects 0.000 description 1
- 208000012761 aggressive behavior Diseases 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
- H04L9/3226—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials using a predetermined code, e.g. password, passphrase or PIN
- H04L9/3231—Biological data, e.g. fingerprint, voice or retina
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/24—Speech recognition using non-acoustical features
- G10L15/25—Speech recognition using non-acoustical features using position of the lips, movement of the lips or face analysis
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/06—Decision making techniques; Pattern matching strategies
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/22—Interactive procedures; Man-machine interfaces
-
- 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
- H04L9/40—Network security protocols
Definitions
- the present application relates to network technologies, and in particular, to an identity authentication method and apparatus.
- an attacker can perform face recognition through simulated video or video, even if at least one authentication method such as face and voiceprint is used for verification, and each authentication method is used.
- the comparison is independent, and the attacker can break through.
- the present application provides an identity authentication method and apparatus to improve the efficiency and reliability of identity authentication.
- an identity authentication method comprising:
- the object identifier is included in the pre-stored object registration information, obtain the template physiological feature corresponding to the object identifier in the object registration information;
- the physiological characteristics of the target object are compared with the physiological characteristics of the template to obtain a comparison result. If the comparison result satisfies the authentication condition, it is confirmed that the target object passes the authentication.
- an identity authentication apparatus comprising:
- An information acquiring module configured to acquire the collected audio and video stream, where the audio and video stream is generated by a target object to be authenticated;
- An identifier determining module configured to determine whether the lip language and the voice in the audio and video stream meet the consistency, and if the consistency is satisfied, the voice content obtained by performing voice recognition on the audio stream in the audio and video stream is used as a The object identifier of the target object;
- An information management module configured to: if the object registration information is pre-stored, include the object identifier, and acquire, in the object registration information, a template physiological feature corresponding to the object identifier;
- a feature recognition module configured to perform physiological recognition on the audio and video stream to obtain a physiological feature of the target object
- the authentication processing module is configured to compare the physiological characteristics of the target object with the physiological characteristics of the template to obtain a comparison result, and if the comparison result satisfies the authentication condition, confirm that the target object passes the authentication.
- the identity authentication method and apparatus simplifies the method of verifying the user identification according to the audio and video stream identification of the user at the time of authentication, and can also verify the facial features and voiceprint features through the same audio and video stream.
- the user operation improves the authentication efficiency, and also maintains the 1:1 authentication mode to ensure the accuracy of the recognition.
- the method also ensures that the target object is a living body through the consistency judgment of the lip language and the voice, preventing the attacker.
- Forged video Video recording improves the security and reliability of authentication.
- FIG. 3 is a flowchart of a face feature recognition according to an exemplary embodiment of the present application.
- FIG. 4 is a flow chart of voiceprint feature recognition according to an exemplary embodiment of the present application.
- FIG. 5 is an identity authentication process according to an exemplary embodiment of the present application.
- FIG. 6 is a structural diagram of an identity authentication apparatus according to an exemplary embodiment of the present application.
- FIG. 7 is a structural diagram of an identity authentication apparatus according to an exemplary embodiment of the present application.
- the identity authentication method provided by the embodiment of the present application can be applied to Internet identity authentication. For example, when logging in to a network application, the identity authentication by the method allows login, thereby ensuring security of application use.
- the application can be run on a user's smart device, such as a smart phone or a smart tablet.
- the audio and video stream can be collected through the camera and the microphone on the smart device.
- the user can read his own application ID to the camera and microphone of the mobile phone.
- the ID can be the account registered by the user in the application. "123456", when the user finishes reading aloud, the mobile phone can collect the user's audio and video stream, including the user's video image and the spoken voice.
- the identity authentication method in the embodiment of the present application may be processed based on the collected audio and video stream.
- the user Before performing the authentication, the user also needs to perform an identity registration process, and then perform identity authentication according to the registered information, and the registration process is also based on the foregoing. Collect audio and video streams.
- the identity registration process and the identity authentication process are respectively described as follows.
- the processing of the identity registration or authentication does not limit the execution device in the actual application. For example, after the smart phone collects the audio and video stream of the user, the voice can be sounded.
- the video stream is transmitted to the server processing of the application backend, or part of the processing is on the client side of the smartphone, and the other part is processed on the server side, or other methods may be used.
- the user may include two types of information when performing identity registration, where the type of information is: an object identifier, for example, an example of logging in to an application by a user, the user may be referred to as a target object, when the user When the application is registered, the information that the user uses to distinguish it from other users in the application is the object identifier, for example, the account 123456 of the user in the application, and the account 123456 is the object identifier of the target object.
- Another type of information is physiological information that can uniquely identify a user, such as a user's voiceprint feature, or a user's facial features. Usually, different people's voiceprints and faces are different, and the physiological characteristics of each user can be identified. Sexual information is called a template physiological feature.
- object registration information The above-mentioned object identifier and template physiological characteristics are associated with each other and stored, and the object identifier and the template physiological feature of the corresponding stored target object may be referred to as “object registration information”.
- object registration information For example, the user Xiao Zhang may store its object registration information as “123456 - Template Physiological Feature A”, wherein in order to more accurately identify the user, the type of physiological information included in the template physiological feature used in the present example may be at least two. Kinds of, for example, faces and voice prints.
- Figure 1 illustrates the identity registration process in an example, including the following:
- step 101 an audio and video stream to be registered of the target object is acquired.
- the user can read his/her application account "123456" against his or her smart device such as a mobile phone.
- the user who is registering can be referred to as the target object, and the camera and the microphone of the smart device can collect the audio and video stream when the user reads aloud, and the audio and video stream collected during registration can be referred to as the audio and video stream to be registered.
- the audio stream is the voice that the user reads aloud
- the video stream is the video image when the user reads aloud.
- step 102 it is determined whether the lip language and the voice in the audio and video stream to be registered satisfy the consistency.
- the consistency here refers to whether the motion of the lip and the motion represented by the voice can correspond. For example, suppose a voice is "the weather is fine today", the voice is slowly and slowly read, and the speech rate is slow. A lip movement is a quick reading of the movement used by "Today's sunny weather”. Obviously these two are not right. When the lip movement has stopped (the content has been read), the voice continues (.. ..Sunny weather). This situation may occur, for example, when an attacker tries to pass the user ID and face detection, it can attack the face detection through a previous video recording of the user (the attacked user), and the attacker reads the user himself. The ID attacks the identification of the voice content ID. If it is attacked separately, it is possible to pass the authentication. However, in this attack mode, the lip language and the voice are inconsistent, and it is possible to identify through the consistency judgment that the person is not reading aloud.
- step 102 if the result of the judgment is that the lip language and the voice in the to-be-registered audio and video stream do not satisfy the consistency, the user may be prompted to register failure, or as shown in FIG. 1, go to step 101 to re-acquire the audio and video stream. Prevent mistakes.
