WO2020000906A1 - 人脸识别身份验证方法、装置、电子设备 - Google Patents
人脸识别身份验证方法、装置、电子设备 Download PDFInfo
- Publication number
- WO2020000906A1 WO2020000906A1 PCT/CN2018/119685 CN2018119685W WO2020000906A1 WO 2020000906 A1 WO2020000906 A1 WO 2020000906A1 CN 2018119685 W CN2018119685 W CN 2018119685W WO 2020000906 A1 WO2020000906 A1 WO 2020000906A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- face
- communication device
- identity information
- target
- identified
- Prior art date
Links
Images
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
Definitions
- the invention relates to the technical field of identity recognition, in particular to a face recognition identity verification method, device, electronic device and storage medium.
- Face recognition technology is a kind of biometric recognition technology. It combines image processing technology, computer graphics, pattern recognition and other research fields. It has broad development prospects and is an effective method for identity verification. Face recognition-based identification methods usually build a face database in advance. During identity verification, the collected images of the face to be verified are compared with the images in the face database, and the comparison is finally determined to be successful. The identity information corresponding to the face image in the face database is used as the identity information of the face to be verified.
- the faces in the face database are usually randomly distributed, when the face database contains a large number of face images, such as the face database of a station or a public security system, face comparison will consume a lot of time, leading to identity Poor real-time verification.
- the face recognition authentication method in the prior art has at least the disadvantage of poor real-time performance, which is not conducive to the promotion of face recognition authentication.
- the present invention provides a face recognition and identity verification method, which at least solves the problem of low efficiency in face recognition based on face recognition in the prior art.
- an embodiment of the present invention provides a face recognition identity verification method including:
- the preset face database includes a correspondence relationship composed of at least a communication device identifier, identity information, and face information, and the target group to be identified is determined according to the communication device identifier and the preset face database.
- Database including:
- a target group database to be identified is constructed according to the identity information and face information corresponding to the communication device identifier of the target group.
- the face information includes at least a face image
- comparing and identifying the face image based on the target group database to be identified, and determining identity information of the target object to be verified includes:
- the determined identity information is used as the identity information of the target object to be verified.
- the face information includes at least a face image
- comparing and identifying the face image based on the target group database to be identified, and determining identity information of the target object to be verified includes:
- the information is the identity information corresponding to a face image in the target group database to be identified from which the reference face features are extracted;
- the determined identity information is used as the identity information of the target object to be verified.
- the face information includes at least face features
- the comparison and recognition of the face image based on the target group database to be identified to determine the identity information of the target object to be verified includes:
- the determined identity information is used as the identity information of the target object to be verified.
- the communication device to which the communication device identifier belongs includes a mobile terminal
- the communication device identifier includes a unique identifier corresponding to the mobile terminal
- the step of obtaining the communication device identifier of the target group in the target detection area includes:
- Communicate with the communication device by simulating a communication base station, and determine a communication device identifier of the communication device.
- an embodiment of the present invention provides a face recognition identity verification device, including:
- a communication device identification acquisition module configured to obtain a communication device identification of a target group within a target detection area
- a to-be-identified target group database determining module configured to determine the to-be-identified target group database according to the communication device identifier and a preset face database;
- a target object face image acquisition module configured to acquire a face image of a target object to be verified in the target detection area
- An identity information determination module configured to compare and identify a face image obtained by the target object face image acquisition module based on the target group database to be identified determined by the target group database to be identified module to determine the target to be verified The identity information of the object.
- the preset face database includes a correspondence relationship consisting of at least a communication device identifier, identity information, and face information.
- the database for determining the target group to be identified includes:
- the communication device identifier comparison submodule is configured to determine the identity information and the person corresponding to the communication device identifier of the target group by comparing the obtained communication device identifier with the communication device identifier in the corresponding relationship. Face information
- a database construction submodule is configured to construct a target group database to be identified according to the identity information and face information corresponding to the communication device identifier of the target group.
- the face information includes at least a face image
- the face image to be identified is compared and identified based on the to-be-recognized target group database determined by the to-be-recognized target group database determination module to determine the to-be-verified target object.
- the identity information determination module includes:
- a first identity verification submodule configured to compare a face image of the target object to be verified with a face image in the target group database to be identified, and determine that the comparison is successful in the target group database to be identified
- the identity information corresponding to the human face image
- the first identity information determination submodule is configured to use the determined identity information as the identity information of the target object to be verified.
- the face information includes at least a face image, and the face image is compared and identified in the target group database to be identified, and the identity information of the target object to be verified is determined,
- the identity information determination module is used for:
- the information is the identity information corresponding to a face image in the target group database to be identified from which the reference face features are extracted;
- the determined identity information is used as the identity information of the target object to be verified.
- the face information includes at least face features, and the face image to be identified is compared and identified based on the to-be-recognized target group database determined by the to-be-recognized target group database determination module to determine the to-be-verified target object.
- the identity information determination module includes:
- a second identity verification submodule configured to obtain a face feature in a face image of the target object to be verified
- the second identity verification sub-module is further configured to compare a face feature in a face image of the target object to be verified with a face feature in a target group database to be identified, and determine whether the comparison is successful.
- the identity information corresponding to a face feature in the target group database to be identified;
- the second identity information determination submodule is configured to use the determined identity information as the identity information of the target object to be verified.
- the communication device to which the communication device identification belongs includes a mobile terminal, and the communication device identification includes a unique identification of the mobile terminal.
- the communication device identification acquisition module is configured to: :
- Communicate with the communication device by simulating a communication base station, and determine a communication device identifier of the communication device.
- an embodiment of the present invention further discloses an electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the processor.
- the computer program implements the face recognition identity verification method described in the embodiment of the present invention.
- an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored.
- the program is executed by a processor, the steps of the face recognition and identity verification method disclosed in the embodiment of the present invention .
- the face recognition identity verification method disclosed in the embodiment of the present invention obtains a communication device identifier of a target group within a target detection area; determines a target group database to be identified according to the communication device identifier and a preset face database; and acquires the target The face image of the target object to be verified in the detection area; the face image is compared and identified based on the target group database to be identified, and the identity information of the target object to be verified is determined, which solves the mass problem in the prior art.
- the face recognition identity verification method disclosed in the embodiment of the present invention reduces the face comparison range through the acquired communication device identification of the target group in the target detection area, and further performs face comparison based on the reduced face database to determine
- the target object's identity information has a small amount of calculation, which helps improve the efficiency of face recognition-based authentication.
- FIG. 1 is a flowchart of a face recognition identity verification method according to the first embodiment of the present invention
- FIG. 2 is a flowchart of a face recognition identity verification method according to a second embodiment of the present invention.
- FIG. 3 is one of the schematic structural diagrams of a face recognition and identity verification device according to a third embodiment of the present invention.
- FIG. 4 is a second schematic structural diagram of a face recognition and identity verification device according to a third embodiment of the present invention.
- a face recognition identity verification method disclosed in this embodiment includes steps 110 to 140.
- Step 110 Obtain a communication device identifier of the target group in the target detection area.
- the communication device in the embodiment of the present invention may be a communication device such as a mobile phone, a phone watch, and the like, which has a unique identifier bound to the user's identity information, and the communication device identifier may be a mobile phone number or a phone number of the phone watch.
- the target object may be a person to be identified.
- a communication device held by an object in an object group in a target detection area can be detected through communication technology.
- a communication query signal is broadcasted through a mobile base station, and a feedback signal from a communication device is received. Then, for a communication device that has already fed back the signal, the communication device identifier of the communication device is further obtained.
- Step 120 Determine a target group database to be identified according to the communication device identifier and a preset face database.
- a face database needs to be constructed in advance.
- the above-mentioned face database includes a plurality of groups of corresponding relationships composed of at least a communication device identifier, identity information, and face information. After the communication device identification of the target group in the target detection area is determined, each communication device identification is matched with the communication device identification in each group of the corresponding relationship in the face database, and the matching face database is successfully matched.
- the identity information and face information corresponding to the communication device identifier are added to the database of the target group to be identified as a data source for authenticating the target object.
- Step 130 Obtain a face image of a target object to be verified in the target detection area.
- the face image of the target object to be verified in the target detection area can be obtained through an image acquisition device such as a high-definition camera.
- Step 140 Compare and identify the obtained face image based on the target group database to be identified, and determine the identity information of the target object to be verified.
- the face information in the preset face database may be a face image. In other embodiments of the present invention, the face information in the preset face database may be a face feature.
- the face information in the target group database to be identified is also a face image.
- the face information in the target group database to be identified is also a face image.
- the face information in the target group database to be identified is also a face image.
- the face information in the target group database to be identified is also a face feature.
- the face information in the target group database to be identified is also a face feature.
