CN109635625B - Intelligent identity verification method, equipment, storage medium and device - Google Patents
Intelligent identity verification method, equipment, storage medium and device Download PDFInfo
- Publication number
- CN109635625B CN109635625B CN201811206298.6A CN201811206298A CN109635625B CN 109635625 B CN109635625 B CN 109635625B CN 201811206298 A CN201811206298 A CN 201811206298A CN 109635625 B CN109635625 B CN 109635625B
- Authority
- CN
- China
- Prior art keywords
- target
- score
- user
- passing
- identity verification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012795 verification Methods 0.000 title claims abstract description 192
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000013507 mapping Methods 0.000 claims description 14
- 238000012937 correction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000012015 optical character recognition Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 210000000056 organ Anatomy 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- 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
- G06V40/168—Feature extraction; Face representation
-
- 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
- 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
- G10L17/08—Use of distortion metrics or a particular distance between probe pattern and reference templates
-
- 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
- G10L17/12—Score normalisation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses an intelligent identity verification method, equipment, a storage medium and a device, wherein the method comprises the following steps: collecting face images of users, extracting current face feature information from the face images, comparing the similarity between the target face feature information and the current face feature information, scoring the similarity to obtain a first similarity score, comparing the first similarity score with a target passing score threshold, and judging that the user identity verification passes when the first similarity score is higher than the target passing score threshold. According to the invention, the first similarity score between the current face feature information and the target face feature information is obtained, so that whether the identity verification passes or not is judged according to whether the first similarity score exceeds the target passing score threshold, and as different passing score thresholds are set according to different wind control requirements of different business scenes, the flexibility and the efficiency of the identity verification are obviously improved, and the transaction safety of a user is ensured.
Description
Technical Field
The present invention relates to the field of identity verification technologies, and in particular, to an intelligent identity verification method, device, storage medium, and apparatus.
Background
The identity verification scheme is based on a computer vision technology, obtains user identity information through an identity card optical character recognition (Optical Character Recognition, OCR) technology or manual input of a user, verifies the user identity information through an identity verification platform appointed by public security, obtains a reserved photo of the user in the public security, and confirms the user identity through a comparison technology.
At present, identity verification is widely used in various business scenes such as user login, application and security, however, the conventional identity verification scheme adopts a unified threshold value for comparison to confirm the identity of the user no matter what business scene the user is in, and each business scene has different wind control requirements, and the identity verification is performed by adopting the unified threshold value, so that the accuracy is lower.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide an intelligent identity verification method, equipment, a storage medium and a device, and aims to solve the technical problems of low accuracy in the prior art that a unified threshold value is adopted for identity verification.
In order to achieve the above object, the present invention provides an intelligent identity verification method, which includes the steps of:
Acquiring a target passing score threshold corresponding to a current service scene;
acquiring user identity information of a user according to a received user identity verification instruction, and acquiring target face characteristic information according to the user identity information;
collecting a face image of the user, and extracting current face characteristic information from the face image;
comparing the similarity of the target face characteristic information with the current face characteristic information, scoring the similarity, and obtaining a first similarity score;
and comparing the first similarity score with the target passing score threshold, and judging that the user identity verification passes when the first similarity score is higher than the target passing score threshold.
Preferably, before the user identity information of the user is obtained according to the received user identity verification instruction, the method further includes:
acquiring verification precision and a target threshold interval in a service scene;
selecting a passing score threshold value meeting the verification precision in the target threshold value interval, and establishing a preset mapping relation between a service scene and the passing score threshold value;
and acquiring a current service scene, and searching a target passing score threshold corresponding to the current service scene in the preset mapping relation.
Preferably, the verification accuracy includes a target passing rate and a target misjudgment rate;
the selecting a pass score threshold value meeting the verification precision in the target threshold value interval comprises the following steps:
acquiring sample data in a service scene, wherein the sample data comprises a sample face image, sample positive and negative and sample similarity scores;
sequentially selecting a to-be-determined passing score threshold value in the target threshold value interval, and calculating the current passing rate and the current misjudgment rate of the sample face image according to the to-be-determined passing score threshold value, the sample positive and negative property and the sample similarity score;
when the current passing rate is not higher than the target passing rate and/or the current misjudgment rate is not lower than the target misjudgment rate, selecting a next undetermined passing score threshold value, and returning to the step of calculating the current passing rate and the current misjudgment rate of the sample face image according to the undetermined passing score threshold value, the sample positive and negative property and the sample similarity score;
and when the current passing rate is higher than the target passing rate and the current misjudgment rate is lower than the target misjudgment rate, taking the undetermined passing score threshold as the passing score threshold meeting the verification precision.