- step 103 may be performed, and the voice content obtained by performing voice recognition according to the audio stream in the collected audio and video stream is used as the object identifier of the target object.
- Speech recognition uses computer technology to automatically recognize the content of a person's spoken voice, that is, the process from voice to content recognition. For example, after the voice of the audio to be registered is “123456”, the voice content in the audio stream is “123456”, and the identified content can be used as the identifier of the user, that is, the user ID.
- the above-mentioned speech recognition of the audio stream may be performed by identifying the audio stream of the speech after determining that the lip language and the speech satisfy the consistency; or, in the process of determining whether the lip language and the voice satisfy the consistency.
- the object identifier is obtained by recognizing the audio stream.
- Another aspect of the process is to perform physiological recognition on the registered audio and video stream to obtain a template physiological feature of the audio and video stream to be registered.
- the physiological features are exemplified by facial features and voiceprint features, but are not limited to these two features, as long as they are physiological features that can uniquely identify the user and distinguish different users.
- the voice stream in the audio and video stream to be registered may be voiceprinted to obtain the voiceprint feature of the target object.
- the face detection is performed on the video stream in the registered audio and video stream, and the face feature of the target object is obtained.
- the face feature obtained by the above detection may be referred to as a template face feature as a standard in the subsequent authentication process.
- the detected voiceprint feature is called a template voiceprint feature, and the template voiceprint is used.
- Features and template face features can be collectively referred to as template physiological features.
- the template physiological feature and the object identifier of the target object are referred to as object registration information.
- the object identifier of the target object and the corresponding template physiological feature are The object registration information is stored in the database.
- the execution order of these three aspects is not For example, after obtaining the audio and video stream to be registered in step 101, the above three aspects of processing may be performed in parallel. If the lip language and the voice are inconsistent, the recognized voiceprint feature and the face feature may not be stored; or, Firstly, the judgment of lip language and speech consistency is performed, and after the consistency is determined, the detection and recognition of the voiceprint feature and the face feature are performed.
- FIG. 2 illustrates the flow of the lip language and speech consistency judgment in FIG. 1, which may include:
- step 201 endpoint detection is performed according to the audio stream in the audio and video stream to be registered. This step can detect the start time and end time of the audio stream from a continuous audio stream.
- step 202 continuous speech feature extraction is performed based on the audio stream, including but not limited to MFCC features, LPCC features.
- the features extracted in this step can be used for speech recognition.
- step 203 the speech single character and the corresponding time point in the audio stream are identified.
- each single character in the audio stream can be identified according to the voice features extracted in step 202, and the corresponding time points of occurrence and disappearance are determined.
- the method for speech recognition includes, but is not limited to, a Hidden Markov Mode (HMM), a Deep Neural Networ (DNN), and a Long Short Time Model (LSTM).
- HMM Hidden Markov Mode
- DNN Deep Neural Networ
- LSTM Long Short Time Model
- step 204 the location of the lip is detected based on the video stream in the audio and video stream to be registered. This step detects the location of the lip from the video image.
- the quality of the detected lip image is determined. For example, parameters such as the definition of the position of the lip and the degree of exposure can be determined. If the definition is insufficient or the exposure is too high, the quality is judged as unqualified. You can return to reacquiring the audio and video streams to be registered. If the quality is acceptable, proceed to step 206 to continue the lip language recognition.
- step 206 lip continuous feature extraction is performed. This step may extract features from successive lip images, including but not limited to bare pixels, or LBP, Gabor, Partial image descriptors such as SIFT and Surf.
- step 207 the lip language single character and the corresponding time point in the video stream are identified.
- the lip language character recognition in this step may use a hidden Markov (HMM) or a long and short time memory model, and the corresponding time point of the single lip language character in the video time series is also determined by the model when performing lip language recognition.
- HMM hidden Markov
- a long and short time memory model the corresponding time point of the single lip language character in the video time series is also determined by the model when performing lip language recognition.
- step 208 it is determined whether the single character of the lip language and the voice and the corresponding time point satisfy the consistency. For example, in this step, the time point information of the voice single character can be compared with the time point information of the lip language single character. If the comparison result is consistent, the audio stream is considered to be a real person, and step 209 is continued; if not, If it is suspected of being an aggressive behavior, it will return to restart the registration process.
- the method for detecting the consistency of the characters and the corresponding time points of the lip language and the voice is more refined, and the judgment of the human voice can be more accurate.
- voice recognition may be performed according to the voice features extracted in step 202 to obtain a user ID, that is, an object identifier.
- the method for speech recognition includes, but is not limited to, a Hidden Markov Mode (HMM), a Deep Neural Networ (DNN), and a Long Short Time Model (LSTM).
- HMM Hidden Markov Mode
- DNN Deep Neural Networ
- LSTM Long Short Time Model
- the speech recognition of the audio stream may be performed in step 209 after determining that the lip language and the speech satisfy the consistency; or, in step 203, the single in the audio stream may be identified.
- the voice recognition is performed according to the voice feature to obtain the user ID.
- the previously identified user ID can be directly used as the object identifier.
- FIG. 3 illustrates the flow of face feature recognition in FIG. 1 and may include:
- a face image is detected according to a video stream in the audio and video stream to be registered.
- the video frame image may be extracted from the video stream in the audio and video stream, and the presence or absence of a face may be detected. If it occurs, the process proceeds to 302, otherwise the process returns to continue.
- step 302 the quality of the face image is detected.
- This step can be checked in step 301.
- the detected face performs face feature point detection, and judges the angle of the face in the horizontal direction and the vertical direction according to the result of the feature point detection. If both are within a certain inclination angle, the quality requirement is satisfied, otherwise, the quality is not satisfied. Requirements; at the same time, the parameters such as the sharpness and exposure of the face area are also required to meet the requirements within a certain threshold range. If the quality of the face image is good, the face feature can be better recognized.
- feature vectors may be extracted from the face image for the face image satisfying the quality requirement, including but not limited to: Local Binary Pattern (LBP), Gabor feature, volume Convolutional Neural Network (CNN) and so on.
- LBP Local Binary Pattern
- CNN volume Convolutional Neural Network
- step 304 the plurality of face feature vectors extracted in step 303 are fused or combined to form a unique face feature of the user, that is, a template face feature.
- FIG. 4 illustrates the flow of voiceprint feature recognition in FIG. 1, which may include:
- step 401 an audio stream in the audio and video stream to be registered is acquired.
- the voiceprint feature recognition of this example can be performed according to the audio stream in the audio and video stream to be registered.
- step 402 it is determined that the audio quality of the audio stream satisfies the quality criteria.
- the audio quality can be judged.
- the quality of the collected audio stream is good, the effect of voiceprint recognition on the audio is better. Therefore, the audio stream can be prior to the subsequent voiceprint recognition.
- Quality is judged.
- the vocal signal strength, the signal-to-noise ratio, and the like in the audio stream can be calculated to determine whether the voice meets the quality standard condition.