- the face recognition identity verification method disclosed in the embodiment of the present invention obtains a communication device identifier of a target group in a target detection area; determines a target group database to be identified according to the communication device identifier and a preset face database; Verify the face image of the target object; compare and identify the obtained face image of the target object to be verified based on the target group database to be identified, determine the identity information of the target object to be verified, and solve the problem of massive faces in the prior art The inefficiency of face recognition in the library for identity verification.
- the face recognition identity verification method disclosed in the embodiment of the present invention reduces the face comparison range through the acquired communication device identification of the target group in the target detection area, and further performs face comparison based on the reduced face database to determine
- the target object's identity information has a small amount of calculation, which helps improve the efficiency of face recognition-based authentication.
- a face recognition identity verification method disclosed in this embodiment includes steps 210 to 250.
- Step 210 Obtain a preset face database.
- the aforementioned face database includes a correspondence relationship composed of at least a communication device identifier, identity information, and face information.
- the preset face database includes multiple groups of correspondences, and the format of each group of correspondences is as follows (communication device identification, identity information, face information).
- the format of each group of correspondences is as follows (communication device identification, identity information, face information).
- real-name systems such as mobile phone numbers or phone number applications, opening bank accounts, and processing credential information
- a face database including mobile phone numbers, identity information, and face images is easily available.
- the communication device to which the communication device identifier belongs may be a mobile phone, and the communication device identifier is a mobile phone number.
- the identity information is determined according to specific business needs.
- the identity information may include any one or more of a name, an ID number, a home address, a work address, and vehicle information.
- the face information may be a face image or a face feature extracted from the face image.
- Step 220 Obtain the communication device identifier of the target group in the target detection area.
- the target object described in this embodiment is a person to be identified.
- the target detection area is traffic intersections
- the target group is pedestrians waiting to cross the road and / or crossing the road at the traffic intersection.
- the purpose of identity verification is to determine the identity of pedestrians who cross the red light .
- a communication device is a mobile phone
- the communication device identifier is a mobile phone number as an example, and the technical solution of the authentication method based on face recognition is described in detail.
- the detection of mobile phone signals can be implemented in various ways, such as detecting mobile phone signals by detecting Bluetooth signals, WiFi signals, and GPRS signals.
- the detection transmission distance based on the Bluetooth signal can reach 10 cm-10 meters, and the technology has low power consumption and simple application, but the accuracy is slightly lower.
- the detection of mobile phone signals based on WiFi signals has become more and more common. It is possible to collect the WiFi signals in a specific area and achieve the positioning of monitored objects through corresponding calculations with high accuracy.
- Mobile phone detection based on GPRS signal can be combined with GPS. After acquiring the GPS output signal, the mobile phone can be connected to the Internet using GPRS network for mobile phone positioning and monitoring.
- the communication device to which the communication device identifier belongs includes a mobile terminal, and the communication device identifier includes a unique identifier corresponding to the mobile terminal.
- Obtaining the communication device identifier of the target group in the target detection area includes: detecting the object in the target detection area through non-contact communication technology Communication equipment held in the group; communicating with the communication equipment by simulating a communication base station, and determining a communication equipment identifier of the communication equipment.
- the communication device may be a mobile phone, and the communication device identifier may be a mobile phone number.
- a mobile phone signal detection device may be provided at a traffic intersection, and the mobile phone signal detection device scans a nearby mobile phone through contactless communication technology to obtain a mobile phone identification, such as an IMIS code.
- the mobile phone number corresponding to each mobile phone identifier is obtained by calling the data interface of the telecommunication system according to the mobile phone identifier.
- the mobile phone signal detection device simulates a mobile base station, sends an identification signal to a nearby mobile phone, and further analyzes a feedback signal of the mobile phone to obtain a mobile phone number of a mobile phone held by a pedestrian in the target group.
- the above non-contact communication technology may be Bluetooth technology, WiFi communication technology, GPRS communication technology, and the like.
- Step 230 Determine a target group database to be identified according to the communication device identifier and a preset face database.
- the preset face database includes a correspondence relationship consisting of at least a communication device identifier, identity information, and face information, and a target to be determined is determined according to the communication device identifier and the preset face database.
- Identifying the target group database includes: determining the identity information and face information corresponding to the communication device identifier of the target group by comparing the obtained communication device identifier with the communication device identifier in the above-mentioned correspondence; The identification information and face information corresponding to the communication device identifier of the target group construct a target group database to be identified.
- the mobile phone numbers number1 and number2 are respectively compared with the preset. It is assumed that all mobile phone numbers (that is, communication device identifications) in the face database are matched.
- the identity information ID1 and face information Face1 corresponding to the mobile phone number1 in the preset face database will be identified in the preset face database corresponding to the mobile phone number2
- the information ID2 and face information Face2 are used as reference information for subsequent comparison and identification of the target object.
- the identity information ID1 and face information Face1 corresponding to the mobile phone number1 in the preset face database and the mobile phone in the preset face database and the mobile phone are used.
- the identification information ID2 and the face information Face2 corresponding to the number number2 constitute a database of target groups to be identified in a format of (identity information, face information).
- the data in the target group database to be identified is: ⁇ (identity information ID1, face information Face1), (identity information ID2, face information Face2) ⁇ .
- Step 240 Obtain a face image of a target object to be verified in the target detection area.
- a face image of a pedestrian crossing a red light at a traffic intersection may be collected by a high-definition camera provided at the traffic intersection. Face images of pedestrians who run through red lights can also be collected by traffic police through handheld image acquisition equipment. In this embodiment, it is assumed that the obtained face image of the target object A to be verified is PictureA.
- Step 250 Compare and identify the acquired face image based on the target group database to be identified, and determine the identity information of the target object to be verified.
- the face information includes at least a face image
- the obtained face image is compared and identified based on the target group database to be identified
- determining the identity information of the target object to be verified includes: The face image of the subject is compared with the face image in the target group database to be identified, and the identity information corresponding to the face image in the target group database that is successfully compared is determined; the determined identity information is used as the target to be verified The identity information of the object.
- the face information in the face database is preset to be a face image
- the face information Face1 and Face2 in the target group database to be identified are also face images.
- the identity information corresponding to the face image in the target group database to be successfully identified is used as the identity information of the target object to be verified.
- the identity information ID1 corresponding to the face image Face1 is used as the identity information of the target object A to be verified.
- the face information includes at least a face image
- the face image is compared and identified based on the target group database to be identified
- determining the identity information of the target object to be verified includes: Extract the face features to be verified from the face image, and extract the reference face features from the face images in the target group database to be identified; compare the face features to be verified with the reference face features to determine that the comparison is successful
- the identity information corresponding to the reference face feature wherein the identity information corresponding to the reference face feature is the identity information corresponding to the face image in the target group database to be identified from which the reference face feature is extracted; the determined identity information is used as The identity information of the target object to be verified.
- the face information Face1 and Face2 in the target group database to be identified are also face images.
- After obtaining the face image PictureA of the target object A to be verified first extract the face feature FeatureA from the face image PictureA, and extract the reference face features Feature1 and Feature2 from the face information Face1 and Face2 respectively; then, Perform the to-be-verified feature FeatureA extracted from the obtained face image PictureA of the to-be-verified target object A and the reference face features Feature1 and Feature2 extracted from the face images Face1 and Face2 in the to-be-recognized target group database Comparison recognition, to determine the reference facial features for comparison success.
- an example in which the reference face feature that is successfully compared is Feature2 is used as an example, and the identity information corresponding to the face image Face2 in the target group database to be identified with the reference face feature as Feature2 is extracted as the The identity information of the target object to be verified.
- the face information includes facial features
- the obtained face image is compared and identified based on the target group database to be identified
- determining the identity information of the target object to be verified includes: obtaining the target to be verified
- the facial features in the face image of the subject; the facial features in the face image of the target object to be verified are compared with the facial features in the target group database to be identified, and the database of the target group to be identified that is successfully compared is determined
- the identity information corresponding to the facial features in; the determined identity information is used as the identity information of the target object to be verified.
- the face information Face1 and Face2 in the target group database to be identified are also face features.
- the face information Face1 and Face2 in the target group database to be identified are also face features.
- the facial feature FeatureA to be verified is compared with the facial features Face1 and Face2 in the target group database to be identified, and the facial features that are successfully compared are determined.
- the identity information corresponding to the facial features in the database of the target group to be identified that is successfully compared is used as the identity information of the target object to be verified.
- the identity information ID1 corresponding to the facial feature Face1 is used as the identity information of the target object A to be verified.
- a technical solution for extracting a face feature from a face image may adopt an existing technology. For example: firstly perform image processing such as contrast enhancement gray transform, median filtering, histogram equalization, and size normalization on the face image of the target object to be verified obtained by the camera to obtain a pre-processed image; The face detection technology determines the size, posture and position of the face in the pre-processed image; finally, extracts the facial features in the pre-processed image based on the key position points of the face and the contour information of the face.