Preferably, the selecting the predetermined passing score threshold sequentially in the target threshold interval, calculating the current passing rate and the current misjudgment rate of the sample face image according to the predetermined passing score threshold, the sample positive and negative and the sample similarity score, includes:
sequentially selecting undetermined passing score thresholds in the target threshold interval;
judging a verification result of a sample according to the sample similarity score and the to-be-determined passing score threshold value, and calculating the current passing rate of the face image of the sample according to the verification result;
and calculating the current misjudgment rate of the sample face image according to the sample positive and negative and the verification result.
Preferably, after comparing the first similarity score with the target pass score threshold, the method further comprises:
when the first similarity score is not higher than the target passing score threshold, judging whether the first similarity score is lower than a first preset standard score, wherein the first preset standard score is lower than the target passing score threshold;
when the first similarity score is lower than a first preset standard score, judging that the user identity verification is not passed;
And when the first similarity score is not lower than a first preset standard score, performing secondary identity verification on the user.
Preferably, when the first similarity score is not lower than a first preset standard score, performing a secondary identity verification on the user, including:
when the first similarity score is not lower than a first preset standard score, acquiring the audio of the user, and extracting current voiceprint feature information from the audio;
acquiring target voiceprint feature information according to the user identity information;
comparing the similarity of the target voiceprint characteristic information with the current voiceprint characteristic information, scoring the similarity, and obtaining a second similarity score;
and comparing the second similarity score with the second preset standard score, and judging that the user identity verification passes when the second similarity score is higher than the second preset standard score.
Preferably, before the step of collecting the face image of the user and extracting the current face feature information from the face image, the method includes:
detecting whether the user is a living body;
when the user is a living body, executing the steps of collecting the face image of the user and extracting the current face characteristic information from the face image;
And when the user is non-living, judging that the user identity verification is not passed.
In addition, in order to achieve the above object, the present invention also proposes a smart identity verification device comprising a memory, a processor and a smart identity verification program stored on the memory and executable on the processor, the smart identity verification program being configured to implement the steps of the smart identity verification method as described above.
In addition, in order to achieve the above object, the present invention also proposes a storage medium having stored thereon a smart identity verification program which, when executed by a processor, implements the steps of the smart identity verification method as described above.
In addition, in order to achieve the above object, the present invention also provides an intelligent identity verification apparatus, which includes:
the threshold value acquisition module is used for acquiring a target passing score threshold value corresponding to the current service scene;
the first feature acquisition module is used for acquiring user identity information of the user according to the received user identity verification instruction and acquiring target face feature information according to the user identity information;
The second feature acquisition module is used for acquiring a face image of the user and extracting current face feature information from the face image;
the scoring module is used for comparing the similarity between the target face characteristic information and the current face characteristic information, scoring the similarity and obtaining a first similarity score;
and the verification module is used for comparing the first similarity score with the target passing score threshold value, and judging that the user identity verification passes when the first similarity score is higher than the target passing score threshold value.
According to the invention, the first similarity score between the current face feature information and the target face feature information is obtained, and the target passing score threshold corresponding to the current service scene is obtained, so that whether the user identity verification passes or not is judged according to whether the first similarity score exceeds the target passing score threshold, and as different passing score thresholds are set according to different wind control demands of different service scenes, the flexibility and the efficiency of the identity verification are obviously improved, the user experience is improved, and the transaction safety of the user is ensured.
Drawings
FIG. 1 is a schematic diagram of an intelligent identity verification device in a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flow chart of a first embodiment of the intelligent identity verification method of the present invention;
FIG. 3 is a flow chart of a second embodiment of the intelligent identity verification method of the present invention;
FIG. 4 is a flow chart of a third embodiment of the intelligent identity verification method of the present invention;
FIG. 5 is a block diagram of a first embodiment of the intelligent identity verification apparatus of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an intelligent identity verification device of a hardware running environment according to an embodiment of the present invention.
As shown in fig. 1, the smart identity verification apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display (Display), and the optional user interface 1003 may also include a standard wired interface, a wireless interface, and the wired interface for the user interface 1003 may be a USB interface in the present invention. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) Memory or a stable Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the structure shown in FIG. 1 is not limiting of the intelligent identity verification device and may include more or fewer components than shown, or may combine certain components, or may be arranged in different components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a smart identity verification program may be included in a memory 1005, which is a type of computer storage medium.
In the intelligent identity verification apparatus shown in fig. 1, the network interface 1004 is mainly used for connecting to a background server, and performing data communication with the background server; the user interface 1003 is mainly used for connecting user equipment; the intelligent identity verification apparatus calls an intelligent identity verification program stored in the memory 1005 through the processor 1001 and executes the intelligent identity verification method provided by the embodiment of the present invention.
Based on the hardware structure, the embodiment of the intelligent identity verification method is provided.
Referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the intelligent identity verification method according to the present invention, and the first embodiment of the intelligent identity verification method according to the present invention is provided.
In a first embodiment, the smart identity verification method comprises the steps of:
Step S10: and obtaining a target passing score threshold corresponding to the current service scene.