- the quality standard condition can be that the signal-to-noise ratio is within a certain range.
- the acoustic signal strength is higher than a certain intensity threshold and the like. If the quality passes, step 403 may continue; otherwise, the audio and video streams to be registered may be re-acquired.
- step 403 the voiceprint feature vector is extracted from the audio stream.
- the number of audio and video streams to be registered may be multiple.
- the user can read his own user ID twice, and correspondingly collect two audio and video streams.
- the voiceprint feature vector of the audio stream in each of the audio and video streams the feature vector can be extracted in a plurality of conventional manners, and will not be described in detail.
- the voice feature parameter MFCC can be extracted from the voice signal of the audio stream ( Mel Frequency Cepstrum Coefficient, then feature vectors are calculated using i-vector (a speaker recognition algorithm) and PLDA (Probabilistic Linear Discriminant Analysis).
- step 404 it is determined whether the voiceprint feature vectors of the plurality of audio streams satisfy the consistency.
- the collected audio stream is at least two corresponding.
- the voiceprint consistency between the plurality of audio streams can be determined.
- the similarity score between the plurality of audio streams may be calculated based on the voiceprint feature vector extracted by each audio stream in step 403.
- step 405 may be continued; otherwise, the multiple audio differences input by the user are too large, and the user who is registering may be re-read. Its user ID, that is, re-acquisition of the audio stream.
- a template voiceprint feature is generated from the voiceprint feature vectors of the plurality of audio streams.
- the voiceprint feature vectors extracted by the respective audio streams are weighted and summed according to the previous steps to obtain a template voiceprint feature.
- the object registration information of the target object is already stored in the database, and the object registration information may include the object identifier and the corresponding template physiological feature, and the template physiological feature may include the template voiceprint feature and the template face.
- the feature as follows, can perform identity authentication processing of the object based on the object registration information.
- Figure 5 illustrates an example of an identity authentication process in which the authentication is used.
- the physiological features are described by taking the integrated facial features and voiceprint features as an example, and the physiological features can be compared after determining that the target object being authenticated is a living object rather than a video.
- the authentication process includes the following processing:
- step 501 the collected audio and video streams are acquired, and the audio and video streams are generated by a target object to be authenticated.
- the user can open the application client on the smart device such as the smart phone, and the user can collect the audio and video stream to be authenticated through the camera and the microphone of the smart phone, and the audio and video stream can be the user reading the application. ID.
- step 502 it is determined whether the lip language and the voice in the audiovisual stream satisfy the consistency.
- step 503 If the consistency is met, indicating that the target object being authenticated is a live object instead of a video, etc., proceed to step 503; otherwise, return to performing 501 re-acquisition.
- voice recognition is performed on the audio stream in the audio and video stream to obtain voice content of the audio stream.
- the recognized voice content may be the user ID "123456.”
- step 504 the voice content is used as the object identifier of the target object, and it is determined whether the object identifier is included in the pre-stored object registration information.
- the template physiological features corresponding to the object identifier such as the template face feature and the template voiceprint feature, may be acquired in the object registration information, and the tone to be authenticated is continued.
- the video stream is physiologically recognized, and the physiological characteristics of the target object are obtained to be compared with the physiological characteristics of the template. If the object identifier is not included in the pre-stored object registration information, the user may be prompted not to register.
- step 505 voiceprint recognition is performed on the audio and video stream to obtain a voice signature of the target object. Sign.
- the extraction of the voiceprint features of this step can be seen in Figure 4.
- step 506 face recognition is performed on the audio and video stream to obtain a face feature of the target object.
- the physiological characteristics of the target object can be compared with the physiological characteristics of the template to obtain a comparison result. If the comparison result satisfies the authentication condition, it is confirmed that the target object passes the authentication. For example, the following steps 507 to 509 are included.
- step 507 the voiceprint feature of the target object is compared with the template voiceprint feature to obtain a voiceprint comparison score.
- step 508 the face feature of the target object is compared with the template face feature to obtain a face comparison score.
- step 509 it is determined whether the voiceprint comparison score and the face comparison score satisfy the authentication condition.
- the voiceprint comparison score and the face comparison score satisfy at least one of the following, it is confirmed that the target object passes the authentication: the voiceprint comparison score is greater than the voiceprint score threshold, and the face comparison score Greater than a face score threshold; or, the product of the voiceprint comparison score and the face comparison score is greater than a corresponding product threshold; or, the weighted sum of the voiceprint comparison score and the face comparison score is greater than a corresponding Weighted threshold.
- the voice recognition of the audio stream is processed by the user ID, which can be performed after determining that the lip language and the voice satisfy the consistency, and can also identify the audio stream.
- the user ID is obtained at the same time in the single character time point. In the above example, the user ID is identified after determining that the lip language and the voice satisfy the consistency.
- the identity authentication method in the embodiment of the present application is such that when the user authenticates, only one
- the secondary audio and video stream can be used.
- the user can read his own user ID once.
- the method can perform voice recognition according to the audio audio and video stream to obtain the user ID, and can also verify the facial features through the same audio and video stream.
- the voice pattern feature this method not only simplifies the user operation, improves the authentication efficiency, but also maintains the 1:1 authentication mode, that is, the recognized physiological features are only compared with the features corresponding to the object identifiers in the database, and the guarantee is ensured.
- the method also ensures that the target object is a living body through the consistency judgment of the lip language and the voice, prevents the video recording of the forgery of the attacker, and improves the security and reliability of the authentication; the user ID in the method
- the identified physiological features are all based on the same audio and video stream, and can identify the attacker's fake audio and video streams to a certain extent.
- the embodiment of the present application further provides an identity authentication device.
- the device may include: an information acquiring module 61, an identifier determining module 62, an information management module 63, and a feature recognition module. 64 and authentication processing module 65.
- the information acquiring module 61 is configured to acquire the collected audio and video stream, where the audio and video stream is generated by the target object to be authenticated;
- the identifier determining module 62 is configured to determine whether the lip language and the voice in the audio and video stream meet the consistency, and if the consistency is satisfied, the voice content obtained by performing voice recognition on the audio stream in the audio and video stream is used as An object identifier of the target object;
- the information management module 63 is configured to: if the pre-stored object registration information includes the object identifier, obtain the template physiological feature corresponding to the object identifier in the object registration information;
- a feature recognition module 64 configured to perform physiological recognition on the audio and video stream to obtain a physiological feature of the target object
- the authentication processing module 65 is configured to compare the physiological features of the target object with the template physiological features to obtain a comparison result. If the comparison result satisfies the authentication condition, the target object is confirmed to be authenticated.
- feature recognition module 64 can include: voiceprint recognition Sub-module 641 and face recognition sub-module 642.
- the voiceprint recognition sub-module 641 is configured to perform voiceprint recognition on the audio and video stream to obtain a voiceprint feature of the target object;
- the face recognition sub-module 642 is configured to perform face recognition on the audio and video stream to obtain a face feature of the target object.