- the contour information of the extracted key position points and the face may include position information such as the corners of the eyes, the ends of the eyebrows, the corners of the mouth, and the tip of the nose.
- a face image of a target object to be verified in the target detection area may be obtained first, and then a communication device identifier of the target group in the target detection area may be obtained, and determined according to the communication device identifier and a preset face database
- the target group database to be identified; the face image and the communication device identification of the target group in the target detection area can also be obtained at the same time.
- the execution order of the steps is not limited.
- the face database used as a comparison data source in the prior art for face comparison needs to include a large amount of face information, and the face database of the public security system
- the nationwide face database will include hundreds of millions of face images or hundreds of millions of face features.
- the face image or face feature of the target object needs to be compared hundreds of millions of times to determine the degree of matching.
- the highest face image or set of face features will take a long time.
- the information of the target group in the target detection area in the face database is first extracted through the mobile phone number, which can effectively reduce the calculation amount of later comparison and recognition.
- the face recognition identity verification method disclosed in the present invention enables the application of long-distance and fast face recognition identity verification in public places such as scenic spots, stations, airports, traffic intersections, and squares to be implemented and promoted.
- the face recognition identity verification method disclosed in the embodiment of the present invention obtains a communication device identifier of an object group in a target detection area; determines a target group database to be identified according to the communication device identifier and a preset face database; and acquires a target group in the target detection area. Verify the face image of the target object; compare and identify the face image based on the target group database to be identified, determine the identity information of the target object to be verified, and solve the face recognition based on the massive face database in the prior art Inefficiency in authentication.
- the face recognition identity verification method disclosed in the embodiment of the present invention reduces the face comparison range through the acquired communication device identification of the target group in the target detection area, and further performs face comparison based on the reduced face database to determine
- the target object's identity information has a small amount of calculation, which helps improve the efficiency of face recognition-based authentication.
- reducing the face database to be compared by mobile phone number, and then performing identity verification through face information can further improve the accuracy of identity verification.
- a face recognition and identity verification device disclosed in this embodiment includes:
- a communication device identifier acquiring module 310 configured to acquire a communication device identifier of a target group within a target detection area
- a target group database determining module 320 configured to determine the target group database to be identified according to the communication device identifier and a preset face database
- a target object face image acquisition module 330 configured to acquire a face image of a target object to be verified in the target detection area
- the identity information determination module 340 is configured to compare and identify the face image obtained by the target object face image acquisition module 330 based on the target group database to be identified determined by the target group database to be identified 320 to determine the identity of the target object to be verified information.
- the preset face database includes a correspondence relationship composed of at least a communication device identifier, identity information, and face information, and the target group to be identified is determined according to the communication device identifier and the preset face database.
- the target group database determining module 320 includes:
- the communication device identifier comparison submodule 3201 is configured to determine identity information corresponding to the communication device identifier of the target group by comparing the obtained identification of the communication device with the communication device identifier in the correspondence relationship. And face information;
- a database construction submodule 3202 is configured to construct a target group database to be identified according to the identity information and face information corresponding to the communication device identifier of the target group.
- the face information includes at least a face image.
- the identity information determination module 340 includes:
- a first identity verification submodule 3401 configured to compare the face image of the target object to be verified with the face image in the target group database to be identified, and determine that the face image in the target group database to be identified that corresponds successfully corresponds to Identity information
- the first identity information determination submodule 3402 is configured to use the determined identity information as identity information of a target object to be verified.
- the face information includes at least face features, and based on the target group database to be identified determined by the target group database determining module 320, the face image acquired by the target object face image acquisition module 330 is performed.
- the identity information determination module 340 includes:
- a second identity verification submodule 3403 configured to obtain face features in a face image of a target object to be verified
- the second identity verification submodule 3403 is further configured to compare the facial features in the face image of the target object to be verified with the facial features in the target group database to be identified, and determine the database of the target group to be identified that is successfully compared. Identity information corresponding to facial features in
- the second identity information determination sub-module 3404 is configured to use the determined identity information as the identity information of the target object to be verified.
- the face information includes at least a face image, and based on the target group database to be identified determined by the target group database determining module 320, the face image acquired by the target object face image acquiring module 330 is performed.
- the identity information determination module 340 includes:
- a third identity verification sub-module 3405 configured to extract facial features to be verified from the facial image of the target to be verified, and extract reference facial features from the facial images in the target group database to be identified;
- the third identity verification sub-module 3405 is further configured to compare the face features to be verified with reference face features, and determine identity information corresponding to the reference face features that are successfully compared, where the reference face features correspond to
- the identity information is identity information corresponding to a face image in a target group database to be identified from which a reference face feature is extracted;
- the third identity information determining submodule 3406 is configured to use the determined identity information as the identity information of the target object to be verified.
- the communication device to which the communication device identification belongs includes a mobile terminal
- the communication device identification includes a unique identification corresponding to the mobile terminal
- the communication device identification acquisition module 310 is further configured to:
- the mobile terminal may be a mobile phone or a telephone watch.
- the communication device identifier may be a mobile phone number.
- the face recognition and identity verification device disclosed in this embodiment is used to implement the face recognition and identity verification method described in the first embodiment.
- each module of the device refer to the specific implementation of corresponding steps in the first embodiment. The embodiment will not be described again.
- the face recognition identity verification device disclosed in the embodiment of the present invention obtains a communication device identifier of a target group in a target detection area, and determines a target group database to be identified according to the communication device identifier and a preset face database to acquire a target in the target detection area. Verify the face image of the target object, and compare and recognize the obtained face image based on the target group database to be identified, determine the identity information of the target object to be verified, and solve the problem of face-based Identify inefficiencies in authentication.
- the face recognition and identity verification device disclosed in the embodiment of the present invention reduces the face comparison range by acquiring the communication device identifier of the target group in the target detection area, and further performs face comparison based on the reduced face database to determine
- the target object's identity information has a small amount of calculation, which helps improve the efficiency of face recognition-based authentication.
- reducing the face database to be compared by mobile phone number, and then performing identity verification through face information can further improve the accuracy of identity verification.
- the present invention also discloses an electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor.
- the processor executes the computer program, the implementation of the present invention is implemented.
- the electronic device may be a PC, a mobile terminal, a personal digital assistant, a tablet computer, or the like.
- the present invention also discloses a computer-readable storage medium on which a computer program is stored.
- the program is executed by a processor, the steps of the face recognition identity verification method according to the first and second embodiments of the present invention are implemented.
- the embodiments can be implemented by means of software plus a necessary universal hardware platform, and of course, they can also be implemented by hardware.