It should be noted that, the execution subject of the embodiment is an intelligent identity verification device, where the intelligent identity verification device may be an electronic device such as a personal computer, a smart phone, a tablet computer, and the like. Under various existing identity verification scenes, comparing the current user characteristics with the target user characteristics to obtain similarity scores, setting a pass score threshold, judging that the identity verification of the current user passes when the similarity scores are higher than the pass score threshold, and judging that the identity verification of the current user does not pass when the similarity scores are not higher than the pass score threshold. However, each business scenario has different wind control requirements, for example, a user logged in business scenario has a lower wind control requirement, a lower pass score threshold may be set, a user-secured business scenario has a higher wind control requirement, and a higher pass score threshold may be set.
Therefore, in this embodiment, various service scenarios are predetermined, and a passing score threshold meeting the wind control requirement of the service scenario is set for each service scenario, so that a mapping relationship between the service scenario and the passing score threshold is established and stored. When a user enters an identity verification interface through intelligent identity verification equipment, extracting a current service scene from the identity verification interface, and searching a target passing score threshold corresponding to the current service scene in the mapping relation.
Step S20: and acquiring user identity information of the user according to the received user identity verification instruction, and acquiring target face characteristic information according to the user identity information.
The identity verification instruction is an instruction input by a user and used for initiating identity verification, wherein the instruction comprises user identity information, and the user identity information comprises at least one of user name, user account number or user identity card number. The face feature information is feature data which is used for assisting in face classification according to the shape description of face organs and the distance characteristics between the face organs, and the feature components generally comprise Euclidean distance, curvature, angle and the like among feature points. The face is composed of eyes, nose, mouth, chin, etc., and the geometric description of these parts and the structural relationship between them can be used as the important feature for identifying the face. And extracting the user identity information from the identity verification instruction, and acquiring target face characteristic information according to the user identity information, wherein the target face characteristic information is face characteristic information corresponding to the user identity information, namely real face characteristic information.
It will be appreciated that the first correspondence between the user identity information and the face feature information may be stored locally at the intelligent identity verification device or at a background server. When the first corresponding relation is stored locally in the intelligent identity verification equipment, the searching speed can be increased, and the user experience is improved; when the first corresponding relation is stored in the background server, the local storage space can be saved, and the local processing pressure can be reduced.
In a specific implementation, when a user inputs user identity information such as a user name, a user account number or a user identity card number in an identity verification interface according to a prompt, and clicks an identity verification instruction, the user identity information and the identity verification instruction are simultaneously sent to an intelligent identity verification device local or background server, and the intelligent identity verification device local searches corresponding target face feature information according to the user identity information or receives target face feature information fed back by the background server.
Step S30: and acquiring a face image of the user, and extracting current face characteristic information from the face image.
After obtaining the target face feature information, the user will be subjected to face recognition, the face image of the user is collected, and the feature extraction is performed on the face image, so as to obtain the current face feature information.
In a specific implementation, before extracting the current face feature information from the face image, the collected face image is limited by various conditions and randomly disturbed, so that the face image cannot be directly used, and the image preprocessing such as gray correction, noise filtering and the like must be performed at the early stage of image processing. For the face image, the preprocessing process mainly comprises light compensation, gray level transformation, histogram equalization, normalization, geometric correction, filtering, sharpening and the like of the face image.
Step S40: and comparing the similarity of the target face characteristic information with the similarity of the current face characteristic information, scoring the similarity, and obtaining a first similarity score.
It should be noted that, because the target face feature information is face feature information corresponding to user identity information, if the similarity between the current face feature information and the target face feature information is higher, the user and the user corresponding to the user identity information can be considered to be the same person, and the user identity verification can be determined to pass. Therefore, the target face feature information and the current face feature information are subjected to similarity comparison, similarity scoring is carried out, and a first similarity score can be obtained.
Step S50: and comparing the first similarity score with the target passing score threshold, and judging that the user identity verification passes when the first similarity score is higher than the target passing score threshold.
It can be understood that, for how similar the identity verification can be passed, the passing score threshold value of the target corresponding to the current service scenario is determined. And comparing the first similarity score with a target passing score threshold corresponding to the current business scene, and judging that the user identity verification passes if the first similarity score is higher than the target passing score threshold.
In the first embodiment, by acquiring the first similarity score between the current face feature information and the target face feature information and acquiring the target passing score threshold corresponding to the current service scene, whether the user identity verification passes or not is judged according to whether the first similarity score exceeds the target passing score threshold, and as different passing score thresholds are set according to different wind control requirements of different service scenes, the flexibility and the efficiency of the identity verification are obviously improved, the user experience is improved, and the transaction safety of the user is ensured.
Referring to fig. 3, fig. 3 is a schematic flow chart of a second embodiment of the intelligent identity verification method according to the present invention, and the second embodiment of the intelligent identity verification method according to the present invention is proposed based on the first embodiment shown in fig. 2.
In a second embodiment, the step S10 includes:
step S101: and acquiring verification precision and a target threshold interval in the service scene.