- the authentication processing module 65 is configured to compare the voiceprint feature of the target object with the template voiceprint feature, obtain a voiceprint comparison score, and compare the face feature of the target object with the template face feature. And obtaining a face comparison score, and if the voiceprint comparison score and the face comparison score satisfy the authentication condition, confirming that the target object passes the authentication.
- the voiceprint comparison score is greater than the voiceprint score threshold, and the face The comparison score is greater than a face score threshold; or the product of the voiceprint comparison score and the face comparison score is greater than a corresponding product threshold; or the weighted sum of the voiceprint comparison score and the face comparison score Greater than the corresponding weighted threshold.
- the identity determination module 62 can include:
- the character recognition sub-module 621 is configured to perform voice single-character and corresponding time point recognition on the audio stream in the audio-video stream, and perform lip-single character and corresponding time point recognition on the video stream in the audio-video stream;
- the consistent judgment sub-module 622 is configured to determine that the consistency is satisfied if the single character of the voice and the lip language and the corresponding time point are consistent.
- the information acquiring module 61 is further configured to acquire an audio and video stream to be registered of the target object.
- the identifier determining module 62 is further configured to: when the lip language and the voice in the to-be-registered audio and video stream meet the consistency, the voice content obtained by performing voice recognition on the audio stream in the audio and video stream is used as the The object identifier of the target object;
- the feature identification module 64 is further configured to perform physiological recognition on the to-be-registered audio and video stream to obtain the template physiological feature of the to-be-registered audio and video stream;
- the information management module 63 is further configured to store the object identifier of the target object and the corresponding template physiological feature in the object registration information.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Networks & Wireless Communication (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Social Psychology (AREA)
- Psychiatry (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Telephonic Communication Services (AREA)
Abstract
Description
Claims (10)
- 一种身份认证方法,其特征在于,所述方法包括:获取采集到的音视频流,所述音视频流由待认证的目标对象产生;判断所述音视频流中的唇语和语音是否满足一致性,若满足一致性,则将对所述音视频流中的音频流进行语音识别得到的语音内容,作为所述目标对象的对象标识;若预存储的对象注册信息中包括所述对象标识,在所述对象注册信息中获取所述对象标识对应的模版生理特征;对所述音视频流进行生理识别,得到所述目标对象的生理特征;将所述目标对象的生理特征与模版生理特征比对,得到比对结果,若所述比对结果满足认证条件,则确认所述目标对象通过认证。
- 根据权利要求1所述的方法,其特征在于,所述生理特征包括:声纹特征和人脸特征;所述模版生理特征包括:模版人脸特征和模版声纹特征;所述对所述音视频流进行生理识别得到所述目标对象的生理特征,包括:对所述音视频流进行声纹识别,得到所述目标对象的声纹特征;对所述音视频流进行人脸识别,得到所述目标对象的人脸特征;所述将所述目标对象的生理特征与模版生理特征比对,得到比对结果,若所述比对结果满足认证条件,则确认所述目标对象通过认证,包括:将所述目标对象的声纹特征与模版声纹特征比对,得到声纹比对分数;并将所述目标对象的人脸特征与模版人脸特征比对,得到人脸比对分数;若所述声纹比对分数和人脸比对分数满足认证条件,则确认所述目 标对象通过认证。
- 根据权利要求2所述的方法,其特征在于,若所述声纹比对分数和人脸比对分数满足如下至少一种,则确认所述目标对象通过认证:所述声纹比对分数大于声纹分数阈值,且人脸比对分数大于人脸分数阈值;或者,所述声纹比对分数和人脸比对分数的乘积大于对应的乘积阈值;或者,所述声纹比对分数和人脸比对分数的加权和大于对应的加权阈值。
- 根据权利要求1所述的方法,其特征在于,所述判断所述音视频流中的唇语和语音是否满足一致性,若满足一致性,包括:对所述音视频流中的音频流进行语音单字符及对应时间点识别;对所述音视频流中的视频流进行唇语单字符及对应时间点识别;若所述语音和唇语的单字符及对应时间点一致,则确定满足一致性。
- 根据权利要求1所述的方法,其特征在于,所述获取采集到的音视频流之前,所述方法还包括:获取所述目标对象的待注册音视频流;在所述待注册音视频流中的唇语和语音满足一致性时,将对所述音视频流中的音频流进行语音识别得到的语音内容,作为所述目标对象的对象标识;对所述待注册音视频流进行生理识别,得到所述待注册音视频流的所述模版生理特征;将所述目标对象的对象标识及对应的所述模版生理特征,对应存储在所述对象注册信息中。
- 一种身份认证装置,其特征在于,所述装置包括:信息获取模块,用于获取采集到的音视频流,所述音视频流由待认证的目标对象产生;标识确定模块,用于判断所述音视频流中的唇语和语音是否满足一致性,若满足一致性,则将对所述音视频流中的音频流进行语音识别得到的语音内容,作为所述目标对象的对象标识;信息管理模块,用于若预存储的对象注册信息中包括所述对象标识,在所述对象注册信息中获取所述对象标识对应的模版生理特征;特征识别模块,用于对所述音视频流进行生理识别,得到所述目标对象的生理特征;认证处理模块,用于将所述目标对象的生理特征与模版生理特征比对,得到比对结果,若所述比对结果满足认证条件,则确认目标对象通过认证。
- 根据权利要求6所述的装置,其特征在于,所述特征识别模块,包括:声纹识别子模块和人脸识别子模块;所述声纹识别子模块,用于对所述音视频流进行声纹识别,得到所述目标对象的声纹特征;所述人脸识别子模块,用于对所述音视频流进行人脸识别,得到所述目标对象的人脸特征;所述认证处理模块,用于将所述目标对象的声纹特征与模版声纹特征比对,得到声纹比对分数,并将所述目标对象的人脸特征与模版人脸特征比对,得到人脸比对分数,若所述声纹比对分数和人脸比对分数满足认证条件,则确认所述目标对象通过认证。
- 根据权利要求7所述的装置,其特征在于,若所述声纹比对分数和人脸比对分数满足如下至少一种,则确认所述目标对象通过认证:所述声纹比对分数大于声纹分数阈值,且人脸比对分数大于人脸分数阈值;或者,所述声纹比对分数和人脸比对分数的乘积大于对应的乘积阈值;或者,所述声纹比对分数和人脸比对分数的加权和大于对应的 加权阈值。
- 根据权利要求6所述的装置,其特征在于,所述标识确定模块包括:字符识别子模块,用于对所述音视频流中的音频流进行语音单字符及对应时间点识别,对音视频流中的视频流进行唇语单字符及对应时间点识别;一致判断子模块,用于若所述语音和唇语的单字符及对应时间点一致,则确定满足一致性。
- 根据权利要求6所述的装置,其特征在于,所述信息获取模块,还用于获取所述目标对象的待注册音视频流;所述标识确定模块,还用于在所述待注册音视频流中的唇语和语音满足一致性时,将对所述音视频流中的音频流进行语音识别得到的语音内容,作为所述目标对象的对象标识;所述特征识别模块,还用于对所述待注册音视频流进行生理识别,得到所述待注册音视频流的所述模版生理特征;所述信息管理模块,还用于将所述目标对象的对象标识及对应的所述模版生理特征,对应存储在所述对象注册信息中。
Priority Applications (13)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CA3024565A CA3024565C (en) | 2016-05-19 | 2017-04-12 | Identity authentication method and apparatus |
MYPI2018001981A MY192351A (en) | 2016-05-19 | 2017-04-12 | Identity authentication method and apparatus |
MX2018014147A MX2018014147A (es) | 2016-05-19 | 2017-04-12 | Metodo y aparato de autentificacion de identidad. |
BR112018073635-0A BR112018073635A2 (pt) | 2016-05-19 | 2017-04-12 | método de autenticação de identidade e aparelho de autenticação de identidade |
SG11201810131VA SG11201810131VA (en) | 2016-05-19 | 2017-04-12 | Identity authentication method and apparatus |
EP17798578.5A EP3460697B1 (en) | 2016-05-19 | 2017-04-12 | Identity authentication method and apparatus |
JP2018560844A JP2019522840A (ja) | 2016-05-19 | 2017-04-12 | アイデンティティ認証方法及び装置 |
RU2018144787A RU2738325C2 (ru) | 2016-05-19 | 2017-04-12 | Способ и устройство аутентификации личности |
AU2017266971A AU2017266971B2 (en) | 2016-05-19 | 2017-04-12 | Identity authentication method and apparatus |
KR1020187036914A KR102196686B1 (ko) | 2016-05-19 | 2017-04-12 | 신원 인증 방법 및 장치 |
US16/192,401 US10789343B2 (en) | 2016-05-19 | 2018-11-15 | Identity authentication method and apparatus |
PH12018502437A PH12018502437A1 (en) | 2016-05-19 | 2018-11-19 | Identity authentication method and apparatus |
ZA2018/07860A ZA201807860B (en) | 2016-05-19 | 2018-11-21 | Identity authentication method and apparatus |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610340549.4 | 2016-05-19 | ||
CN201610340549.4A CN107404381A (zh) | 2016-05-19 | 2016-05-19 | 一种身份认证方法和装置 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US16/192,401 Continuation US10789343B2 (en) | 2016-05-19 | 2018-11-15 | Identity authentication method and apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017198014A1 true WO2017198014A1 (zh) | 2017-11-23 |
Family
ID=60324817
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2017/080196 WO2017198014A1 (zh) | 2016-05-19 | 2017-04-12 | 一种身份认证方法和装置 |
Country Status (16)
Country | Link |
---|---|
US (1) | US10789343B2 (zh) |
EP (1) | EP3460697B1 (zh) |
JP (2) | JP2019522840A (zh) |
KR (1) | KR102196686B1 (zh) |
CN (1) | CN107404381A (zh) |
AU (1) | AU2017266971B2 (zh) |
BR (1) | BR112018073635A2 (zh) |
CA (1) | CA3024565C (zh) |
MX (1) | MX2018014147A (zh) |
MY (1) | MY192351A (zh) |
PH (1) | PH12018502437A1 (zh) |
RU (1) | RU2738325C2 (zh) |
SG (1) | SG11201810131VA (zh) |
TW (1) | TWI706268B (zh) |
WO (1) | WO2017198014A1 (zh) |
ZA (1) | ZA201807860B (zh) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109271915A (zh) * | 2018-09-07 | 2019-01-25 | 北京市商汤科技开发有限公司 | 防伪检测方法和装置、电子设备、存储介质 |
CN110364163A (zh) * | 2019-07-05 | 2019-10-22 | 西安交通大学 | 一种语音和唇语相融合的身份认证方法 |
CN111160928A (zh) * | 2019-12-16 | 2020-05-15 | 深圳前海微众银行股份有限公司 | 一种验证身份的方法及装置 |
CN111178277A (zh) * | 2019-12-31 | 2020-05-19 | 支付宝实验室(新加坡)有限公司 | 一种视频流识别方法及装置 |
CN111178287A (zh) * | 2019-12-31 | 2020-05-19 | 云知声智能科技股份有限公司 | 一种声像融合的端对端身份识别方法及装置 |
CN111814732A (zh) * | 2020-07-23 | 2020-10-23 | 上海优扬新媒信息技术有限公司 | 一种身份验证方法及装置 |
JP2020535538A (ja) * | 2018-09-07 | 2020-12-03 | 北京市商▲湯▼科技▲開▼▲発▼有限公司Beijing Sensetime Technology Development Co., Ltd. | 偽装防止の検出方法および装置、電子機器、記憶媒体 |
CN114677634A (zh) * | 2022-05-30 | 2022-06-28 | 成都新希望金融信息有限公司 | 面签识别方法、装置、电子设备及存储介质 |
Families Citing this family (55)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6778010B2 (ja) * | 2016-04-13 | 2020-10-28 | キヤノン株式会社 | 画像識別システム、画像識別方法 |
CN107404381A (zh) * | 2016-05-19 | 2017-11-28 | 阿里巴巴集团控股有限公司 | 一种身份认证方法和装置 |
WO2018108263A1 (en) * | 2016-12-14 | 2018-06-21 | Telefonaktiebolaget Lm Ericsson (Publ) | Authenticating a user subvocalizing a displayed text |
US10776467B2 (en) | 2017-09-27 | 2020-09-15 | International Business Machines Corporation | Establishing personal identity using real time contextual data |
US10803297B2 (en) | 2017-09-27 | 2020-10-13 | International Business Machines Corporation | Determining quality of images for user identification |
US10839003B2 (en) | 2017-09-27 | 2020-11-17 | International Business Machines Corporation | Passively managed loyalty program using customer images and behaviors |
US10795979B2 (en) | 2017-09-27 | 2020-10-06 | International Business Machines Corporation | Establishing personal identity and user behavior based on identity patterns |
KR20190046063A (ko) * | 2017-10-25 | 2019-05-07 | 현대자동차주식회사 | 사용자 인증 시스템, 사용자 인증 방법 및 서버 |