- the above-mentioned technical solution essentially or part that contributes to the existing technology can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic A disc, an optical disc, and the like include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
本发明公开了一种人脸识别身份验证方法,属于身份识别技术领域,解决现有技术中身份验证实时性较差的问题,该方法包括:获取目标检测区域内对象群体的通信设备标识,并根据通信设备标识和预设人脸库,确定待识别目标群体数据库,及获取目标检测区域内待验证目标对象的人脸图像,并基于待识别目标群体数据库对获取的人脸图像进行比对识别,确定待验证目标对象的身份信息。本发明实施例公开的人脸识别身份验证方法,通过获取的目标检测区域内对象群体的通信设备标识,缩小人脸比对范围,进一步基于缩小后的人脸库进行人脸比对,以确定目标对象的身份信息,运算量小,有助于提高基于人脸识别的身份验证的效率。
Description
本发明涉及身份识别技术领域,特别是涉及一种人脸识别身份验证方法、装置、电子设备及存储介质。
人脸识别技术是生物特征识别技术的一种,它结合了图像处理技术、计算机图形学、模式识别等多个研究领域,具有广阔的发展前景,是进行身份验证的有效的手段。基于人脸识别的身份识别方法,通常是预先构建人脸库,在身份验证时,将采集的待验证人脸的图像和人脸库中的图像进行一一比对,最终确定比对成功的人脸库中的人脸图像对应的身份信息作为待验证人脸的身份信息。
然而,由于人脸库中的人脸通常是随机分布的,当人脸库中包含大量人脸图像时,例如车站、公安系统的人脸库,人脸比对将会消耗大量时间,导致身份验证实时性较差。
可见,现有技术中的人脸识别身份验证方法至少存在实时性较差的缺陷,不利于人脸识别身份验证的推广。
发明内容
本发明提供一种人脸识别身份验证方法,至少解决现有技术中的基于人脸识别进行身份验证时效率低下的问题。
为了解决上述问题,根据本发明的一个方面,本发明实施例提供了一种人脸识别身份验证方法包括:
获取目标检测区域内对象群体的通信设备标识;
根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库;
获取所述目标检测区域内待验证目标对象的人脸图像;
基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息。
可选的,所述预设人脸库中包括至少由通信设备标识、身份信息和人脸信息组成的对应关系,所述根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库,包括:
通过将获取的所述通信设备标识和所述对应关系中的所述通信设备标识进行比对,确定所述对象群体的通信设备标识对应的身份信息和人脸信息;
根据所述对象群体的通信设备标识对应的身份信息和人脸信息构建待识别目标群体数据库。
可选的,所述人脸信息至少包括人脸图像,所述基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息,包括:
将所述待验证目标对象的人脸图像与所述待识别目标群体数据库中的人脸图像进行比对,确定比对成功的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;
将确定的所述身份信息,作为所述待验证目标对象的身份信息。
可选的,所述人脸信息至少包括人脸图像,所述基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息,包括:
从所述待验证目标对象的人脸图像中提取待验证人脸特征,以及,从所述待识别目标群体数据库中的人脸图像中提取参考人脸特征;
将所述待验证人脸特征与所述参考人脸特征进行比对,确定比对成功的所述参考人脸特征对应的所述身份信息,其中,所述参考人脸特征对应的所述身份信息为从中提取所述参考人脸特征的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;
将确定的所述身份信息,作为所述待验证目标对象的身份信息。
可选的,所述人脸信息至少包括人脸特征,所述基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息,包括:
获取所述待验证目标对象的人脸图像中的人脸特征;
将所述待验证目标对象的人脸图像中的人脸特征与所述待识别目标 群体数据库中的人脸特征进行比对,确定比对成功的所述待识别目标群体数据库中的人脸特征对应的所述身份信息;
将确定的所述身份信息,作为所述待验证目标对象的身份信息。
可选的,所述通信设备标识所属通信设备包括移动终端,所述通信设备标识包括移动终端对应的唯一标识,所述获取目标检测区域内对象群体的通信设备标识的步骤,包括:
通过非接触通信技术,检测目标检测区域内对象群体中持有的通信设备;
通过模拟通信基站的方式与所述通信设备进行通信,确定所述通信设备的通信设备标识。
根据本发明的另一个方面,本发明实施例提供了一种人脸识别身份验证装置,包括:
通信设备标识获取模块,用于获取目标检测区域内对象群体的通信设备标识;
待识别目标群体数据库确定模块,用于根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库;
目标对象人脸图像获取模块,用于获取所述目标检测区域内待验证目标对象的人脸图像;
身份信息确定模块,用于基于所述待识别目标群体数据库确定模块确定的待识别目标群体数据库对所述目标对象人脸图像获取模块获取的人脸图像进行比对识别,确定所述待验证目标对象的身份信息。
可选的,所述预设人脸库中包括至少由通信设备标识、身份信息和人脸信息组成的对应关系,根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库时,所述待识别目标群体数据库确定模块包括:
通信设备标识比对子模块,用于通过将获取的所述通信设备标识和所述对应关系中的所述通信设备标识进行比对,确定所述对象群体的通信设备标识对应的身份信息和人脸信息;
数据库构建子模块,用于根据所述对象群体的通信设备标识对应的身份信息和人脸信息构建待识别目标群体数据库。
可选的,所述人脸信息至少包括人脸图像,基于所述待识别目标群体 数据库确定模块确定的待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息时,所述身份信息确定模块包括:
第一身份验证子模块,用于将所述待验证目标对象的人脸图像与所述待识别目标群体数据库中的人脸图像进行比对,确定比对成功的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;
第一身份信息确定子模块,用于将确定的所述身份信息,作为所述待验证目标对象的身份信息。
可选的,所述人脸信息至少包括人脸图像,所述于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息时,所述身份信息确定模块用于:
从所述待验证目标对象的人脸图像中提取待验证人脸特征,以及,从所述待识别目标群体数据库中的人脸图像中提取参考人脸特征;
将所述待验证人脸特征与所述参考人脸特征进行比对,确定比对成功的所述参考人脸特征对应的所述身份信息,其中,所述参考人脸特征对应的所述身份信息为从中提取所述参考人脸特征的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;
将确定的所述身份信息,作为所述待验证目标对象的身份信息。
可选的,所述人脸信息至少包括人脸特征,基于所述待识别目标群体数据库确定模块确定的待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息时,所述身份信息确定模块包括:
第二身份验证子模块,用于获取所述待验证目标对象的人脸图像中的人脸特征;
所述第二身份验证子模块,还用于将所述待验证目标对象的人脸图像中的人脸特征与所述待识别目标群体数据库中的人脸特征进行比对,确定比对成功的所述待识别目标群体数据库中的人脸特征对应的所述身份信息;
第二身份信息确定子模块,用于将确定的所述身份信息,作为所述待验证目标对象的身份信息。
可选的,所述通信设备标识所属通信设备包括移动终端,所述通信设备标识包括移动终端的唯一标识,获取目标检测区域内对象群体的通信设备标识时,所述通信设备标识获取模块用于:
通过非接触通信技术,检测目标检测区域内对象群体中持有的通信设备;
通过模拟通信基站的方式与所述通信设备进行通信,确定所述通信设备的通信设备标识。
根据本发明的另一个方面,本发明实施例还公开了一种电子设备,包括存储器、处理器及存储在所述存储器上并可在处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现本发明实施例所述的人脸识别身份验证方法。
根据本发明的另一个方面,本发明实施例提供了一种计算机可读存储介质,其上存储有计算机程序,该程序被处理器执行时本发明实施例公开的人脸识别身份验证方法的步骤。
本发明实施例公开的人脸识别身份验证方法,通过获取目标检测区域内对象群体的通信设备标识;根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库;获取所述目标检测区域内待验证目标对象的人脸图像;基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息,解决了现有技术中在海量人脸库中基于人脸识别进行身份验证时效率低下的问题。本发明实施例公开的人脸识别身份验证方法,通过获取的目标检测区域内对象群体的通信设备标识,缩小人脸比对范围,进一步基于缩小后的人脸库进行人脸比对,以确定目标对象的身份信息,运算量小,有助于提高基于人脸识别的身份验证的效率。
为了更清楚地说明本发明实施例的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是本发明实施例一的人脸识别身份验证方法流程图;
图2是本发明实施例二的人脸识别身份验证方法的流程图;
图3是本发明实施例三的人脸识别身份验证装置的结构示意图之一;