Step S102: and selecting a passing score threshold value meeting the verification precision in the target threshold value interval, and establishing a preset mapping relation between a service scene and the passing score threshold value.
It can be understood that the verification accuracy refers to the passing rate and the misjudgment rate of verification, and the requirements of each service scene on the verification accuracy are different, for example, in the service scene where the user logs in, the passing rate of the verification of the user is required to be higher, and in the service scene where the user is ensured, the misjudgment rate of the verification of the user is required to be lower, so that different verification accuracies are set for different service scenes. The target threshold interval refers to a rough interval containing a target passing score threshold, for example, the target threshold interval required by a business scene logged by a user is about 40-60 minutes, the target threshold interval required by a business scene protected by the user is about 65-90 minutes, and the final passing score threshold is determined from the target threshold interval.
Step S103: and acquiring a current service scene, and searching a target passing score threshold corresponding to the current service scene in the preset mapping relation.
It should be noted that, after a preset mapping relationship between a service scene and a passing score threshold is established, a corresponding target passing score threshold may be searched in the preset mapping relationship according to a current service scene.
Further, the verification accuracy comprises a target passing rate and a target misjudgment rate;
the step S102 includes:
sample data in a business scene is obtained, wherein the sample data comprises a sample face image, sample positive and negative and sample similarity scores.
And sequentially selecting a pending passing score threshold value in the target threshold value interval, and calculating the current passing rate and the current misjudgment rate of the sample face image according to the pending passing score threshold value, the sample positive and negative and the sample similarity score.
And when the current passing rate is not higher than the target passing rate and/or the current misjudgment rate is not lower than the target misjudgment rate, selecting a next undetermined passing score threshold value, and returning to the step of calculating the current passing rate and the current misjudgment rate of the sample face image according to the undetermined passing score threshold value, the sample positive and negative and the sample similarity score.
And when the current passing rate is higher than the target passing rate and the current misjudgment rate is lower than the target misjudgment rate, taking the undetermined passing score threshold as the passing score threshold meeting the verification precision.
When the current passing rate is higher than the target passing rate and the current misjudgment rate is lower than the target misjudgment rate, the fact that the passing score threshold to be determined meets the verification precision is indicated, and the passing score threshold to be determined is used as the passing score threshold meeting the verification precision in a business scene.
In specific implementation, taking a user-secured service scene as an example for explanation, because the user-secured service scene has higher requirements on verification accuracy, preliminarily setting the verification accuracy to be not lower than 70% of target passing rate, not higher than 10% of target erroneous judgment rate, and 65-90 minutes of target threshold interval, sequentially taking each score as a pending threshold score in the interval, and respectively calculating the current passing rate and the current erroneous judgment rate under different pending threshold scores; and taking the undetermined threshold value of which the current passing rate and the current misjudgment rate meet the verification precision as a target passing score threshold value. For example, when 60 points are used as the undetermined threshold value points, the passing rate is 80%, and the misjudgment rate is 20%; when 72 points are used as undetermined threshold values, the passing rate is 75%, and the misjudgment rate is 8%; when 85 points are used as undetermined threshold values, the passing rate is 65%, and the misjudgment rate is 8%. Accordingly, the pending passing score threshold satisfying the verification accuracy is 72 points, so 72 points are taken as the target passing score threshold.
Further, the sequentially selecting the pending passing score threshold in the target threshold interval, and calculating the current passing rate and the current misjudgment rate of the sample face image according to the pending passing score threshold, the sample positive and negative and the sample similarity score, including:
sequentially selecting undetermined passing score thresholds in the target threshold interval;
judging a verification result of a sample according to the sample similarity score and the to-be-determined passing score threshold value, and calculating the current passing rate of the face image of the sample according to the verification result;
and calculating the current misjudgment rate of the sample face image according to the sample positive and negative and the verification result.
In a specific implementation, a plurality of positive samples, a plurality of negative samples and corresponding sample similarity scores under each service scene are obtained, for each service scene, a to-be-determined passing score threshold is selected in a target threshold interval of the service scene, when the sample similarity score is lower than the to-be-determined passing score threshold, a verification result of a sample corresponding to the sample similarity score is considered to be not passed, otherwise, the verification result is considered to be passed, and the current passing rate is calculated according to the number of samples, of which the verification result is passed, divided by the total number of samples; meanwhile, the current misjudgment rate is calculated according to the positive and negative of the sample and the verification result, when the sample is a positive sample and the verification result is a pass, the judgment is correct, when the sample is a positive sample and the verification result is a fail, the judgment is wrong, when the sample is a negative sample and the verification result is a pass, the judgment is wrong, when the sample is a negative sample and the verification result is a fail, the judgment is correct, and therefore the current misjudgment rate is calculated according to the correct sample number and the total sample number.