US10565432B2 (en) * | 2017-11-29 | 2020-02-18 | International Business Machines Corporation | Establishing personal identity based on multiple sub-optimal images |
CN108052813A (zh) * | 2017-11-30 | 2018-05-18 | 广东欧珀移动通信有限公司 | 终端设备的解锁方法、装置及移动终端 |
KR20230004947A (ko) | 2017-12-08 | 2023-01-06 | 구글 엘엘씨 | 디지털 음성 어시스턴트 컴퓨팅 디바이스들 사이에서 신호 프로세싱 조정 |
US10971173B2 (en) | 2017-12-08 | 2021-04-06 | Google Llc | Signal processing coordination among digital voice assistant computing devices |
CN108074575A (zh) * | 2017-12-14 | 2018-05-25 | 广州势必可赢网络科技有限公司 | 一种基于循环神经网络的身份验证方法及装置 |
TWI661329B (zh) | 2017-12-15 | 2019-06-01 | 鴻海精密工業股份有限公司 | 身份資訊關聯系統與方法、電腦存儲介質及使用者設備 |
CN110288349A (zh) * | 2018-03-16 | 2019-09-27 | 杭州海康威视数字技术股份有限公司 | 用户信息管理方法、装置及系统、存储介质 |
CN108597523B (zh) * | 2018-03-23 | 2019-05-17 | 平安科技(深圳)有限公司 | 说话人认证方法、服务器及计算机可读存储介质 |
CN108712381A (zh) * | 2018-04-16 | 2018-10-26 | 出门问问信息科技有限公司 | 一种身份验证方法及装置 |
CN108446674A (zh) * | 2018-04-28 | 2018-08-24 | 平安科技(深圳)有限公司 | 电子装置、基于人脸图像与声纹信息的身份识别方法及存储介质 |
US10678899B2 (en) * | 2018-05-24 | 2020-06-09 | Nice Ltd. | System and method for performing voice biometrics analysis |
CN108682424A (zh) * | 2018-07-13 | 2018-10-19 | 广州势必可赢网络科技有限公司 | 一种音频采集设备及方法 |
CN109344703B (zh) * | 2018-08-24 | 2021-06-25 | 深圳市商汤科技有限公司 | 对象检测方法及装置、电子设备和存储介质 |
CN109461437B (zh) * | 2018-11-28 | 2023-05-09 | 平安科技(深圳)有限公司 | 唇语识别的验证内容生成方法及相关装置 |
JP7299708B2 (ja) * | 2019-01-15 | 2023-06-28 | グローリー株式会社 | 認証システム、管理装置及び認証方法 |
CN109801638B (zh) * | 2019-01-24 | 2023-10-13 | 平安科技(深圳)有限公司 | 语音验证方法、装置、计算机设备及存储介质 |
CN109872714A (zh) * | 2019-01-25 | 2019-06-11 | 广州富港万嘉智能科技有限公司 | 一种提高语音识别准确性的方法、电子设备及存储介质 |
CN110113319A (zh) * | 2019-04-16 | 2019-08-09 | 深圳壹账通智能科技有限公司 | 身份认证方法、装置、计算机设备和存储介质 |
CN110223710A (zh) * | 2019-04-18 | 2019-09-10 | 深圳壹账通智能科技有限公司 | 多重联合认证方法、装置、计算机装置及存储介质 |
CN110110513A (zh) * | 2019-04-24 | 2019-08-09 | 上海迥灵信息技术有限公司 | 基于人脸和声纹的身份认证方法、装置和存储介质 |
CN110288286A (zh) * | 2019-05-21 | 2019-09-27 | 深圳壹账通智能科技有限公司 | 基于身份验证的物品入库方法、装置、设备及存储介质 |
CN110324314B (zh) * | 2019-05-23 | 2023-04-18 | 深圳壹账通智能科技有限公司 | 用户注册方法及装置、存储介质、电子设备 |
CN110569707A (zh) * | 2019-06-25 | 2019-12-13 | 深圳和而泰家居在线网络科技有限公司 | 一种身份识别方法和电子设备 |
CN110348378A (zh) * | 2019-07-10 | 2019-10-18 | 北京旷视科技有限公司 | 一种认证方法、装置和存储介质 |
CN111684459A (zh) * | 2019-07-18 | 2020-09-18 | 深圳海付移通科技有限公司 | 一种身份验证方法、终端设备、存储介质 |
CN110517106A (zh) * | 2019-07-24 | 2019-11-29 | 合肥善达信息科技有限公司 | 一种用于评标的专家身份认证系统及其实时监测方法 |
TWI822646B (zh) * | 2019-08-07 | 2023-11-11 | 華南商業銀行股份有限公司 | 基於唇部動態影像的身分驗證裝置及方法 |
TWI801647B (zh) * | 2019-08-07 | 2023-05-11 | 華南商業銀行股份有限公司 | 基於動態影像的身分驗證裝置及方法 |
CN110491413B (zh) * | 2019-08-21 | 2022-01-04 | 中国传媒大学 | 一种基于孪生网络的音频内容一致性监测方法及系统 |
CN110717407A (zh) * | 2019-09-19 | 2020-01-21 | 平安科技(深圳)有限公司 | 基于唇语密码的人脸识别方法、装置及存储介质 |
CN110602405A (zh) * | 2019-09-26 | 2019-12-20 | 上海盛付通电子支付服务有限公司 | 拍摄方法和装置 |
CN110738159A (zh) * | 2019-10-11 | 2020-01-31 | 中国建设银行股份有限公司 | 用于实现变更企业实际控制人的在线股东大会方法、装置 |
US11687778B2 (en) | 2020-01-06 | 2023-06-27 | The Research Foundation For The State University Of New York | Fakecatcher: detection of synthetic portrait videos using biological signals |
US11403369B2 (en) | 2020-01-21 | 2022-08-02 | Disney Enterprises, Inc. | Secure content processing pipeline |
US11425120B2 (en) | 2020-02-11 | 2022-08-23 | Disney Enterprises, Inc. | Systems for authenticating digital contents |
US20220318349A1 (en) * | 2020-03-24 | 2022-10-06 | Rakuten Group, Inc. | Liveness detection using audio-visual inconsistencies |
CN111667835A (zh) * | 2020-06-01 | 2020-09-15 | 马上消费金融股份有限公司 | 语音识别方法、活体检测方法、模型训练方法及装置 |
CN111881726B (zh) * | 2020-06-15 | 2022-11-25 | 马上消费金融股份有限公司 | 一种活体检测方法、装置及存储介质 |
CN112102546A (zh) * | 2020-08-07 | 2020-12-18 | 浙江大华技术股份有限公司 | 一种人机交互控制方法、对讲呼叫方法及相关装置 |
GB202014436D0 (en) * | 2020-09-14 | 2020-10-28 | Voice Biometrics Limted | Multifactor voice and face authentication systems and methods |
CN112133311B (zh) * | 2020-09-18 | 2023-01-17 | 科大讯飞股份有限公司 | 说话人识别方法、相关设备及可读存储介质 |
CN112435653A (zh) * | 2020-10-14 | 2021-03-02 | 北京地平线机器人技术研发有限公司 | 语音识别方法、装置和电子设备 |
CN112348527A (zh) * | 2020-11-17 | 2021-02-09 | 上海桂垚信息科技有限公司 | 一种基于语音识别在银行交易系统中的身份认证方法 |
CN112491844A (zh) * | 2020-11-18 | 2021-03-12 | 西北大学 | 一种基于可信执行环境的声纹及面部识别验证系统及方法 |
CN112397072B (zh) * | 2021-01-18 | 2021-04-30 | 深圳市声扬科技有限公司 | 语音检测方法、装置、电子设备及存储介质 |
KR20220138669A (ko) * | 2021-04-06 | 2022-10-13 | 삼성전자주식회사 | 개인화 오디오 정보를 제공하기 위한 전자 장치 및 방법 |
CN113347608B (zh) * | 2021-06-11 | 2023-05-12 | 焦作大学 | 一种用于车辆的物联网可信认证方法 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060206724A1 (en) * | 2005-02-16 | 2006-09-14 | David Schaufele | Biometric-based systems and methods for identity verification |
CN103841108A (zh) * | 2014-03-12 | 2014-06-04 | 北京天诚盛业科技有限公司 | 用户生物特征的认证方法和系统 |
CN104361276A (zh) * | 2014-11-18 | 2015-02-18 | 新开普电子股份有限公司 | 一种多模态生物特征身份认证方法及系统 |
CN104598796A (zh) * | 2015-01-30 | 2015-05-06 | 科大讯飞股份有限公司 | 身份识别方法及系统 |
CN104834900A (zh) * | 2015-04-15 | 2015-08-12 | 常州飞寻视讯信息科技有限公司 | 一种联合声像信号进行活体检测的方法和系统 |
CN105141619A (zh) * | 2015-09-15 | 2015-12-09 | 北京云知声信息技术有限公司 | 一种帐号登录方法及装置 |
CN105426723A (zh) * | 2015-11-20 | 2016-03-23 | 北京得意音通技术有限责任公司 | 基于声纹识别、人脸识别以及同步活体检测的身份认证方法及系统 |
Family Cites Families (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI326427B (en) | 2005-06-22 | 2010-06-21 | Egis Technology Inc | Biometrics signal input device, computer system having the biometrics signal input device, and control method thereof |
JP2007156974A (ja) * | 2005-12-07 | 2007-06-21 | Kddi Corp | 個人認証・識別システム |
KR101092820B1 (ko) * | 2009-09-22 | 2011-12-12 | 현대자동차주식회사 | 립리딩과 음성 인식 통합 멀티모달 인터페이스 시스템 |
JP2011203992A (ja) * | 2010-03-25 | 2011-10-13 | Sony Corp | 情報処理装置、情報処理方法、およびプログラム |
JP2011215942A (ja) * | 2010-03-31 | 2011-10-27 | Nec Personal Products Co Ltd | ユーザ認証装置、ユーザ認証システム、ユーザ認証方法及びプログラム |
US9100825B2 (en) * | 2012-02-28 | 2015-08-04 | Verizon Patent And Licensing Inc. | Method and system for multi-factor biometric authentication based on different device capture modalities |
JP5492274B2 (ja) * | 2012-10-25 | 2014-05-14 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | 認証装置、認証方法及び認証プログラム |
US20140143551A1 (en) * | 2012-11-21 | 2014-05-22 | Leigh M. Rothschild | Encoding biometric identification information into digital files |
EP2974124A4 (en) * | 2013-03-14 | 2016-10-19 | Intel Corp | VOICE AND / OR FACE RECOGNITION BASED SERVICE DELIVERY |
US9003196B2 (en) * | 2013-05-13 | 2015-04-07 | Hoyos Labs Corp. | System and method for authorizing access to access-controlled environments |
RU2543958C2 (ru) * | 2013-05-14 | 2015-03-10 | Российская Федерация, от имени которой выступает Федеральная служба по техническому и экспортному контролю (ФСТЭК России) | Способ контроля исполнения домашнего ареста с биометрической аутентификацией контролируемого |
US9406295B2 (en) * | 2013-11-22 | 2016-08-02 | Intel Corporation | Apparatus and method for voice based user enrollment with video assistance |
US9721079B2 (en) * | 2014-01-15 | 2017-08-01 | Steve Y Chen | Image authenticity verification using speech |
US9615224B2 (en) * | 2015-02-19 | 2017-04-04 | Cisco Technology, Inc. | Zero touch deployment over a wireless wide area network |
WO2016139655A1 (en) * | 2015-03-01 | 2016-09-09 | I Am Real Ltd. | Method and system for preventing uploading of faked photos |
CN107404381A (zh) * | 2016-05-19 | 2017-11-28 | 阿里巴巴集团控股有限公司 | 一种身份认证方法和装置 |
US9686238B1 (en) * | 2016-07-07 | 2017-06-20 | Oceus Networks Inc. | Secure network enrollment |
US11868995B2 (en) * | 2017-11-27 | 2024-01-09 | Nok Nok Labs, Inc. | Extending a secure key storage for transaction confirmation and cryptocurrency |
-
2016
- 2016-05-19 CN CN201610340549.4A patent/CN107404381A/zh active Pending
-
2017
- 2017-03-14 TW TW106108380A patent/TWI706268B/zh active
- 2017-04-12 RU RU2018144787A patent/RU2738325C2/ru active
- 2017-04-12 JP JP2018560844A patent/JP2019522840A/ja active Pending
- 2017-04-12 KR KR1020187036914A patent/KR102196686B1/ko active IP Right Grant
- 2017-04-12 CA CA3024565A patent/CA3024565C/en active Active
- 2017-04-12 WO PCT/CN2017/080196 patent/WO2017198014A1/zh unknown
- 2017-04-12 SG SG11201810131VA patent/SG11201810131VA/en unknown
- 2017-04-12 MX MX2018014147A patent/MX2018014147A/es unknown
- 2017-04-12 BR BR112018073635-0A patent/BR112018073635A2/pt active Search and Examination
- 2017-04-12 AU AU2017266971A patent/AU2017266971B2/en active Active
- 2017-04-12 MY MYPI2018001981A patent/MY192351A/en unknown
- 2017-04-12 EP EP17798578.5A patent/EP3460697B1/en active Active
-
2018
- 2018-11-15 US US16/192,401 patent/US10789343B2/en active Active
- 2018-11-19 PH PH12018502437A patent/PH12018502437A1/en unknown
- 2018-11-21 ZA ZA2018/07860A patent/ZA201807860B/en unknown
-
2021
- 2021-07-28 JP JP2021123330A patent/JP7109634B2/ja active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060206724A1 (en) * | 2005-02-16 | 2006-09-14 | David Schaufele | Biometric-based systems and methods for identity verification |
CN103841108A (zh) * | 2014-03-12 | 2014-06-04 | 北京天诚盛业科技有限公司 | 用户生物特征的认证方法和系统 |
CN104361276A (zh) * | 2014-11-18 | 2015-02-18 | 新开普电子股份有限公司 | 一种多模态生物特征身份认证方法及系统 |
CN104598796A (zh) * | 2015-01-30 | 2015-05-06 | 科大讯飞股份有限公司 | 身份识别方法及系统 |
CN104834900A (zh) * | 2015-04-15 | 2015-08-12 | 常州飞寻视讯信息科技有限公司 | 一种联合声像信号进行活体检测的方法和系统 |
CN105141619A (zh) * | 2015-09-15 | 2015-12-09 | 北京云知声信息技术有限公司 | 一种帐号登录方法及装置 |
CN105426723A (zh) * | 2015-11-20 | 2016-03-23 | 北京得意音通技术有限责任公司 | 基于声纹识别、人脸识别以及同步活体检测的身份认证方法及系统 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3460697A4 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109271915A (zh) * | 2018-09-07 | 2019-01-25 | 北京市商汤科技开发有限公司 | 防伪检测方法和装置、电子设备、存储介质 |
JP2020535538A (ja) * | 2018-09-07 | 2020-12-03 | 北京市商▲湯▼科技▲開▼▲発▼有限公司Beijing Sensetime Technology Development Co., Ltd. | 偽装防止の検出方法および装置、電子機器、記憶媒体 |
CN109271915B (zh) * | 2018-09-07 | 2021-10-08 | 北京市商汤科技开发有限公司 | 防伪检测方法和装置、电子设备、存储介质 |
CN110364163A (zh) * | 2019-07-05 | 2019-10-22 | 西安交通大学 | 一种语音和唇语相融合的身份认证方法 |
CN111160928A (zh) * | 2019-12-16 | 2020-05-15 | 深圳前海微众银行股份有限公司 | 一种验证身份的方法及装置 |
CN111178277A (zh) * | 2019-12-31 | 2020-05-19 | 支付宝实验室(新加坡)有限公司 | 一种视频流识别方法及装置 |
CN111178287A (zh) * | 2019-12-31 | 2020-05-19 | 云知声智能科技股份有限公司 | 一种声像融合的端对端身份识别方法及装置 |
CN111178277B (zh) * | 2019-12-31 | 2023-07-14 | 支付宝实验室(新加坡)有限公司 | 一种视频流识别方法及装置 |
CN111814732A (zh) * | 2020-07-23 | 2020-10-23 | 上海优扬新媒信息技术有限公司 | 一种身份验证方法及装置 |
CN111814732B (zh) * | 2020-07-23 | 2024-02-09 | 度小满科技(北京)有限公司 | 一种身份验证方法及装置 |
CN114677634A (zh) * | 2022-05-30 | 2022-06-28 | 成都新希望金融信息有限公司 | 面签识别方法、装置、电子设备及存储介质 |
CN114677634B (zh) * | 2022-05-30 | 2022-09-27 | 成都新希望金融信息有限公司 | 面签识别方法、装置、电子设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
AU2017266971A1 (en) | 2018-12-06 |
SG11201810131VA (en) | 2018-12-28 |
MX2018014147A (es) | 2019-08-12 |
KR102196686B1 (ko) | 2020-12-31 |
BR112018073635A2 (pt) | 2019-02-26 |
AU2017266971B2 (en) | 2021-05-13 |
US10789343B2 (en) | 2020-09-29 |
PH12018502437A1 (en) | 2019-05-15 |
CA3024565C (en) | 2022-09-06 |
EP3460697A4 (en) | 2019-05-08 |
CN107404381A (zh) | 2017-11-28 |
MY192351A (en) | 2022-08-17 |
JP2021182420A (ja) | 2021-11-25 |
CA3024565A1 (en) | 2017-11-23 |
JP7109634B2 (ja) | 2022-07-29 |
JP2019522840A (ja) | 2019-08-15 |
TW201741921A (zh) | 2017-12-01 |
RU2738325C2 (ru) | 2020-12-11 |
KR20190009361A (ko) | 2019-01-28 |
RU2018144787A (ru) | 2020-06-19 |
US20190102531A1 (en) | 2019-04-04 |
EP3460697A1 (en) | 2019-03-27 |
RU2018144787A3 (zh) | 2020-06-19 |
EP3460697B1 (en) | 2021-12-08 |
ZA201807860B (en) | 2019-08-28 |
TWI706268B (zh) | 2020-10-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017198014A1 (zh) | 一种身份认证方法和装置 | |
KR102455633B1 (ko) | 라이브니스 검사 방법 및 장치 | |
WO2017215558A1 (zh) | 一种声纹识别方法和装置 | |
EP3540621B1 (en) | Identity authentication method and apparatus, terminal and server | |
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 | |
TW201907330A (zh) | 身份認證的方法、裝置、設備及資料處理方法 | |
CN107346568B (zh) | 一种门禁系统的认证方法和装置 | |
CN111611568A (zh) | 一种人脸声纹复核终端及其身份认证方法 | |
Hassanat | Visual passwords using automatic lip reading | |
JP6280068B2 (ja) | パラメータ学習装置、話者認識装置、パラメータ学習方法、話者認識方法、およびプログラム | |
CN111611437A (zh) | 一种防止人脸声纹验证替换攻击的方法及装置 | |
Shen et al. | Secure mobile services by face and speech based personal authentication | |
Bredin et al. | Making talking-face authentication robust to deliberate imposture | |
CN111611569A (zh) | 一种人脸声纹复核终端及其身份认证方法 | |
Shenai et al. | Fast biometric authentication system based on audio-visual fusion | |
Mohammed et al. | Evaluation of Voice & Ear Biometrics Authentication System | |
Gofman et al. | Quality-based score-level fusion for secure and robust multimodal biometrics-based authentication on consumer mobile devices | |
KR100360480B1 (ko) | 사용자 인증 장치 및 그 방법 | |
WO2021060256A1 (ja) | 顔認証装置、顔認証方法、及びコンピュータ読み取り可能な記録媒体 | |
Ramya et al. | Enhanced Speaker Verification Incorporated with Face Recognition | |
WO2023238953A1 (ja) | 認証装置および認証方法 | |
JP2021131619A (ja) | 情報処理方法、情報処理装置及びプログラム | |
Das et al. | Multi-feature audio-visual person recognition | |
Poh et al. | Multimodal Interaction in Biometrics: Technological and Usability Challenges | |
BANU et al. | IMPLEMENTATION OF ACCURATE PERSONAL IDENTIFICATION BY USING PALM PRINT IMAGE PROCESSING |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 3024565 Country of ref document: CA |
|
ENP | Entry into the national phase |
Ref document number: 2018560844 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
REG | Reference to national code |
Ref country code: BR Ref legal event code: B01A Ref document number: 112018073635 Country of ref document: BR |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17798578 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2017266971 Country of ref document: AU Date of ref document: 20170412 Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 20187036914 Country of ref document: KR Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2017798578 Country of ref document: EP Effective date: 20181219 |
|
ENP | Entry into the national phase |
Ref document number: 112018073635 Country of ref document: BR Kind code of ref document: A2 Effective date: 20181116 |