图4是本发明实施例三的人脸识别身份验证装置的结构示意图之二。
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
实施例一
本实施例公开的一种人脸识别身份验证方法,如图1所示,该方法包括:步骤110至步骤140。
步骤110,获取目标检测区域内对象群体的通信设备标识。
本发明实施例中的通信设备可以为:手机、电话手表等具有与用户的身份信息绑定的唯一标识的通信设备,其中,通信设备标识可以为手机号码、电话手表的电话号码。目标对象可以为待识别的人。
本发明在具体实施过程中,可以通过通信技术检测目标检测区域内对象群体中对象持有的通信设备。
例如,通过移动基站广播通信查询信号,并接收通信设备的反馈信号,然后,对于已经反馈信号的通信设备,进一步获取该通信设备的通信设备标识。
步骤120,根据通信设备标识和预设人脸库,确定待识别目标群体数据库。
本发明具体实施时,需要预先构建人脸库。上述人脸库中包括至少由通信设备标识、身份信息和人脸信息组成的多组对应关系。当确定了目标检测区域内对象群体的通信设备标识之后,对于每个通信设备标识,分别与上述人脸库中每组对应关系中的通信设备标识进行匹配,并将匹配成功的人脸库中的通信设备标识所对应的身份信息和人脸信息加入到待识别目标群体数据库中,作为对目标对象进行身份验证的数据源。
步骤130,获取目标检测区域内待验证目标对象的人脸图像。
在具体应用过程中,可以通过高清摄像头等图像采集设备获取目标检测区域内待验证目标对象的人脸图像。
步骤140,基于待识别目标群体数据库对上述获取的人脸图像进行比对识别,确定上述待验证目标对象的身份信息。
在本发明的一些实施例中,预设人脸库中的人脸信息可以为人脸图像,在本发明的另一些实施例中,预设人脸库中的人脸信息可以为人脸特征。
当预设人脸库中的人脸信息为人脸图像时,待识别目标群体数据库中的人脸信息也为人脸图像。在获取到待验证目标对象的人脸图像之后,直接通过将获取到的待验证目标对象的人脸图像与所述待识别目标群体数据库中的人脸图像进行比对识别,确定上述待验证目标对象的身份信息;或者分别从待验证目标对象的人脸图像中和待识别目标群体数据库中的人脸图像中提取人脸特征,然后通过将提取的人脸特征进行比对,确定待验证目标对象的身份信息。
当预设人脸库中的人脸信息为人脸特征时,待识别目标群体数据库中的人脸信息也为人脸特征。在获取到待验证目标对象的人脸图像之后,首先从待验证目标对象的人脸图像中提取人脸特征,然后,通过将待验证目标对象的人脸图像中的人脸特征与待识别目标群体数据库中的人脸特征进行比对识别,确定所述待验证目标对象的身份信息。
本发明实施例公开的人脸识别身份验证方法,通过获取目标检测区域内对象群体的通信设备标识;根据通信设备标识和预设人脸库,确定待识别目标群体数据库;获取目标检测区域内待验证目标对象的人脸图像;基于待识别目标群体数据库对获得的待验证目标对象的人脸图像进行比对识别,确定上述待验证目标对象的身份信息,解决了现有技术中在海量人脸库中基于人脸识别进行身份验证时效率低下的问题。
本发明实施例公开的人脸识别身份验证方法,通过获取的目标检测区域内对象群体的通信设备标识,缩小人脸比对范围,进一步基于缩小后的人脸库进行人脸比对,以确定目标对象的身份信息,运算量小,有助于提高基于人脸识别的身份验证的效率。
实施例二
本实施例公开的一种人脸识别身份验证方法,如图2所示,该方法包括:步骤210至步骤250。
步骤210,获取预设人脸库。
在本发明的一些实施例中,上述人脸库包括至少由通信设备标识、身份信息和人脸信息组成的对应关系。
例如,预设人脸库包括多组对应关系,每组对应关系的格式如下(通信设备标识,身份信息,人脸信息)。随着越来越多的系统要求实名制,例如手机号或电话号码申请、开立银行账户、办理证件信息,因此,包括手机号码、身份信息和人脸图像的人脸库很容易获得。
具体实施时,通信设备标识所属通信设备可以为手机,则通信设备标识为手机号码。其中,身份信息根据具体业务需求确定。
例如,对于交通系统的人脸库,身份信息可以包括:姓名、身份证号码、家庭住址、工作地址、拥有车辆信息等中的任意一项或多项。人脸信息可以为人脸图像,也可以为从人脸图像中提取的人脸特征。
步骤220,获取目标检测区域内对象群体的通信设备标识。
本实施例中所述目标对象为待识别的人。以交通路口对违章行人进行身份验证的应用场景为例,目标检测区域为交通路口,对象群体为交通路口等待过马路和/或正在过马路的行人,身份验证目的是确定闯红灯的行人的身份信息。
为了便于读者理解本发明,本实施例中以通信设备为手机,所述通信设备标识为手机号码为例,详细说明基于人脸识别的身份验证方法的技术方案。
目前手机信号的检测可以通过多种方式实现,如通过检测蓝牙信号、WiFi信号、GPRS信号来检测手机信号。基于蓝牙信号的检测传输距离可达到10厘米-10米,且技术功耗低,应用简单,但是精度稍低。基于WiFi信号的手机信号检测随着智能手机的推广变得越来越普遍,可以通过对特定区域的WiFi信号进行采集,通过相应的计算实现对监控物体的定位,精度较高。基于GPRS信号的手机检测可以和GPS相结合,获取GPS输出的信号后,利用GPRS网络可以将手机接入互联网,进行手机的定位与监控。
优选的,通信设备标识所属通信设备包括移动终端,通信设备标识包括移动终端对应的唯一标识,获取目标检测区域内对象群体的通信设备标识,包括:通过非接触通信技术,检测目标检测区域内对象群体中持有的通信设备;通过模拟通信基站的方式与上述通信设备进行通信,确定上述通信设备的通信设备标识。优选的,通信设备可为手机,通信设备标识可为手机号码。
具体实施时,可以在交通路口设置手机信号检测设备,手机信号检测设备通过非接触通信技术扫描附近的手机,以获取手机标识,如IMIS码。
然后,根据手机标识通过调用电信系统的数据接口获取每个手机标识对应的手机号码。或者,手机信号检测设备模拟移动基站,向附近的手机发送识别信号,并进一步通过对手机的反馈信号进行解析,得到所述对象群体中行人持有的手机的手机号码。
具体实施时,上述非接触通信技术可以为蓝牙技术、WiFi通信技术、GPRS通信技术等。
具体实施时,获取目标检测区域内对象群体的通信设备标识的具体技术方案,可以参考现有技术中确定一定区域内通信设备标识的技术方案,本发明实施例中不再赘述。
步骤230,根据通信设备标识和预设人脸库,确定待识别目标群体数据库。
在本发明的一些实施例中,所述预设人脸库中包括至少由通信设备标识、身份信息和人脸信息组成的对应关系,根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库,包括:通过将获取的所述通信设备标识和上述对应关系中的通信设备标识进行比对,确定所述对象群体的通信设备标识对应的身份信息和人脸信息;根据所述对象群体的通信设备标识对应的身份信息和人脸信息构建待识别目标群体数据库。
例如,通过前述步骤确定的在交通路口等待过马路和/或正在过马路的人群中,存在持有手机号码number1和number2的行人,则本步骤中,将手机号码number1和number2分别和所述预设人脸库中的所有手机号码(即通信设备标识)进行匹配。
如果预设人脸库中存在手机号码number1和number2,则将预设人脸 库中与手机号码number1对应的身份信息ID1和人脸信息Face1、预设人脸库中与手机号码number2对应的身份信息ID2和人脸信息Face2作为后续对目标对象进行比对识别的参考信息,由预设人脸库中与手机号码number1对应的身份信息ID1和人脸信息Face1、预设人脸库中与手机号码number2对应的身份信息ID2和人脸信息Face2构成格式为(身份信息,人脸信息)的待识别目标群体数据库。
本实施例中,上述待识别目标群体数据库中的数据为:{(身份信息ID1,人脸信息Face1),(身份信息ID2,人脸信息Face2)}。
步骤240,获取目标检测区域内待验证目标对象的人脸图像。
具体实施时,可以通过设置在交通路口的高清摄像头采集交通路口闯红灯的行人的人脸图像。也可以由交通警察通过手持图像采集设备采集闯红灯的行人的人脸图像。本实施例中,假设获取的待验证目标对象A的人脸图像为PictureA。
步骤250,基于待识别目标群体数据库对上述获取的人脸图像进行比对识别,确定上述待验证目标对象的身份信息。
在本发明的一些实施例中,人脸信息至少包括人脸图像,基于待识别目标群体数据库对获得的人脸图像进行比对识别,确定待验证目标对象的身份信息,包括:将待验证目标对象的人脸图像与待识别目标群体数据库中的人脸图像进行比对,确定比对成功的待识别目标群体数据库中的人脸图像对应的身份信息;将确定的身份信息,作为待验证目标对象的身份信息。
例如,当预设人脸库中的人脸信息为人脸图像时,待识别目标群体数据库中的人脸信息Face1和Face2也为人脸图像。在获取到待验证目标对象A的人脸图像PictureA之后,直接将获取到的待验证目标对象A的人脸图像PictureA与所述待识别目标群体数据库中的人脸图像Face1和Face2进行比对识别,确定比对成功的人脸图像。
然后,将比对成功的所述待识别目标群体数据库中的人脸图像对应的身份信息,作为所述待验证目标对象的身份信息。本实施例中,假设人脸图像PictureA与人脸图像Face1比对成功,则将人脸图像Face1对应的身份信息ID1作为所述待验证目标对象A的身份信息。两幅人脸图像进行比 对识别的具体实施方式为现有技术,本发明实施例中不再赘述。
在本发明的一些实施例中,人脸信息至少包括人脸图像,基于待识别目标群体数据库对人脸图像进行比对识别,确定待验证目标对象的身份信息,包括:从待验证目标对象的人脸图像中提取待验证人脸特征,以及,从待识别目标群体数据库中的人脸图像中提取参考人脸特征;将待验证人脸特征与参考人脸特征进行比对,确定比对成功的参考人脸特征对应的身份信息,其中,参考人脸特征对应的身份信息为从中提取参考人脸特征的待识别目标群体数据库中的人脸图像对应的身份信息;将确定的身份信息,作为待验证目标对象的身份信息。
例如,当预设人脸库中的人脸信息为人脸图像时,待识别目标群体数据库中的人脸信息Face1和Face2也为人脸图像。在获取到待验证目标对象A的人脸图像PictureA之后,首先从人脸图像PictureA中提取待验证人脸特征FeatureA,从人脸信息Face1和Face2中分别提取参考人脸特征Feature1和Feature2;然后,将获取到的待验证目标对象A的人脸图像PictureA中提取的待验证人脸特征FeatureA与所述待识别目标群体数据库中的人脸图像Face1和Face2中提取的参考人脸特征Feature1和Feature2进行比对识别,确定比对成功的参考人脸特征。本实施例中,以比对成功的参考人脸特征为Feature2举例,将提取所述参考人脸特征为Feature2的所述待识别目标群体数据库中的人脸图像Face2对应的身份信息,作为所述待验证目标对象的身份信息。
在本发明的另一些实施例中,人脸信息包括人脸特征,基于待识别目标群体数据库对获取的人脸图像进行比对识别,确定待验证目标对象的身份信息,包括:获取待验证目标对象的人脸图像中的人脸特征;将待验证目标对象的人脸图像中的人脸特征与待识别目标群体数据库中的人脸特征进行比对,确定比对成功的待识别目标群体数据库中的人脸特征对应的身份信息;将确定的身份信息,作为待验证目标对象的身份信息。
例如,当预设人脸库中的人脸信息为人脸特征时,待识别目标群体数据库中的人脸信息Face1和Face2也为人脸特征。在获取到待验证目标对象A的人脸图像PictureA之后,首先采用与提取人脸库中人脸Face1和Face2相同的方法从待验证目标对象A的人脸图像PictureA中提取待验证 人脸特征FeatureA,然后,将所述待验证人脸特征FeatureA与所述待识别目标群体数据库中的人脸特征Face1和Face2进行比对识别,确定比对成功的人脸特征。
然后,将比对成功的待识别目标群体数据库中的人脸特征对应的身份信息,作为待验证目标对象的身份信息。假设人脸特征FeatureA与人脸特征Face1比对成功,则将人脸特征Face1对应的身份信息ID1作为所述待验证目标对象A的身份信息。
从人脸图像中提取人脸特征的技术方案可以采用现有技术。例如:首先对摄像头获取的所述待验证目标对象的人脸图像进行对比度增强灰度变换、中值滤波、直方图均衡化以及尺寸归一化等图像处理,得到预处理图像;然后,通过人脸检测技术确定所述预处理图像中人脸的大小,姿态和位置;最后,基于人脸的关键位置点和人脸的轮廓信息提取所述预处理图像中的人脸特征。其中,提取关键位置点和人脸的轮廓信息可以包括眼角、眉毛的末端、嘴角、鼻尖等位置信息。
本发明具体实施时,还可以首先获取目标检测区域内待验证目标对象的人脸图像,然后再获取目标检测区域内对象群体的通信设备标识,以及根据通信设备标识和预设人脸库,确定待识别目标群体数据库;也可以同时获取人脸图像和目标检测区域内对象群体的通信设备标识,本发明对获取目标检测区域内待验证目标对象的人脸图像的步骤和确定待识别目标群体数据库的步骤的执行先后顺序不做限定。
通过前述描述可以,对于公共场所需要进行人脸识别的应用,现有技术中进行人脸比对时作为比对数据源的人脸库中需要包括大量人脸信息,以公安系统的人脸库为例,全国范围的人脸库中会包括上亿幅人脸图像或上亿组人脸特征,目标对象的人脸图像或人脸特征需要进行上亿次的比对,才能最终确定匹配度最高的一幅人脸图像或一组人脸特征,会消耗很长时间。
通过本发明公开的方法,首先通过手机号码提取出目标检测区域内对象群体在人脸库中的信息,可以有效减小后期比对识别的运算量。
以交通路口为例,每个交通路口通常只有十几个,最多不超过一百人在等待和正在过马路,即通过手机信号识别,最终最多从人脸库中提取出 一百人的身份信息和人脸信息用于后续比对,确定闯红灯的行人的身份信息,数据量减小了数十万倍,极大提升了身份验证的速度和实时性。
本发明公开的人脸识别身份验证方法使在景区、车站、机场、交通路口、广场等公共场所中远距离、快速进行人脸识别身份验证的应用得以实施和推广。
本发明实施例公开的人脸识别身份验证方法,通过获取目标检测区域内对象群体的通信设备标识;根据通信设备标识和预设人脸库,确定待识别目标群体数据库;获取目标检测区域内待验证目标对象的人脸图像;基于待识别目标群体数据库对所述人脸图像进行比对识别,确定待验证目标对象的身份信息,解决了现有技术中在海量人脸库中基于人脸识别进行身份验证时效率低下的问题。
本发明实施例公开的人脸识别身份验证方法,通过获取的目标检测区域内对象群体的通信设备标识,缩小人脸比对范围,进一步基于缩小后的人脸库进行人脸比对,以确定目标对象的身份信息,运算量小,有助于提高基于人脸识别的身份验证的效率。同时,通过手机号码缩小待比对的人脸库,然后通过人脸信息进行身份验证,可以进一步提升身份验证的准确性。
实施例三
本实施例公开的一种人脸识别身份验证装置,如图3所示,上述装置包括:
通信设备标识获取模块310,用于获取目标检测区域内对象群体的通信设备标识;
待识别目标群体数据库确定模块320,用于根据通信设备标识和预设人脸库,确定待识别目标群体数据库;
目标对象人脸图像获取模块330,用于获取目标检测区域内待验证目标对象的人脸图像;
身份信息确定模块340,用于基于待识别目标群体数据库确定模块320确定的待识别目标群体数据库对目标对象人脸图像获取模块330获取的人脸图像进行比对识别,确定待验证目标对象的身份信息。
可选的,如图4所示,预设人脸库中包括至少由通信设备标识、身份 信息和人脸信息组成的对应关系,根据通信设备标识和预设人脸库,确定待识别目标群体数据库时,待识别目标群体数据库确定模块320包括:
通信设备标识比对子模块3201,用于通过将获取的所述通信设备的识和所述对应关系中的所述通信设备标识进行比对,确定所述对象群体的通信设备标识对应的身份信息和人脸信息;
数据库构建子模块3202,用于根据所述对象群体的通信设备标识对应的身份信息和人脸信息构建待识别目标群体数据库。
可选的,如图4所示,所述人脸信息至少包括人脸图像,基于待识别目标群体数据库确定模块320确定的待识别目标群体数据库对目标对象人脸图像获取模块330获取的人脸图像进行比对识别,确定所述待验证目标对象的身份信息时,身份信息确定模块340包括:
第一身份验证子模块3401,用于将待验证目标对象的人脸图像与待识别目标群体数据库中的人脸图像进行比对,确定比对成功的待识别目标群体数据库中的人脸图像对应的身份信息;
第一身份信息确定子模块3402,用于将确定的所述身份信息,作为待验证目标对象的身份信息。
可选的,如图4所示,人脸信息至少包括人脸特征,基于待识别目标群体数据库确定模块320确定的待识别目标群体数据库对目标对象人脸图像获取模块330获取的人脸图像进行比对识别,确定待验证目标对象的身份信息时,身份信息确定模块340包括:
第二身份验证子模块3403,用于获取待验证目标对象的人脸图像中的人脸特征;
第二身份验证子模块3403,还用于将待验证目标对象的人脸图像中的人脸特征与待识别目标群体数据库中的人脸特征进行比对,确定比对成功的待识别目标群体数据库中的人脸特征对应的身份信息;
第二身份信息确定子模块3404,用于将确定的所述身份信息,作为待验证目标对象的身份信息。
可选的,如图4所示,人脸信息至少包括人脸图像,基于待识别目标群体数据库确定模块320确定的待识别目标群体数据库对目标对象人脸图像获取模块330获取的人脸图像进行比对识别,确定待验证目标对象的身 份信息时,身份信息确定模块340包括:
第三身份验证子模块3405,用于从待验证目标对象的人脸图像中提取待验证人脸特征,以及,从待识别目标群体数据库中的人脸图像中提取参考人脸特征;
所述第三身份验证子模块3405,还用于将待验证人脸特征与参考人脸特征进行比对,确定比对成功的参考人脸特征对应的身份信息,其中,参考人脸特征对应的身份信息为从中提取参考人脸特征的待识别目标群体数据库中的人脸图像对应的身份信息;
第三身份信息确定子模块3406,用于将确定的所述身份信息,作为待验证目标对象的身份信息。可选的,所述通信设备标识所属通信设备包括移动终端,通信设备标识包括移动终端对应的唯一标识,所述通信设备标识获取模块310进一步用于:
通过非接触通信技术,检测目标检测区域内对象群体中持有的通信设备;
通过模拟通信基站的方式与所述通信设备进行通信,确定所述通信设备的通信设备标识。其中,移动终端可以为手机、电话手表。当移动终端为手机时,通信设备标识可以为手机号码。
本实施例公开的人脸识别身份验证装置用于实现实施例一所述的人脸识别身份验证方法,所述装置的各个模块的具体实施方式参见实施例一中相应步骤的具体实施方式,本实施例不再赘述。
本发明实施例公开的人脸识别身份验证装置,通过获取目标检测区域内对象群体的通信设备标识,并根据通信设备标识和预设人脸库,确定待识别目标群体数据库获取目标检测区域内待验证目标对象的人脸图像,以及基于待识别目标群体数据库对获得的人脸图像进行比对识别,确定待验证目标对象的身份信息,解决了现有技术中在海量人脸库中基于人脸识别进行身份验证时效率低下的问题。
本发明实施例公开的人脸识别身份验证装置,通过获取的目标检测区域内对象群体的通信设备标识,缩小人脸比对范围,进一步基于缩小后的人脸库进行人脸比对,以确定目标对象的身份信息,运算量小,有助于提高基于人脸识别的身份验证的效率。同时,通过手机号码缩小待比对的人 脸库,然后通过人脸信息进行身份验证,可以进一步提升身份验证的准确性。
相应的,本发明还公开了一种电子设备,包括存储器、处理器及存储在所述存储器上并可在处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现如本发明实施例一和实施例二所述的人脸识别身份验证方法。所述电子设备可以为PC机、移动终端、个人数字助理、平板电脑等。
本发明还公开了一种计算机可读存储介质,其上存储有计算机程序,该程序被处理器执行时实现如本发明实施例一和实施例二所述的人脸识别身份验证方法的步骤。
本说明书中的各个实施例均采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似的部分互相参见即可。对于装置实施例而言,由于其与方法实施例基本相似,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
以上对本发明提供的一种人脸识别身份验证方法及装置进行了详细介绍,本文中应用了具体个例对本发明的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本发明的方法及其核心思想;同时,对于本领域的一般技术人员,依据本发明的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本发明的限制。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件实现。基于这样的理解,上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品可以存储在计算机可读存储介质中,如ROM/RAM、磁碟、光盘等,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
Claims (17)
- 一种人脸识别身份验证方法,其特征在于,包括:获取目标检测区域内对象群体的通信设备标识;根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库;获取所述目标检测区域内待验证目标对象的人脸图像;基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息。
- 根据权利要求1所述的方法,其特征在于,所述预设人脸库中包括至少由通信设备标识、身份信息和人脸信息组成的对应关系。
- 根据权利要求2所述的方法,其特征在于,所述根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库的步骤,包括:通过将获取的所述通信设备标识和所述对应关系中的所述通信设备标识进行比对,确定所述对象群体的通信设备标识对应的身份信息和人脸信息;根据所述对象群体的通信设备标识对应的身份信息和人脸信息构建待识别目标群体数据库。
- 根据权利要求3所述的方法,其特征在于,所述人脸信息至少包括人脸图像,所述基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息的步骤,包括:将所述待验证目标对象的人脸图像与所述待识别目标群体数据库中的人脸图像进行比对,确定比对成功的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;将确定的所述身份信息,作为所述待验证目标对象的身份信息。
- 根据权利要求3所述的方法,其特征在于,所述人脸信息至少包括人脸图像,所述基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息的步骤,包括:从所述待验证目标对象的人脸图像中提取待验证人脸特征,以及,从所述待识别目标群体数据库中的人脸图像中提取参考人脸特征;将所述待验证人脸特征与所述参考人脸特征进行比对,确定比对成功 的所述参考人脸特征对应的所述身份信息,其中,所述参考人脸特征对应的所述身份信息为从中提取所述参考人脸特征的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;将确定的所述身份信息,作为所述待验证目标对象的身份信息。
- 根据权利要求3所述的方法,其特征在于,所述人脸信息至少包括人脸特征,所述基于所述待识别目标群体数据库对所述人脸图像进行比对识别,确定所述待验证目标对象的身份信息的步骤,包括:获取所述待验证目标对象的人脸图像中的人脸特征;将所述待验证目标对象的人脸图像中的人脸特征与所述待识别目标群体数据库中的人脸特征进行比对,确定比对成功的所述待识别目标群体数据库中的人脸特征对应的所述身份信息;将确定的所述身份信息,作为所述待验证目标对象的身份信息。
- 根据权利要求1至6任一项所述的方法,其特征在于,所述通信设备标识所属通信设备包括移动终端,所述通信设备标识包括移动终端对应的唯一标识。
- 根据权利要求7所述的方法,其特征在于,所述获取目标检测区域内对象群体的通信设备标识的步骤,包括:通过非接触通信技术,检测目标检测区域内对象群体中持有的通信设备;通过模拟通信基站的方式与所述通信设备进行通信,确定所述通信设备的通信设备标识。
- 根据权利要求1至6任一项所述的方法,其特征在于,所述获取目标检测区域内对象群体的通信设备标识的步骤之前,还包括:获取预设人脸库。
- 一种人脸识别身份验证装置,其特征在于,包括:通信设备标识获取模块,用于获取目标检测区域内对象群体的通信设备标识;待识别目标群体数据库确定模块,用于根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库;目标对象人脸图像获取模块,用于获取所述目标检测区域内待验证目 标对象的人脸图像;身份信息确定模块,用于基于所述待识别目标群体数据库确定模块确定的待识别目标群体数据库对所述目标对象人脸图像获取模块获取的人脸图像进行比对识别,确定所述待验证目标对象的身份信息。
- 根据权利要求10所述的装置,其特征在于,所述预设人脸库中包括至少由通信设备标识、身份信息和人脸信息组成的对应关系,根据所述通信设备标识和预设人脸库,确定待识别目标群体数据库时,所述待识别目标群体数据库确定模块包括:通信设备标识比对子模块,用于通过将获取的所述通信设备标识和所述对应关系中的所述通信设备标识进行比对,确定所述对象群体的通信设备标识对应的身份信息和人脸信息;数据库构建子模块,用于根据所述对象群体的通信设备标识对应的身份信息和人脸信息构建待识别目标群体数据库。
- 根据权利要求11所述的装置,其特征在于,所述人脸信息至少包括人脸图像,基于所述待识别目标群体数据库确定模块确定的待识别目标群体数据库对所述目标对象人脸图像获取模块获取的人脸图像进行比对识别,确定所述待验证目标对象的身份信息时,所述身份信息确定模块包括:第一身份验证子模块,用于将所述待验证目标对象的人脸图像与所述待识别目标群体数据库中的人脸图像进行比对,确定比对成功的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;第一身份信息确定子模块,用于将确定的所述身份信息,作为所述待验证目标对象的身份信息。
- 根据权利要求11所述的装置,其特征在于,所述人脸信息至少包括人脸图像,基于所述待识别目标群体数据库确定模块确定的待识别目标群体数据库对所述目标对象人脸图像获取模块获取的人脸图像进行比对识别,确定所述待验证目标对象的身份信息时,所述身份信息确定模块包括:第三身份信息确定子模块,用于从所述待验证目标对象的人脸图像中提取待验证人脸特征,以及,从所述待识别目标群体数据库中的人脸图像 中提取参考人脸特征;所述第三身份信息确定子模块,还用于将所述待验证人脸特征与所述参考人脸特征进行比对,确定比对成功的所述参考人脸特征对应的所述身份信息,其中,所述参考人脸特征对应的所述身份信息为从中提取所述参考人脸特征的所述待识别目标群体数据库中的人脸图像对应的所述身份信息;第三身份信息确定子模块,用于将确定的所述身份信息,作为所述待验证目标对象的身份信息。
- 根据权利要求11所述的装置,其特征在于,所述人脸信息至少包括人脸特征,基于所述待识别目标群体数据库确定模块确定的待识别目标群体数据库对所述目标对象人脸图像获取模块获取的人脸图像进行比对识别,确定所述待验证目标对象的身份信息时,所述身份信息确定模块包括:第二身份验证子模块,用于获取所述待验证目标对象的人脸图像中的人脸特征;所述第二身份验证子模块,还用于将所述待验证目标对象的人脸图像中的人脸特征与所述待识别目标群体数据库中的人脸特征进行比对,确定比对成功的所述待识别目标群体数据库中的人脸特征对应的所述身份信息;第二身份信息确定子模块,用于将确定的所述身份信息,作为所述待验证目标对象的身份信息。
- 根据权利要求10至14任一项所述的装置,其特征在于,所述通信设备标识所属通信设备包括移动终端,所述通信设备标识包括移动终端的唯一标识,获取目标检测区域内对象群体的通信设备标识时,所述通信设备标识获取模块用于:通过非接触通信技术,检测目标检测区域内对象群体中持有的通信设备;通过模拟通信基站的方式与所述通信设备进行通信,确定所述通信设备的通信设备标识。
- 一种电子设备,包括存储器、处理器及存储在所述存储器上并可 在处理器上运行的计算机程序,其特征在于,所述处理器执行所述计算机程序时实现权利要求1至9任意一项所述的人脸识别身份验证方法。
- 一种计算机可读存储介质,其上存储有计算机程序,其特征在于,该程序被处理器执行时实现权利要求1至9任意一项所述的人脸识别身份验证方法的步骤。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810711952.2A CN109086669B (zh) | 2018-06-29 | 2018-06-29 | 人脸识别身份验证方法、装置、电子设备 |
CN201810711952.2 | 2018-06-29 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020000906A1 true WO2020000906A1 (zh) | 2020-01-02 |
Family
ID=64837044
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2018/119685 WO2020000906A1 (zh) | 2018-06-29 | 2018-12-07 | 人脸识别身份验证方法、装置、电子设备 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109086669B (zh) |
WO (1) | WO2020000906A1 (zh) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111652173A (zh) * | 2020-06-10 | 2020-09-11 | 杭州十域科技有限公司 | 一种适合综合商场内人流管控的采集方法 |
CN112052733A (zh) * | 2020-07-31 | 2020-12-08 | 中国建设银行股份有限公司 | 数据库构建方法、人脸识别方法、装置及电子设备 |
CN114724261A (zh) * | 2022-04-15 | 2022-07-08 | 澜途集思生态科技集团有限公司 | 基于snip算法的生态生物识别方法 |
US11630021B2 (en) | 2011-08-12 | 2023-04-18 | Mueller International, Llc | Enclosure for leak detector |
Families Citing this family (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111625793B (zh) * | 2019-02-27 | 2023-06-23 | 阿里巴巴集团控股有限公司 | 身份识别、订单支付、子人脸库建立方法、装置及设备、订单支付系统 |
CN109919093A (zh) * | 2019-03-07 | 2019-06-21 | 苏州科达科技股份有限公司 | 一种人脸识别方法、装置、设备及可读存储介质 |
CN111915305B (zh) * | 2019-05-10 | 2023-07-25 | 腾讯科技(深圳)有限公司 | 支付方法、装置、设备以及存储介质 |
CN112233299A (zh) * | 2019-06-27 | 2021-01-15 | 西安光启未来技术研究院 | 一种人像动态权限认证方法及系统 |
CN112395904A (zh) * | 2019-08-12 | 2021-02-23 | 北京蜂盒科技有限公司 | 一种生物特征识别方法及系统 |
CN110659580A (zh) * | 2019-08-29 | 2020-01-07 | 深圳市新系区块链技术有限公司 | 一种快速认证身份信息的方法及装置 |
CN110991390B (zh) * | 2019-12-16 | 2023-04-07 | 腾讯云计算(北京)有限责任公司 | 身份信息检索方法、装置、业务系统及电子设备 |
CN113076533B (zh) * | 2020-01-03 | 2023-09-05 | 中国移动通信集团广东有限公司 | 一种业务处理方法及装置 |
CN113554046A (zh) * | 2020-04-24 | 2021-10-26 | 阿里巴巴集团控股有限公司 | 图像处理方法和系统、存储介质及计算设备 |
CN111598047B (zh) * | 2020-05-28 | 2023-06-27 | 重庆康普达科技有限公司 | 一种人脸识别方法 |
CN111695538A (zh) * | 2020-06-17 | 2020-09-22 | 杭州海康威视系统技术有限公司 | 非法入侵人员检测方法、装置及可读存储介质 |
CN111898451A (zh) * | 2020-07-01 | 2020-11-06 | 中国建设银行股份有限公司 | 身份识别方法、装置、电子设备及计算机可读存储介质 |
CN114267058B (zh) * | 2020-09-16 | 2024-07-02 | 腾讯科技(深圳)有限公司 | 人脸识别方法、装置、计算机设备及存储介质 |
CN112218046B (zh) * | 2020-09-27 | 2023-10-24 | 杭州海康威视系统技术有限公司 | 对象监控方法及装置 |
CN111931153B (zh) * | 2020-10-16 | 2021-02-19 | 腾讯科技(深圳)有限公司 | 基于人工智能的身份验证方法、装置和计算机设备 |
CN114495188B (zh) * | 2020-10-26 | 2024-09-20 | 腾讯科技(深圳)有限公司 | 一种图像数据处理方法、装置和相关设备 |
CN112541172A (zh) * | 2020-11-27 | 2021-03-23 | 浪潮电子信息产业股份有限公司 | 一种计算机设备开机方法及相关装置 |
CN112907805A (zh) * | 2021-01-18 | 2021-06-04 | 广州云从鼎望科技有限公司 | 基于人脸识别的过闸控制方法及系统、介质、控制装置 |
CN113128437A (zh) * | 2021-04-27 | 2021-07-16 | 北京市商汤科技开发有限公司 | 身份识别方法及装置、电子设备和存储介质 |
CN113688712A (zh) * | 2021-08-18 | 2021-11-23 | 上海浦东发展银行股份有限公司 | 一种人像识别方法、装置、电子设备及存储介质 |
CN114648336A (zh) * | 2022-03-18 | 2022-06-21 | 成都商汤科技有限公司 | 人脸支付方法、装置、电子设备及存储介质 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103886658A (zh) * | 2014-01-28 | 2014-06-25 | 北京中科虹霸科技有限公司 | 基于多模态生物识别的分布式物联网锁设备及其开锁方法 |
CN105160739A (zh) * | 2015-09-09 | 2015-12-16 | 北京旷视科技有限公司 | 自动识别设备、方法以及门禁系统 |
CN105654584A (zh) * | 2015-12-24 | 2016-06-08 | 歌尔声学股份有限公司 | 一种门禁监控方法和系统 |
CN206179048U (zh) * | 2016-11-09 | 2017-05-17 | 中国联合网络通信集团有限公司 | 身份识别装置 |
-
2018
- 2018-06-29 CN CN201810711952.2A patent/CN109086669B/zh active Active
- 2018-12-07 WO PCT/CN2018/119685 patent/WO2020000906A1/zh active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103886658A (zh) * | 2014-01-28 | 2014-06-25 | 北京中科虹霸科技有限公司 | 基于多模态生物识别的分布式物联网锁设备及其开锁方法 |
CN105160739A (zh) * | 2015-09-09 | 2015-12-16 | 北京旷视科技有限公司 | 自动识别设备、方法以及门禁系统 |
CN105654584A (zh) * | 2015-12-24 | 2016-06-08 | 歌尔声学股份有限公司 | 一种门禁监控方法和系统 |
CN206179048U (zh) * | 2016-11-09 | 2017-05-17 | 中国联合网络通信集团有限公司 | 身份识别装置 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11630021B2 (en) | 2011-08-12 | 2023-04-18 | Mueller International, Llc | Enclosure for leak detector |
CN111652173A (zh) * | 2020-06-10 | 2020-09-11 | 杭州十域科技有限公司 | 一种适合综合商场内人流管控的采集方法 |
CN111652173B (zh) * | 2020-06-10 | 2023-07-18 | 杭州十域科技有限公司 | 一种适合综合商场内人流管控的采集方法 |
CN112052733A (zh) * | 2020-07-31 | 2020-12-08 | 中国建设银行股份有限公司 | 数据库构建方法、人脸识别方法、装置及电子设备 |
CN114724261A (zh) * | 2022-04-15 | 2022-07-08 | 澜途集思生态科技集团有限公司 | 基于snip算法的生态生物识别方法 |
Also Published As
Publication number | Publication date |
---|---|
CN109086669A (zh) | 2018-12-25 |
CN109086669B (zh) | 2020-06-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020000906A1 (zh) | 人脸识别身份验证方法、装置、电子设备 | |
TWI752418B (zh) | 伺服器、客戶端、用戶核身方法及系統 | |
CN103390153B (zh) | 用于生物特征验证的纹理特征的方法和系统 | |
CN102646190A (zh) | 一种基于生物特征的认证方法、装置及系统 | |
CN109376604B (zh) | 一种基于人体姿态的年龄识别方法和装置 | |
CN107169458B (zh) | 数据处理方法、装置及存储介质 | |
WO2018166291A1 (zh) | 一种基于多因素交叉验证的用户签到身份识别方法 | |
JP7213596B2 (ja) | 動的ラスタライズ管理に基づく身元識別方法、装置及びサーバ | |
WO2018121006A1 (zh) | 一种车牌定位方法及装置 | |
CN110399763B (zh) | 人脸识别方法与系统 | |
KR20210101313A (ko) | 얼굴 인식 방법, 신경망 훈련방법, 장치 및 전자 디바이스 | |
CN104348778A (zh) | 一种在手机端进行人脸初步验证的远程身份认证的系统、终端和方法 | |
CN105117624A (zh) | 一种基于人脸识别的用户体系识别系统及方法 | |
Stojmenovic | Mobile cloud computing for biometric applications | |
CN103425972A (zh) | 身份证件持有人真伪自动识别方法和系统 | |
CN103761466A (zh) | 一种身份验证的方法及装置 | |
CN103440482A (zh) | 基于隐蔽视频的身份证件持有人识别方法、系统和装置 | |
CN105575008A (zh) | 一种智能寄存柜 | |
CN113536261A (zh) | 权限管理方法、装置、计算机装置及存储介质 | |
CN113591603B (zh) | 证件的验证方法、装置、电子设备及存储介质 | |
WO2018082232A1 (zh) | 一种签名的识别方法、装置、移动终端及云端服务器 | |
CN110929108A (zh) | 一种用于铁路客运服务的询问方法、系统及机器人 | |
CN102930193A (zh) | 一种基于人脸识别的社区登陆实现方法 | |
CN110852193A (zh) | 一种人脸识别的方法和装置 | |
CN106710019A (zh) | 数据处理方法、装置及电子设备 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18923789 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18923789 Country of ref document: EP Kind code of ref document: A1 |