In a second embodiment, verification accuracy and a target threshold interval in a service scene are acquired; and selecting a pass score threshold meeting the verification precision in the target threshold interval. Based on a large amount of sample data of laboratories, a pass score threshold value which can best meet the verification precision of each business scene is obtained.
Referring to fig. 4, fig. 4 is a schematic flow chart of a third embodiment of the intelligent identity verification method according to the present invention, and the third embodiment of the intelligent identity verification method according to the present invention is proposed based on the first embodiment shown in fig. 2.
In a third embodiment, after the step S50, the method further includes:
step S501: and when the first similarity score is not higher than the target passing score threshold, judging whether the first similarity score is lower than a first preset standard score, wherein the first preset standard score is lower than the target passing score threshold.
It is to be appreciated that when the first similarity score is above the target pass score threshold, determining that the user identity verification passed; when the first similarity score is not higher than the target passing score threshold, if the user identity verification is directly determined to be failed, part of real users may be misjudged as fake users, so that secondary identity verification is set to improve accuracy of identity verification, and specifically, whether secondary identity verification is performed or not is determined through a first preset standard score.
Step S502: and when the first similarity score is lower than a first preset standard score, judging that the user identity verification is not passed.
It should be noted that, if the first preset standard score is lower than the target passing score threshold, if the first similarity score is still lower than the first preset standard score, it is indicated that the possibility that the user is a real user is very low, and at this time, it is determined that the user identity verification is not passed.
Step S503: and when the first similarity score is not lower than a first preset standard score, performing secondary identity verification on the user.
It will be appreciated that when the first similarity score is not less than a first predetermined standard score, the user has a certain likelihood of being a real user, and a secondary identity verification will be performed on the user.
Further, the step S503 specifically includes:
when the first similarity score is not lower than a first preset standard score, acquiring target voiceprint feature information according to the user identity information;
collecting the audio of the user, and extracting current voiceprint feature information from the audio;
comparing the similarity of the target voiceprint characteristic information with the current voiceprint characteristic information, scoring the similarity, and obtaining a second similarity score;
And comparing the second similarity score with the second preset standard score, and judging that the user identity verification passes when the second similarity score is higher than the second preset standard score.
When the first similarity score is not lower than a first preset standard score, performing secondary identity verification on the user through voiceprint recognition.
In a specific implementation, target voiceprint feature information is obtained according to the user identity information, wherein the target voiceprint feature information is voiceprint feature information corresponding to the user identity information, namely real voiceprint feature information. The second corresponding relation between the user identity information and the voiceprint feature information can be stored locally in the intelligent identity verification device or stored in a background server; when the second corresponding relation is stored locally in the intelligent identity verification equipment, the searching speed can be increased, and the user experience is improved; when the second corresponding relation is stored in the background server, the local storage space can be saved, and the local processing pressure can be reduced. And the intelligent identity verification equipment locally searches corresponding target voiceprint feature information according to the user identity information or receives the target voiceprint feature information fed back by the background server.
It can be understood that after the target voiceprint feature information is obtained, voiceprint recognition is performed on the user, the audio of the user is collected, and feature extraction is performed on the audio to obtain the current voiceprint feature information. Because the target voiceprint feature information is voiceprint feature information corresponding to user identity information, if the similarity between the current voiceprint feature information and the target voiceprint feature information is higher, the user and the user corresponding to the user identity information can be considered to be the same person, and the user identity verification can be judged to pass. And comparing the similarity between the target voiceprint characteristic information and the current voiceprint characteristic information, scoring the similarity, and obtaining a second similarity score. And the second preset standard score is a pass score threshold value for the pass of voiceprint recognition, the second similarity score is compared with the second preset standard score, and if the second similarity score is higher than the second preset standard score, the pass of the user identity verification is judged.
In a third embodiment, before the step S30, the method further includes:
detecting whether the user is a living body;
when the user is a living body, executing the steps of collecting the face image of the user and extracting the current face characteristic information from the face image;
And when the user is non-living, judging that the user identity verification is not passed.
It should be noted that, in order to avoid an attack of the user not a real person but a photo, a video, a static or a 3D model, whether the user is a living body is detected, if the current user is a non-living body, identity verification is ended, and if the current user is a living body, identity verification is continued.
In a specific implementation, the detecting whether the current user is a living body includes:
prompting the user to perform a specified limb action, such as right hand left shoulder, detecting whether the action of the user is correct, and recognizing the user as a living body when the action of the user is consistent with the specified limb action; alternatively, it is detected by an application (e.g., weChat) whether the user is living.
In the third embodiment, by detecting whether the user is a living body, attacks of photos, videos, static or 3D models can be effectively avoided; and after face recognition, the secondary identity verification is performed through voiceprint recognition to improve the accuracy of the identity verification, so that part of real users are prevented from being misjudged as false users, and the user experience is improved.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium is stored with an intelligent identity verification program, and the intelligent identity verification program realizes the steps of the intelligent identity verification method when being executed by a processor.
In addition, referring to fig. 5, an embodiment of the present invention further provides an intelligent identity verification apparatus, where the intelligent identity verification apparatus includes:
the threshold value acquisition module 10 is used for acquiring a target passing score threshold value corresponding to the current service scene;
the first feature acquisition module 20 is configured to acquire user identity information of a user according to a received identity verification instruction of the user, and acquire target face feature information according to the user identity information;
a second feature acquisition module 30, configured to acquire a face image of the user, and extract current face feature information from the face image;
the scoring module 40 is configured to compare the similarity between the target face feature information and the current face feature information, and score the similarity to obtain a first similarity score;
the verification module 50 is configured to compare the first similarity score with the target passing score threshold, and determine that the user identity verification passes when the first similarity score is higher than the target passing score threshold.
In an embodiment, the smart identity verification apparatus further comprises: the threshold value determining module is used for acquiring verification precision and a target threshold value interval in a service scene;
Selecting a passing score threshold value meeting the verification precision in the target threshold value interval, and establishing a preset mapping relation between a service scene and the passing score threshold value;
and acquiring a current service scene, and searching a target passing score threshold corresponding to the current service scene in the preset mapping relation.
In an embodiment, the threshold determining module is further configured to obtain sample data in a service scene, where the sample data includes a sample face image, a sample positive and negative property, and a sample similarity score;
sequentially selecting a to-be-determined passing score threshold value in the target threshold value interval, and calculating the current passing rate and the current misjudgment rate of the sample face image according to the to-be-determined passing score threshold value, the sample positive and negative property and the sample similarity score;
when the current passing rate is not higher than the target passing rate and/or the current misjudgment rate is not lower than the target misjudgment rate, selecting a next undetermined passing score threshold value, and returning to the step of calculating the current passing rate and the current misjudgment rate of the sample face image according to the undetermined passing score threshold value, the sample positive and negative property and the sample similarity score;
And when the current passing rate is higher than the target passing rate and the current misjudgment rate is lower than the target misjudgment rate, taking the undetermined passing score threshold as the passing score threshold meeting the verification precision.
In an embodiment, the threshold determining module is further configured to sequentially select a pending passing score threshold in the target threshold interval;
judging a verification result of a sample according to the sample similarity score and the to-be-determined passing score threshold value, and calculating the current passing rate of the face image of the sample according to the verification result;
and calculating the current misjudgment rate of the sample face image according to the sample positive and negative and the verification result.
In an embodiment, the smart identity verification apparatus further comprises: the secondary verification module is used for judging whether the first similarity score is lower than a first preset standard score or not when the first similarity score is not higher than the target passing score threshold, and the first preset standard score is lower than the target passing score threshold;
when the first similarity score is lower than a first preset standard score, judging that the user identity verification is not passed;
and when the first similarity score is not lower than a first preset standard score, performing secondary identity verification on the user.
In an embodiment, the secondary verification module is further configured to collect audio of the user when the first similarity score is not lower than a first preset standard score, and extract current voiceprint feature information from the audio;
acquiring target voiceprint feature information according to the user identity information;
comparing the similarity of the target voiceprint characteristic information with the current voiceprint characteristic information, scoring the similarity, and obtaining a second similarity score;
and comparing the second similarity score with the second preset standard score, and judging that the user identity verification passes when the second similarity score is higher than the second preset standard score.
In an embodiment, the smart identity verification apparatus further comprises: a living body detection module for detecting whether the user is a living body;
when the user is a living body, executing the steps of collecting the face image of the user and extracting the current face characteristic information from the face image;
and when the user is non-living, judging that the user identity verification is not passed.
Other embodiments or specific implementation manners of the intelligent identity verification apparatus of the present invention may refer to the above method embodiments, and are not described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the terms first, second, third, etc. do not denote any order, but rather the terms first, second, third, etc. are used to interpret the terms as names.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. read only memory mirror (Read Only Memory image, ROM)/random access memory (Random Access Memory, RAM), magnetic disk, optical disk), comprising instructions for causing a terminal intelligent identity verification device (which may be a mobile phone, a computer, a server, an air conditioner, or a network intelligent identity verification device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (9)
1. An intelligent identity verification method, which is characterized by comprising the following steps:
acquiring a target passing score threshold corresponding to a current service scene;
acquiring user identity information of a user according to a received user identity verification instruction, and acquiring target face characteristic information according to the user identity information;
collecting a face image of the user, extracting current face characteristic information from the face image, wherein the face image is a preprocessed image, and the preprocessing comprises the following steps: light compensation, gray level change, histogram equalization, normalization, geometric correction filtering and sharpening;
comparing the similarity of the target face characteristic information with the current face characteristic information, scoring the similarity, and obtaining a first similarity score;
comparing the first similarity score with the target passing score threshold, and judging that the user identity verification passes when the first similarity score is higher than the target passing score threshold;
The obtaining the target passing score threshold corresponding to the current service scene comprises the following steps:
acquiring verification precision and a target threshold interval in a service scene;
selecting a passing score threshold value meeting the verification precision in the target threshold value interval, and establishing a preset mapping relation between a service scene and the passing score threshold value;
and acquiring a current service scene, and searching a target passing score threshold corresponding to the current service scene in the preset mapping relation.
2. The intelligent identity verification method of claim 1, wherein the verification accuracy includes a target passing rate and a target misjudgment rate;
the selecting a pass score threshold value meeting the verification precision in the target threshold value interval comprises the following steps:
acquiring sample data in a service scene, wherein the sample data comprises a sample face image, sample positive and negative and sample similarity scores;
sequentially selecting a to-be-determined passing score threshold value in the target threshold value interval, and calculating the current passing rate and the current misjudgment rate of the sample face image according to the to-be-determined passing score threshold value, the sample positive and negative property and the sample similarity score;
when the current passing rate is not higher than the target passing rate and/or the current misjudgment rate is not lower than the target misjudgment rate, selecting a next undetermined passing score threshold value, and returning to the step of calculating the current passing rate and the current misjudgment rate of the sample face image according to the undetermined passing score threshold value, the sample positive and negative property and the sample similarity score;
And when the current passing rate is higher than the target passing rate and the current misjudgment rate is lower than the target misjudgment rate, taking the undetermined passing score threshold as the passing score threshold meeting the verification precision.
3. The intelligent identity verification method according to claim 2, wherein sequentially selecting the passing score threshold value in the target threshold value interval, and calculating the current passing rate and the current misjudgment rate of the sample face image according to the passing score threshold value, the sample positive and negative and the sample similarity score comprises:
sequentially selecting undetermined passing score thresholds in the target threshold interval;
judging a verification result of a sample according to the sample similarity score and the to-be-determined passing score threshold value, and calculating the current passing rate of the face image of the sample according to the verification result;
and calculating the current misjudgment rate of the sample face image according to the sample positive and negative and the verification result.
4. A smart identity verification method according to any one of claims 1-3, wherein after said comparing said first similarity score to said target pass score threshold, said method further comprises:
When the first similarity score is not higher than the target passing score threshold, judging whether the first similarity score is lower than a first preset standard score, wherein the first preset standard score is lower than the target passing score threshold;
when the first similarity score is lower than a first preset standard score, judging that the user identity verification is not passed;
and when the first similarity score is not lower than a first preset standard score, performing secondary identity verification on the user.
5. The intelligent identity verification method of claim 4, wherein performing secondary identity verification on the user when the first similarity score is not lower than a first preset standard score comprises:
when the first similarity score is not lower than a first preset standard score, acquiring the audio of the user, and extracting current voiceprint feature information from the audio;
acquiring target voiceprint feature information according to the user identity information;
comparing the similarity of the target voiceprint characteristic information with the current voiceprint characteristic information, scoring the similarity, and obtaining a second similarity score;
and comparing the second similarity score with a second preset standard score, and judging that the user identity verification passes when the second similarity score is higher than the second preset standard score.
6. A smart identity verification method as claimed in any one of claims 1 to 3 wherein said acquiring a face image of said user and extracting current face feature information from said face image is preceded by:
detecting whether the user is a living body;
when the user is a living body, executing the steps of collecting the face image of the user and extracting the current face characteristic information from the face image;
and when the user is non-living, judging that the user identity verification is not passed.
7. An intelligent identity verification apparatus, characterized in that the intelligent identity verification apparatus comprises: a memory, a processor and a smart identity verification program stored on the memory and executable on the processor, which when executed by the processor performs the steps of the smart identity verification method of any one of claims 1 to 6.
8. A storage medium having stored thereon a smart identity verification program which when executed by a processor performs the steps of the smart identity verification method of any one of claims 1 to 6.
9. An intelligent identity verification apparatus, characterized in that the intelligent identity verification apparatus comprises:
the threshold value acquisition module is used for acquiring a target passing score threshold value corresponding to the current service scene;
the first feature acquisition module is used for acquiring user identity information of the user according to the received user identity verification instruction and acquiring target face feature information according to the user identity information;
the second feature acquisition module is used for acquiring a face image of the user and extracting current face feature information from the face image, wherein the face image is a preprocessed image, and the preprocessing comprises: light compensation, gray level change, histogram equalization, normalization, geometric correction filtering and sharpening;
the scoring module is used for comparing the similarity between the target face characteristic information and the current face characteristic information, scoring the similarity and obtaining a first similarity score;
the verification module is used for comparing the first similarity score with the target passing score threshold, and judging that the user identity verification passes when the first similarity score is higher than the target passing score threshold;
The threshold value acquisition module is also used for acquiring verification precision and a target threshold value interval in a service scene; selecting a passing score threshold value meeting the verification precision in the target threshold value interval, and establishing a preset mapping relation between a service scene and the passing score threshold value; and acquiring a current service scene, and searching a target passing score threshold corresponding to the current service scene in the preset mapping relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811206298.6A CN109635625B (en) | 2018-10-16 | 2018-10-16 | Intelligent identity verification method, equipment, storage medium and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811206298.6A CN109635625B (en) | 2018-10-16 | 2018-10-16 | Intelligent identity verification method, equipment, storage medium and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109635625A CN109635625A (en) | 2019-04-16 |
| CN109635625B true CN109635625B (en) | 2023-08-18 |
Family
ID=66066349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811206298.6A Active CN109635625B (en) | 2018-10-16 | 2018-10-16 | Intelligent identity verification method, equipment, storage medium and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109635625B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298243A (en) * | 2019-05-22 | 2019-10-01 | 深圳壹账通智能科技有限公司 | Testimony verification method, apparatus, computer equipment and computer readable storage medium |
| CN111583451A (en) * | 2020-04-09 | 2020-08-25 | 惠州拓邦电气技术有限公司 | Identity verification method and device of electronic lock, computer equipment and storage medium |
| CN111914769B (en) * | 2020-08-06 | 2024-01-26 | 腾讯科技（深圳）有限公司 | User validity determination method, device, computer readable storage medium and equipment |
| CN113705455B (en) * | 2021-08-30 | 2024-03-19 | 平安银行股份有限公司 | Identity verification method and device, electronic equipment and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101751551A (en) * | 2008-12-05 | 2010-06-23 | 比亚迪股份有限公司 | Method, device, system and device for identifying face based on image |
| CN105512535A (en) * | 2016-01-08 | 2016-04-20 | 广东德生科技股份有限公司 | User authentication method and user authentication device |
| WO2016131063A1 (en) * | 2015-02-15 | 2016-08-18 | Alibaba Group Holding Limited | System and method for user identity verification, and client and server by use thereof |
| CN106961418A (en) * | 2017-02-08 | 2017-07-18 | 北京捷通华声科技股份有限公司 | Identity identifying method and identity authorization system |
| CN107066983A (en) * | 2017-04-20 | 2017-08-18 | 腾讯科技（上海）有限公司 | A kind of auth method and device |
-
2018
- 2018-10-16 CN CN201811206298.6A patent/CN109635625B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101751551A (en) * | 2008-12-05 | 2010-06-23 | 比亚迪股份有限公司 | Method, device, system and device for identifying face based on image |
| WO2016131063A1 (en) * | 2015-02-15 | 2016-08-18 | Alibaba Group Holding Limited | System and method for user identity verification, and client and server by use thereof |
| CN105512535A (en) * | 2016-01-08 | 2016-04-20 | 广东德生科技股份有限公司 | User authentication method and user authentication device |
| CN106961418A (en) * | 2017-02-08 | 2017-07-18 | 北京捷通华声科技股份有限公司 | Identity identifying method and identity authorization system |
| CN107066983A (en) * | 2017-04-20 | 2017-08-18 | 腾讯科技（上海）有限公司 | A kind of auth method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109635625A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109635625B (en) | Intelligent identity verification method, equipment, storage medium and device | |
| CN102945366B (en) | A kind of method and device of recognition of face | |
| WO2019174131A1 (en) | Identity authentication method, server, and computer readable storage medium | |
| CN111626371B (en) | Image classification method, device, equipment and readable storage medium | |
| CN109858375B (en) | Living body face detection method, terminal and computer readable storage medium | |
| CN110866466B (en) | Face recognition method, device, storage medium and server | |
| CN111861240A (en) | Suspicious user identification method, device, equipment and readable storage medium | |
| CN101661557A (en) | Face recognition system and face recognition method based on intelligent card | |
| CN109376604B (en) | Age identification method and device based on human body posture | |
| CN109005104B (en) | Instant messaging method, device, server and storage medium | |
| CN104036254A (en) | Face recognition method | |
| CN108171138B (en) | Biological characteristic information acquisition method and device | |
| CN109816543B (en) | Image searching method and device | |
| CN110929244A (en) | Digital identity identification method, device, equipment and storage medium | |
| CN111684459A (en) | Identity authentication method, terminal equipment and storage medium | |
| CN113591603A (en) | Certificate verification method and device, electronic equipment and storage medium | |
| CN111259894B (en) | Certificate information identification method and device and computer equipment | |
| CN112613471A (en) | Face living body detection method and device and computer readable storage medium | |
| US20140072184A1 (en) | Automated image identification method | |
| CN111241930A (en) | Method and system for face recognition | |
| CN111062301A (en) | Identity authentication method and device, electronic equipment and computer readable storage medium | |
| CN111488800A (en) | Model training and image processing method and device, terminal and storage medium | |
| CN111274940A (en) | Face recognition method, device, equipment and storage medium | |
| CN111428670A (en) | Face detection method, face detection device, storage medium and equipment | |
| CN107844735B (en) | Authentication method and device for biological characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |