WO2018068261A1 - 一种人脸登记方法及人脸登记设备 - Google Patents
一种人脸登记方法及人脸登记设备 Download PDFInfo
- Publication number
- WO2018068261A1 WO2018068261A1 PCT/CN2016/101996 CN2016101996W WO2018068261A1 WO 2018068261 A1 WO2018068261 A1 WO 2018068261A1 CN 2016101996 W CN2016101996 W CN 2016101996W WO 2018068261 A1 WO2018068261 A1 WO 2018068261A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- face
- registration device
- user
- face registration
- location
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- 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
- G06V40/169—Holistic features and representations, i.e. based on the facial image taken as a whole
Definitions
- the present invention relates to the field of communications, and in particular, to a face registration method and a face registration device.
- the cameras used for the collection are angularly required, so that the identification device is mounted on the wall at a fixed angle during installation so that the fixed surface is fixed. Face images from 150 cm to 190 cm in height can be obtained from the camera's range of angles. Therefore, in the process of using the face recognition device, generally, by prompting the user to perform operations such as up, down, left, and right when registering, a variety of face templates are obtained, thereby improving the verification pass rate.
- the current face registration device when performing face registration, prompts the user to perform the up, down, left, and right turns, and obtains a variety of templates by the user's up, down, left, and right.
- the embodiment of the invention provides a face registration method and a face registration device, which are used for obtaining a more perfect face template, which is advantageous for less restricting posture during comparison and improving the success rate of verification.
- a first aspect of the embodiments of the present invention provides a method for registering a face, including:
- the face registration device receives a registration message, the registration message instructing the face registration device to perform face registration;
- the face registration device detects whether there is a user's face in a preset time, and if present, the face registration device collects a first facial feature when the face is in the first position, the first a location is a location where the face is located when the face registration device detects the face of the user;
- the face registration device generates first face template data according to the first facial feature
- the face registration device stores the first face template data in a storage device
- the face registration device obtains second face template data according to the second facial feature
- the face registration device stores the second face template data in a storage device.
- a second aspect of the embodiments of the present invention provides a face registration device, which specifically includes:
- a receiving module configured to receive a registration message, where the registration message instructs the face registration device to perform face registration
- a first detecting module configured to detect whether a user's face exists in a preset time
- a first collecting module configured to: when the face of the user exists, collect a first facial feature when the face is in a first position, where the first location is that the face registration device detects the user The position of the face when the face is;
- a first generating module configured to generate first face template data according to the first facial feature
- a first storage module configured to store the first face template data in a storage device
- a first determining module configured to determine whether a rotation angle of the second position after the rotation of the face is relative to the first position reaches a preset value
- a second collecting module configured to collect, when the preset value is reached, a second facial feature when the face is in the second position
- a second generating module configured to generate second face template data according to the second facial feature
- the second storage module is configured to store the second face template data in the storage device.
- a third aspect of the embodiments of the present invention provides a face registration device, which specifically includes:
- a central processing unit a memory, a storage medium, a power supply, a wireless network interface, and an input/output interface
- the central processor is configured to perform the following steps by invoking an operation instruction stored on the memory or a storage medium:
- the face registration device collects a first facial feature when the face is in a first position, the first location being the face Registration The location where the face is located when the face of the user is detected;
- the second face template data is stored in the storage device.
- the face registration device collects the facial features of the user's first location, and generates the first facial template data according to the facial features of the first location, and collects the user's first
- the facial feature of the second position after the position is rotated, and the second face template data is generated according to the facial features of the second position, and the first face template data and the second face template data are respectively stored in the storage device. Since the facial features of the user's face at the current location are collected and the facial features of the user's face that have reached the preset value are obtained, a more perfect face template is obtained, which is advantageous for the restriction on the posture when the comparison is performed, and the verification is passed. Success rate.
- FIG. 1 is a schematic diagram of an embodiment of a face registration method according to an embodiment of the present invention.
- FIG. 2 is a schematic diagram of another embodiment of a method for registering a face according to an embodiment of the present invention
- FIG. 3 is a schematic structural diagram of a face registration device according to an embodiment of the present invention.
- FIG. 4 is another schematic structural diagram of a face registration device according to an embodiment of the present invention.
- FIG. 5 is a schematic structural diagram of a face registration device according to an embodiment of the present invention.
- an embodiment of a face registration method in an embodiment of the present invention includes:
- the face registration device receives the registration message.
- the face registration device may receive a registration message indicating that the face registration device performs face registration.
- the face registration device detects whether there is a user's face in a preset time. If yes, step 103 is performed, and if not, step 104 is performed.
- step 103 is performed. If the user's face is not detected within the preset time, step 104 is performed.
- the face registration device collects a first facial feature when the face is in the first position.
- the face registration device when the face registration device detects that the face of the user exists in the preset time, the face registration device collects the facial feature when the face is in the first position, and the first location is detected by the face registration device. The location of the face when the user's face is reached.
- the face registration device generates first face template data according to the first facial feature.
- the face registration device may generate the first face template data according to the first facial feature.
- the face registration device stores the first face template data in the storage device.
- the face registration device may store the first face template data in the storage device.
- the face registration device determines whether the rotation angle of the second position after the rotation of the face is relative to the first position reaches a preset value. If yes, step 108 is performed. If not, step 109 is performed.
- the face registration device can determine whether the rotation angle of the second position after the face is rotated relative to the first position reaches a pre-preparation. Set the value. If yes, go to step 108. If not, go to step 109.
- the face registration device collects a second facial feature when the face is in the second position.
- the face registration device determines that the rotation angle of the second position after the face rotation is relative to the first position reaches a preset value
- the face registration device collects the face of the face when the face is in the second position.
- the face registration device performs other operations.
- the face registration device when the face registration device determines that the rotation angle of the second position after the face rotation is less than the preset value, the face registration device performs other operations.
- the face registration device generates second face template data according to the second facial feature.
- the second facial template data may be generated according to the second facial feature.
- the face registration device stores the second face template data in the storage device.
- the second face template data may be stored in the storage device.
- the face registration device may store the first face template data in the storage device by using step 106, and generate the second face template data through step 107 to step 111 and store the data in the storage device, but step 106 and The step of the step 107 to the step 111 is not performed.
- the step 106 may be performed first, or the step 107 to the step 111 may be performed first, or may be performed at the same time, which is not limited herein.
- the face registration device may store the first face template data in the storage device through step 106, and the second face template data may be stored in the storage device through step 111, and the face registration device is The two face template data is stored in the storage device in association with the first face template data.
- the face registration device collects the facial features of the user's first location, and generates the first facial template data according to the facial features of the first location, and collects the second location after the user's first position is rotated.
- the facial feature generates second face template data according to the facial features of the second location, and the face registration device stores the first face template data and the second face template data in the storage device, respectively.
- the face registration device separately collects the face alignment of the user face face registration device and the facial features of different rotation angles, elevation angles or depression angles by judging different rotation angles, elevation angles or depression angles in the registration process.
- the feature of the entire face can be obtained more comprehensively, and a more perfect face template can be obtained, which is advantageous for the restriction of the posture when the comparison is performed, and the success rate of the verification pass is improved.
- an embodiment of the face registration method in the embodiment of the present invention includes:
- the face registration device when the user needs to perform face registration, the face registration device receives a registration message indicating that the face registration device performs face registration. At this time, the face registration device displays the face registration interface.
- the face registration device detects whether a face exists in a preset time, and if yes, executes Step 203, if not, step 204 is performed.
- the face registration device After the face registration device receives the registration message, the face registration device turns on the infrared and the camera, and the camera is in a horizontal position with the face of the user, and detects whether there is a face in a preset time. If yes, step 203 is performed, and if not, step 204 is performed.
- the face registration device detects whether there is a face in a preset time, and the face may be a camera facing the face registration device, or may be a side opposite to the camera, or may face the face.
- the location of the device is not limited. The following is an example in which the face registration device first detects that the face is frontally aligned with the camera.
- the face registration device when the face registration device detects that there is a face front alignment camera within a preset time, the face registration device can collect facial features of the face of the face and generate a frontal face according to the facial features of the face. Template data, the face registration device saves the generated front face template data in the storage device.
- the face registration device when the face registration device does not detect the presence of a face within a preset time, the face registration device exits the registration interface.
- the face registration device determines whether the user face movement track deviates from the registration range. If not, perform step 206. If the offset is performed, perform step 207.
- the face registration device may determine whether the face movement trajectory of the user who needs to be registered deviates from the registration range. If not, perform step 206. If the offset is performed, perform step 207.
- the face registration device detects the facial trajectory of the user in real time. If the trajectory of the face deviates from the registration range during the registration process, the face registration device exits the registration interface.
- the face registration device collects facial features on each side of the user's face, it may first determine whether the face movement trajectory of the user is offset from the registration range, and when it is determined that the registration range is not offset, the face registration device starts.
- the facial feature collection has been described here. For the sake of brevity, the detailed description will not be made below before the facial features of each side of the user's face are collected.
- the face registration device issues a prompt message, prompting the user who needs to register the facial feature to rotate to the left.
- a prompt message may be issued to prompt the user to rotate to the left.
- the face registration device is logged out.
- the registration interface may be exited.
- the face registration device determines whether the angle of rotation of the face to the left reaches a preset value. If yes, step 209 is performed, and if not, step 210 is performed.
- the user may rotate according to the prompt.
- the face registration device may determine whether the rotation angle of the user's face to the left is a preset value. If yes, execute step 209, if not If yes, step 210 is performed.
- the face registration device collects the facial features and generates the facial template data and saves the data to the storage device.
- the face registration device determines that the angle of the face rotation to the left reaches a preset value, and the face registration device collects the facial feature of the face when the angle of the face rotation to the left reaches a preset value, and according to the The face facial feature generates left face template data, and the face registration device saves the generated left face template data to the storage device.
- the face registration device issues a prompt to indicate that the user rotates to the left until the angle of rotation to the left reaches a preset value.
- the face registration device when the face registration device determines that the angle of the face rotation to the left does not reach the preset value, the face registration device sends a prompt message to prompt the user to rotate to the left until the user's face rotates to the left.
- the preset value is reached.
- the face registration device prompts the user to turn to the right.
- the face registration device may issue a prompt message prompting the user to rotate to the right.
- the face registration device determines whether the angle of the user's rotation to the right reaches a preset value. If yes, step 213 is performed. If not, step 214 is performed.
- the user may perform a right turn, and the face registration device determines whether the rotation angle of the user's face to the right is a preset value. If yes, execute step 213, if not, execute Step 214.
- the face registration device collects the facial features and generates the face template data and saves the data to the storage device. in.
- the face registration device determines that the angle of the user's face is rotated to the right reaches a preset value
- the facial feature of the face rotated to the right may be collected, and the right face template is generated according to the facial feature.
- the face registration device stores the right face template data in the storage device.
- the face registration device prompts the user to turn right until the right turn angle reaches a preset value.
- the face registration device when the face registration device determines that the angle of the user's rotation to the right does not reach the preset value, the face registration device sends a prompt indicating that the user continues to rotate until the angle of rotation to the right reaches a preset value.
- the face registration device prompts the user's face to rotate downward.
- the face registration device prompts the user to rotate downward, that is, the face registration device prompts the user to bow.
- the face registration device determines whether the angle of the face rotation of the user reaches a preset value. If yes, step 217 is performed. If not, step 218 is performed.
- the face registration device determines whether the angle of the downward rotation reaches a preset value. If yes, step 217 is performed, and if not, step 218 is performed. .
- the face registration device collects facial features when the head is lowered and generates face template data and saves the data to the storage device.
- the face registration device determines that the angle at which the face of the user is rotated downward reaches a preset value
- the face registration device collects the facial feature when the downward rotation angle of the face of the user reaches a preset value, and according to the direction
- the face feature whose bottom rotation angle reaches the preset value generates the face face template data
- the face registration device saves the face template data at the time of the head down to the storage device.
- the face registration device prompts the user's face to rotate downward until the angle of downward rotation reaches a preset value.
- the prompt information is sent to instruct the user to perform downward rotation until the angle at which the user's face rotates downward reaches a preset value.
- the face registration device prompts the user to control the face to rotate upward.
- the face registration device issues a prompt message instructing the user to control the face to rotate upward.
- the face registration device determines whether the angle of the user's face is rotated upward reaches a preset value. If yes, step 221 is performed. If not, step 222 is performed.
- the user controls the face to rotate upward according to the prompt information, and the face registration device determines whether the angle of the user's face is up to a preset value. If yes, step 221 is performed, and if not, step 222 is performed.
- the face registration device collects the face up feature and generates the face template data and saves the data to the storage device.
- the face registration device determines that the angle of the user's face up is up to a preset value
- the facial feature whose upper elevation angle reaches the preset value is collected, and the facial feature is generated according to the facial feature whose upper elevation angle reaches the preset value.
- the face template data the face registration device saves the face template data at the time of the head to the storage device.
- the face registration device prompts the user's face to rotate upward until the upside angle reaches a preset value.
- a prompt is sent to instruct the user to continue to rotate until the angle of the upward tilt reaches a preset value.
- prompt information of the above-mentioned storytelling may be voice, or the text information may be displayed on the registration interface, or may be video information, which is not limited herein.
- the face registration device collects the face features of the face and generates the face template data by step 203, and collects the face features of the face when the face is turned left by step 209 and generates the face template data of the left face.
- Step 213 collecting facial features when the face is turned right and generating the right face template data, and collecting the face facial features when the face is lowered and generating the face template data by step 217, and collecting the face face when the face is raised by step 221
- the feature is generated and the face-to-face template data is generated.
- the step 203 may be performed first, or the step 209 may be performed first, or the step 217 may be performed first.
- Step 221 may also be performed first, which is not limited herein.
- the face registration device stores the front face template data, the left face template data, the right face template data, the face template data when looking up, and the face template data when looking down.
- the face template data can also be stored in the storage device when there are other face template data of the same user.
- each face template data generated by the face registration device is stored in the user ID.
- the face template data carrying the user's ID is obtained.
- the face registration device collects the facial features of the user when the user needs to verify the position relative to the face registration device. And generating the face template data at the time of verification according to the facial features of the verification, and comparing with the face template data carrying the user ID, when the face registration device determines the face template data and the user ID carrying the user ID.
- the data similarity of any face template data in the face template data reaches a preset threshold, it is determined that the user verification passes, and vice versa, it is determined that the user verification fails.
- the registration interface may be exited.
- the face registration device collects the front side of the face and the facial features of each side, and generates face template data corresponding to the collected front face and facial features of each side, and each person
- the face template data is stored in the storage space where the ID of the user is located, and the face template data carrying the user ID is obtained.
- the face registration device separately collects the face alignment of the user face face registration device and the facial features of different rotation angles, elevation angles or depression angles by judging different rotation angles, elevation angles or depression angles in the registration process. , can get a more complete face template, which is conducive to less restrictive posture when comparing, and improve the success rate of verification.
- User A needs to register the face, and user A first clicks on the face registration device to enter the face registration device.
- the face of user A is facing the camera of the face registration device, and the face registration device prompts the user A to "see” "Camera"
- the face registration device collects the facial features of the front side of the user A and generates front face template data according to the facial features of the front face, and the face registration device prompts the user A to turn left, turn right, look up and bow, and collect user A respectively.
- a series of facial features generate a series of face template data, and store the generated front face template data and the template data of the generated face feature generated by the rotated face in the storage space where the user ID is located.
- the face template data carrying the user ID is obtained, and the face registration interface is exited.
- the user can stand in front of the face registration device, and when the face registration device detects the presence of the face, the current facial feature is collected and the face template data is generated. And comparing with the face template data in the storage device, the data similarity between the currently generated face template data and any personal face template data in the storage device reaches a preset threshold, and the user's verification is passed.
- an embodiment of the present invention further provides a face registration device, which specifically includes:
- the receiving module 301 is configured to receive a registration message, where the registration message indicates that the face registration device performs face registration;
- the first detecting module 302 is configured to detect whether a user's face exists in a preset time
- the first collection module 303 is configured to collect a first facial feature when the face is in the first position when the face of the user exists, where the first location is a location where the face is located when the face registration device detects the face of the user ;
- a first generating module 304 configured to generate first face template data according to the first facial feature
- the first storage module 305 is configured to store the first face template data in the storage device
- the first determining module 306 is configured to determine whether a rotation angle of the second position after the face is rotated relative to the first position reaches a preset value
- the second collecting module 307 is configured to collect a second facial feature when the face is in the second position when the preset value is reached;
- a second generating module 308 configured to generate second face template data according to the second facial feature
- the second storage module 309 is configured to store the second face template data in the storage device.
- the face registration device including:
- the receiving module 401 is configured to receive a registration message, where the registration message instructs the face registration device to perform face registration;
- the first detecting module 402 is configured to detect whether a user's face exists in a preset time
- the first prompting module 403 is configured to send first prompt information, where the first prompt information indicates that the user performs face registration.
- the first collecting module 404 is configured to collect a first facial feature when the face is in the first position when the face of the user is present, where the first location is a location where the face is located when the face registration device detects the face of the user ;
- a first generating module 405, configured to generate first face template data according to the first facial feature
- the first storage module 406 is configured to store the first face template data in the storage device.
- a second detecting module 407 configured to detect a facial movement track of the user
- the second determining module 408 is configured to determine whether the facial movement trajectory deviates from the registration range
- a first determining module 409 configured to exit the face registration when determining that the facial movement track deviates from the registration range
- the second determining module 410 is configured to perform face registration in determining that the face movement trajectory does not deviate from the registration range.
- the second prompting module 411 is configured to send the second prompt information, and the second prompt information indicates that the user controls the face to rotate.
- the first determining module 412 is configured to determine whether a rotation angle of the second position after the face is rotated relative to the first position reaches a preset value
- the second collecting module 413 is configured to collect a second facial feature when the face is in the second position when the preset value is reached;
- a second generating module 414 configured to generate second face template data according to the second facial feature
- the second storage module 415 is configured to store the second face template data in the storage device
- the third collecting module 416 is configured to collect, when the face registration device detects the user's face within the detection range, the third facial feature that detects the user's face;
- a fourth generation module 417 configured to generate third face template data according to the third facial feature
- the third determining module 418 is configured to determine whether the data similarity between the third face template data and any template data in the first template data and the second face template data reaches a preset threshold;
- the third determining module 419 is configured to: when the data similarity between the third face template data and the first face template data and the second template data reaches a preset threshold, determine that the user is detected Face verification passed;
- the fourth determining module 420 is configured to determine that the user is detected when the data similarity between the third face template data and the first face template data and the second template data does not reach a preset threshold. Face verification failed.
- the first determining module 412 can further include:
- the first determining unit 4121 is configured to determine, when the first position is that the face of the user faces the position of the face registration device, whether the rotation angle of the position rotated by the face to the left is relative to the first position.
- a first determining unit 4122 configured to determine, when the rotation angle of the position after the face is rotated to the left relative to the first position, reaches a preset value, determining that the position of the face rotated to the left is the second position;
- the second determining unit 4123 is configured to determine, when the first position is the position of the face of the user facing the face registration device, whether the rotation angle of the position rotated by the face to the right is relative to the first position.
- a second determining unit 4124 configured to determine, when the rotation angle of the position after the face is rotated to the right relative to the first position reaches a preset value, the position after the face is rotated to the right is the second position;
- the third determining unit 4125 is configured to determine, when the first location is that the face of the user faces the location of the face registration device, determine whether the rotation angle of the face after the upward rotation of the face is relative to the first location Set value
- the third determining unit 4126 is configured to determine that the position after the face is rotated upward is the second position when the rotation angle of the position after the face is rotated upward relative to the first position reaches a preset value;
- the fourth determining unit 4127 is configured to determine whether the rotation angle of the position after the face is rotated downward relative to the first position is reached when the first position is the position of the face of the user facing the face registration device.
- the fourth determining unit 4128 is configured to determine that the position after the face is rotated downward is the second position when the rotation angle of the position after the face is rotated downward relative to the first position reaches a preset value.
- the second detection module 407 detects the facial trajectory of the user in real time. If the facial motion trajectory of the face deviates from the registration range during the registration process, the face registration device exits the registration interface.
- the face registration device receives the registration message of the user that needs to be authenticated by the receiving module 401, and the first detection module 402 detects whether there is a human face within a preset time. If there is a human face, the first collection module 404 Collecting a first facial feature when the face is in the first position, the first generation module 405: The first face module data is generated according to the first facial feature, the first storage module 406 stores the first face template data in the storage device, and the second detecting module 407 detects whether the face deviates from the registration range in real time. The first judging module 412 determines whether the rotation angle of the second position after the rotation of the face is relative to the first position reaches a preset value.
- the second collection module 413 collects the second facial feature when the face is in the second position.
- the second generation module 413 generates the second face template data according to the second facial feature
- the second storage module 415 stores the second face template data and the first face template data in the storage device. It can be seen that the face registration device collects the facial features of the user's face as much as possible, and generates a series of template data carrying the user ID according to the facial features collected to reach the face, thereby obtaining a more perfect person.
- the face template helps to reduce the user's posture when comparing, and improves the success rate of verification.
- FIG. 5 is a schematic structural diagram of a face registration device according to an embodiment of the present invention.
- the face registration device 500 may have a large difference due to different configurations or performances, and may include one or more central nodes.
- Central processing units (CPU) 522 eg, one or more processors
- memory 32 e.g., one or more storage media 530 storing application 542 or data 544 (eg, one or one storage device in Shanghai).
- the memory 532 and the storage medium 530 may be short-term storage or persistent storage.
- the program stored on storage medium 530 may include one or more modules (not shown), each of which may include a series of instruction operations in the server.
- the central processing unit 522 can be configured to communicate with the storage medium 530 to perform a series of instruction operations in the storage medium 530 on the face registration device 500.
- the face registration device 500 can also include one or more power sources 526, one or more wired or wireless network interfaces 550, one or more input and output interfaces 558, and/or one or more operating systems 541, such as Windows ServerTM. , Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM and more.
- the steps performed by the face registration device in the above embodiment may be based on the face registration device structure shown in FIG.
- the disclosed system, device and method The law can be implemented in other ways.
- the device embodiments described above are merely illustrative.
- the division of the unit is only a logical function division.
- there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
- the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
- the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
- each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
- the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
- the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
- the technical solution of the present invention which is essential or contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product stored in a storage medium.
- a number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
- the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
本发明实施例公开了一种人脸登记方法及人脸登记设备,用于取到更加完善的人脸模板,有利于比对时对用户的姿势的限制更小,提高验证通过的成功率。本发明实施例方法包括:人脸登记设备接收登记消息,人脸登记设备在预置时间内检测是否存在用户的人脸,若存在,则人脸登记设备采集所述人脸处于第一位置时的第一面部特征;人脸登记设备根据第一面部特征生成第一人脸模板数据;人脸登记设备将第一人脸模板数据储存在储存设备中;人脸登记设备采集人脸旋转后的第二位置相对于第一位置的旋转角度达到预设值时的人脸的第二面部特征;人脸登记设备根据所述第二面部特征得到第二人脸模板数据;人脸登记设备将第二人脸模板数据储存在储存设备中。
Description
本发明涉及通信领域,尤其涉及一种人脸登记方法及人脸登记设备。
现有的人脸识别装置,为了防止采集图像的畸变,采集使用的摄像头都是有角度要求的,所以在安装时都是将识别装置以固定的角度安装在墙面上,以使得在固定的摄像头采集角度范围内可以获取到身高为150厘米至190厘米的人脸图像。所以在人脸识别设备的使用过程中,一般通过在登记时,提示用户进行上仰、低头、左转和右转等操作来获得多样化的人脸模板,从而提高验证通过率。
目前的人脸登记装置,在进行人脸登记时,会提示用户进行上仰、低头、左转和右转,通过用户的上仰、低头、左转和右转获取多样化的模板。
但是在实际应用中用户往往并没有配合到位,所以造成登记后的模板,往往没有多样化。在比对时不能合理使用。同时现有技术由于没有自动分析,在登记时,会存在串登记的行为,譬如,前面的用户进行登记时,突然离开,此时***并没有判断出用户已经离开,导致后面的用户在原先的登记界面上,继续登记。
发明内容
本发明实施例提供了一种人脸登记方法及人脸登记设备,用于取到更加完善的人脸模板,有利于比对时对姿势的限制更小,提高验证通过的成功率。
本发明实施例第一方面提供了一种人脸登记方法,具体包括:
人脸登记设备接收登记消息,所述登记消息指示所述人脸登记设备进行人脸登记;
所述人脸登记设备在预置时间内检测是否存在用户的人脸,若存在,则所述人脸登记设备采集所述人脸处于第一位置时的第一面部特征,所述第一位置为所述人脸登记设备检测到所述用户的人脸时所述人脸所处的位置;
所述人脸登记设备根据所述第一面部特征生成第一人脸模板数据;
所述人脸登记设备将所述第一人脸模板数据储存在储存设备中;
所述人脸登记设备判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值,若达到,则采集所述人脸处于所述第二位置时的第二面部特征;
所述人脸登记设备根据所述第二面部特征得到第二人脸模板数据;
所述人脸登记设备将所述第二人脸模板数据储存在储存设备中。
本发明实施例第二方面提供了一种人脸登记设备,具体包括:
接收模块,用于接收登记消息,所述登记消息指示所述人脸登记设备进行人脸登记;
第一检测模块,用于在预置时间内检测是否存在用户的人脸;
第一采集模块,用于若存在所述用户的人脸时采集所述人脸处于第一位置时的第一面部特征,所述第一位置为所述人脸登记设备检测到所述用户的人脸时所述人脸所处的位置;
第一生成模块,用于根据所述第一面部特征生成第一人脸模板数据;
第一储存模块,用于将所述第一人脸模板数据储存在储存设备中;
第一判断模块,用于判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值;
第二采集模块,用于若达到所述预设值时采集所述人脸处于所述第二位置时的第二面部特征;
第二生成模块,用于根据所述第二面部特征生成第二人脸模板数据;
第二储存模块,用于将所述第二人脸模板数据储存在储存设备中。
本发明实施例第三方面提供了一种人脸登记设备,具体包括:
中央处理器、存储器、存储介质、电源、无线网络接口以及输入输出接口;
通过调用所述存储器或存储介质上存储的操作指令,所述中央处理器,用于执行如下步骤:
接收登记消息,所述登记消息指示所述人脸登记设备进行人脸登记;
在预置时间内检测是否存在用户的人脸,若存在,则所述人脸登记设备采集所述人脸处于第一位置时的第一面部特征,所述第一位置为所述人脸登记设
备检测到所述用户的人脸时所述人脸所处的位置;
根据所述第一面部特征生成第一人脸模板数据;
将所述第一人脸模板数据储存在储存设备中;
判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值,若达到,则采集所述人脸处于所述第二位置时的第二面部特征;
根据所述第二面部特征得到第二人脸模板数据;
将所述第二人脸模板数据储存在储存设备中。
从以上技术方案可以看出,本发明实施例具有以下优点:人脸登记设备会采集用户第一位置的面部特征,并根据第一位置的面部特征生成第一人脸模板数据,并采集用户第一位置旋转后的第二位置的面部特征,根据第二位置的面部特征生成第二人脸模板数据,分别将第一人脸模板数据以及第二人脸模板数据储存在储存设备中。由于是采集当前位置用户人脸的面部特征以及旋转达到预设值的用户人脸的面部特征,取到更加完善的人脸模板,有利于比对时对姿势的限制更小,提高验证通过的成功率。
图1为本发明实施例中人脸登记方法的一个实施例示意图;
图2为本发明实施例中人脸登记方法的另一实施例示意图;
图3为本发明实施例中人脸登记设备的一个结构示意图;
图4为本发明实施例中人脸登记设备的另一结构示意图;
图5为本发明实施例中人脸登记设备的结构示意图。
请参阅图1,本发明实施例中人脸登记方法一个实施例包括:
101、人脸登记设备接收登记消息。
本实施例中,当需要进行人脸登记的时候,人脸登记设备可以接收到登记消息,该登记消息指示人脸登记设备进行人脸登记。
102、人脸登记设备在预置时间内检测是否存在用户的人脸,若存在则执行步骤103,若不存在则执行步骤104。
本实施例中,当人脸登记设备开始进行人脸登记时,会在预置的时间检测是否存在用户的人脸,若在预置的时间内检测到存在用户的人脸,则执行步骤103,若在预置的时间内未检测到用户的人脸,则执行步骤104。
103、人脸登记设备采集人脸处于第一位置时的第一面部特征。
本实施例中,当人脸登记设备在预置的时间内检测到存在用户的人脸,则人脸登记设备采集人脸处于第一位置时的面部特征,该第一位置为人脸登记设备检测到用户的人脸时该人脸所处的位置。
104、执行其他操作。
本实施例中,当人脸登记设备在预置的时间内未检测到存在用户的人脸,则执行其他操作。
105、人脸登记设备根据第一面部特征生成第一人脸模板数据。
本实施例中,当人脸登记设备采集到第一面部特征之后,人脸登记设备可以根据第一面部特征生成第一人脸模板数据。
106、人脸登记设备将第一人脸模板数据储存在储存设备中。
本实施例中,人脸登记设备在生成第一人脸模板数据之后,可以将第一人脸模板数据储存在储存设备中。
107、人脸登记设备判断人脸旋转后的第二位置相对于第一位置的旋转角度是否达到预设值,若达到,则执行步骤108,若未达到,则执行步骤109。
本实施例中,当人脸登记设备根据第一面部特征生成第一人脸模板数据后,人脸登记设备可以判断人脸旋转后的第二位置相对于第一位置的旋转角度是否达到预设值,若达到,则执行步骤108,若未达到,则执行步骤109。
108、人脸登记设备采集人脸处于第二位置时的第二面部特征。
本实施例中,当人脸登记设备确定人脸旋转后的第二位置相对于第一位置的旋转角度达到预设值,则人脸登记设备采集人脸处于该第二位置时人脸的第二面部特征。
109、人脸登记设备执行其他操作。
本实施例中,当人脸登记设备确定人脸旋转后的第二位置相对于第一位置的旋转角度未达到预设值,则人脸登记设备执行其他操作。
110、人脸登记设备根据第二面部特征生成第二人脸模板数据。
本实施例中,当人脸登记设备采集到第二面部特征后,可以根据第二面部特征生成第二人脸模板数据。
111、人脸登记设备将第二人脸模板数据储存于储存设备中。
本实施例中,当人脸登记设备生成第二人脸模板数据后,可以将该第二人脸模板数据储存到储存设备中。
需要说明的是,人脸登记设备通过步骤106可以将第一人脸模板数据储存在储存设备中,通过步骤107至步骤111生成第二人脸模板数据并储存在储存设备中,然而步骤106与步骤107至步骤111之间并没有先后执行的顺序,可以先执行步骤106,也可以先执行步骤107至步骤111,或者同时执行,具体此处不做限定。
需要说明的是,人脸登记设备通过步骤106可以将第一人脸模板数据储存在储存设备中,通过步骤111可以将第二人脸模板数据储存在储存设备中,人脸登记设备是将第二人脸模板数据与第一人脸模板数据关联的储存在存储设备中。
综上所述可以看出,人脸登记设备会采集用户第一位置的面部特征,并根据第一位置的面部特征生成第一人脸模板数据,并采集用户第一位置旋转后的第二位置的面部特征,根据第二位置的面部特征生成第二人脸模板数据,且人脸登记设备分别将第一人脸模板数据以及第二人脸模板数据储存在储存设备中。可以看出,人脸登记设备分别采集了用户的人脸正面对准人脸登记设备以及通过对登记过程中的不同旋转角度,仰角或俯角的判断且获取不同旋转角度、仰角或俯角的面部特征,能够更加全面的获取到整个面部的特征,取到更加完善的人脸模板,有利于比对时对姿势的限制更小,提高验证通过的成功率。
下面进行详细说明,请参阅图2,本发明实施例中人脸登记方法一个实施例包括:
201、进入面部登记界面。
本实施例中,当用户需要进行人脸登记的时候,人脸登记设备接收到登记消息,该登记消息指示人脸登记设备进行人脸登记,此时,人脸登记设备显示人脸登记界面。
202、人脸登记设备在预置的时间内检测是否存在人脸,若存在,则执行
步骤203,若不存在,则执行步骤204。
本实施例中,当人脸登记设备接收到登记消息之后,人脸登记设备会开启红外以及摄像头,该摄像头与该用户的人脸处于水平的位置,在预置的时间内检测是否存在人脸,若存在,则执行步骤203,若不存在则执行步骤204。
需要说明的是,人脸登记设备在预置的时间内检测到是否存在人脸,该人脸可以是正对人脸登记设备的摄像头,也可以是侧面相对于摄像头,也可以是对着人脸登记设备的任意位置,具***置此处不做限定,下面以人脸登记设备初次检测到人脸为正面对准摄像头为例来进行说明。
203、采集正面特征生成模板并保存。
本实施例中,当人脸登记设备在预置的时间内检测到存在人脸正面对准摄像头时,人脸登记设备可以采集人脸正面的面部特征,并且根据正面的面部特征生成正面人脸模板数据,人脸登记设备将生成的正面人脸模板数据保存在储存设备中。
204、退出登记。
本实施例中,当人脸登记设备在预置的时间内未检测到存在人脸,则人脸登记设备退出登记界面。
205、人脸登记设备判断用户面部移动轨迹是否偏离登记范围,若未偏移,则执行步骤206,若偏移,则执行步骤207。
本实施例中,人脸登记设备可以判断该需要登记的用户的面部移动轨迹是否偏离登记范围,若未偏移,则执行步骤206,若偏移,则执行步骤207。
需要说明的是,人脸登记设备在登记过程中,会实时去检测用户的面部轨迹,如果用户在登记过程中人脸的轨迹偏离登记范围时,人脸登记设备会退出登记界面。
需要说明的是,人脸登记设备在采集用户的人脸的各个侧面的面部特征之前,可以先判断该用户的面部移动轨迹是否偏移登记范围,当判断为未偏移登记范围时,才开始进行面部特征采集,此处已经进行了说明,为了简洁,下面在采集用户的人脸各个侧面的人脸面部特征之前不再进行详细描述。
206、人脸登记设备发出提示信息,提示需要登记面部特征的用户向左旋转。
本实施例中,当人脸登记设备判断用户的面部移动轨迹没有偏离登记范围之后,可以发出提示信息,提示用户向左旋转。
207、人脸登记设备退出登记。
本实施例中,当人脸登记设备判断用户的面部移动轨迹偏离登记范围之后,可以退出登记界面。
208、人脸登记设备判断人脸向左旋转角度是否达到预设值,若达到,则执行步骤209,若未达到则执行步骤210。
本实施例中,用户收到提示之后,会根据提示进行旋转,此时,人脸登记设备可以判断用户的人脸向左的旋转角度是否预设值,若达到,则执行步骤209,若未达到则执行步骤210。
209、人脸登记设备采集面部特征且生成人脸模板数据并保存至储存设备中。
本实施例中,人脸登记设备判断人脸向左旋转的角度达到预设值,则人脸登记设备采集人脸向左旋转的角度达到预设值时的人脸的面部特征,且根据该人脸面部特征生成左侧人脸模板数据,人脸登记设备将生成的左侧人脸模板数据保存至储存设备中。
210、人脸登记设备发出提示指示用户向左旋转直至向左旋转的角度达到预设值。
本实施例中,当人脸登记设备判断人脸向左旋转的角度未达到预设值,则人脸登记设备会发送提示信息,提示用户向左旋转,直到用户的人脸向左旋转的角度达到预设值。
211、人脸登记设备提示用户向右选转。
本实施例中,人脸登记设备可以发出提示信息,提示用户向右旋转。
212、人脸登记设备判断用户向右旋转的角度是否达到预设值,若达到,则执行步骤213,若未达到,则执行步骤214。
本实施例中,用户在收到提示信息之后,可以进行右转,人脸登记设备判断用户的人脸向右的旋转角度是否预设值,若达到,则执行步骤213,若未达到则执行步骤214。
213、人脸登记设备采集面部特征且生成人脸模板数据并保存至储存设备
中。
本实施例中,当人脸登记设备判断用户的人脸向右旋转的角度达到预设值之后,可以采集向右旋转后的人脸的面部特征,且根据该面部特征生成右侧人脸模板数据,人脸登记设备将右侧人脸模板数据储存于储存设备中。
214、人脸登记设备提示用户右转直至右转角度达到预设值。
本实施例中,当人脸登记设备判断用户向右旋转的角度未达到预设值时,人脸登记设备会发送提示,指示用户继续进行旋转直至向右旋转的角度达到预设值。
215、人脸登记设备提示用户的面部向下旋转。
本实施例中,人脸登记设备提示用户向下旋转,即人脸登记设备提示用户低头。
216、人脸登记设备判断该用户的人脸向下旋转的角度是否达到预设值,若达到,则执行步骤217,若未达到,则执行步骤218。
本实施例中,当用户根据提示信息将面部向下旋转后,人脸登记设备会判断向下旋转的角度是否达到预设值,若达到,则执行步骤217,若未达到,则执行步骤218。
217、人脸登记设备采集向下低头时面部特征且生成人脸模板数据并保存至储存设备中。
本实施例中,当人脸登记设备判断用户的面部向下旋转的角度达到预设值时,人脸登记设备采集用户的面部向下旋转角度达到预设值时的面部特征,且根据该向下旋转角度达到预设值的面部特征生成低头人脸模板数据,人脸登记设备将该低头时的人脸模板数据保存至储存设备中。
218、人脸登记设备提示用户的面部向下旋转直至向下旋转的角度达到预设值。
本实施例中,人脸登记设备判断用户的面部向下旋转的角度未达到预设值时,会发送提示信息,指示用户进行向下旋转,直至用户的面部向下旋转的角度达到预设值。
219、人脸登记设备提示用户控制面部向上旋转。
本实施例中,人脸登记设备发出提示信息,指示用户控制面部向上旋转。
220、人脸登记设备判断用户的面部向上旋转的角度是否达到预设值,若达到,则执行步骤221,若未达到,则执行步骤222。
本实施例中,用户根据提示信息,控制面部向上旋转,人脸登记设备判断用户的面部上仰的角度是否达到预设值,若达到,则执行步骤221,若未达到,则执行步骤222。
221、人脸登记设备采集向上抬头是人脸面部特征且生成人脸模板数据并保存至储存设备中。
本实施例中,人脸登记设备判断用户的面部上仰的角度达到预设值时,采集上仰角度达到预设值的面部特征,且根据上仰角度达到预设值的面部特征生成仰头人脸模板数据,人脸登记设备将该仰头时的人脸模板数据保存至储存设备中。
222、人脸登记设备提示用户的面部向上旋转直至上仰角度达到预设值。
本实施例中,人脸登记设备判断用户的面部上仰的角度未达到预设值时,会发送提示,指示用户继续旋转,直至上仰的角度达到预设值。
需要说明的是,上述说书的所有提示信息可以是语音,也可以是文字信息显示在登记界面上,也可以是视频信息,具体此处不做限定。
需要说明的是,人脸登记设备通过步骤203采集正面人脸面部特征且生成正面人脸模板数据,通过步骤209采集人脸左转时的人脸面部特征且生成左侧人脸模板数据,通过步骤213采集人脸右转时面部特征且生成右侧人脸模板数据,通过步骤217采集人脸低头时人脸面部特征且生成低头人脸模板数据,通过步骤221采集人脸仰头时人脸面部特征且生成仰头人脸模板数据,然而这些步骤之间并没有先后执行的顺序,可以先执行步骤203,也可以先执行步骤209,也可以先执行步骤213,也可以先执行步骤217,也可以先执行步骤221,具体此处不做限定。
需要说明的是,人脸登记设备将正面人脸模板数据、左侧人脸模板数据、右侧人脸模板数据、仰头时人脸模板数据以及低头时的人脸模板数据关联的储存在人脸模板数据,当有相同用户的其他人脸模板数据时,也可以关联的储存在储存设备中。
需要说明的是,人脸登记设备生成的各人脸模板数据存储在用户ID所处
的存储空间内,得到携带有用户的ID的人脸模板数据,当用户需要验证的时候,人脸登记设备采集用户需要验证时用户相对于人脸登记设备所处的位置时的人脸面部特征,且根据验证时的人脸面部特征生成验证时人脸模板数据,并与携带有用户ID的人脸模板数据进行对比,当人脸登记设备判断验证时人脸模板数据与携带有用户ID的人脸模板数据中的任一人脸模板数据的数据相似度达到预设的阈值时,则确定该用户验证通过,反之,则确定该用户验证未通过。
223、人脸登记设备登记完成。
本实施例中,当人脸登记设备完成人脸各个面的登记之后,可以退出登记界面。
综上所述可以看出,人脸登记设备采集人脸的正面以及各个侧面的面部特征,且根据采集到的正面以及各个侧面的面部特征生成与之对应的人脸模板数据,并将各个人脸模板数据存储在用户的ID所处的存储空间中,得到携带有用户ID的人脸模板数据。可以看出,人脸登记设备分别采集了用户的人脸正面对准人脸登记设备以及通过对登记过程中的不同旋转角度,仰角或俯角的判断且获取不同旋转角度、仰角或俯角的面部特征,能够取到更加完善的人脸模板,有利于比对时对姿势的限制更小,提高验证通过的成功率。
为了便于理解,下面结合具体应用场景进行详细描述:
用户A需要进行人脸登记,用户A会先点击人脸登记设备进入人脸登记设备,此时,用户A的人脸正对人脸登记设备的摄像头,人脸登记设备提示用户A“请正视摄像头”,人脸登记设备采集用户A正面的面部特征且根据正面的面部特征生成正面人脸模板数据,人脸登记设备会提示用户A左转、右转、抬头以及低头,并分别采集用户A左转时的人脸的左侧面的面部特征、右转时的人脸的右侧面的面部特征、低头时人脸的面部特征以及抬头时的人脸的面部特征,并分别根据采集到一系列面部特征生成一系列的人脸模板数据,并将生成的正面人脸模板数据以及经过旋转后的采集到的面部特征生成人脸的模板数据存储在用户的ID所处的存储空间中,得到携带有用户ID的人脸模板数据,退出人脸登记界面。当需要验证时,用户可以站在人脸登记设备前,人脸登记设备检测到存在人脸时,会采集当前的面部特征,并生成人脸模板数据,
并与储存设备中的人脸模板数据进行对比,当前生成的人脸模板数据与储存设备中的任一个人脸模板数据的数据相似度达到预设的阈值,则用户的验证通过。
请参阅图3,本发明实施例还提供了一种人脸登记设备,具体包括:
接收模块301,用于接收登记消息,登记消息指示人脸登记设备进行人脸登记;
第一检测模块302,用于在预置时间内检测是否存在用户的人脸;
第一采集模块303,用于若存在用户的人脸时采集人脸处于第一位置时的第一面部特征,第一位置为人脸登记设备检测到用户的人脸时人脸所处的位置;
第一生成模块304,用于根据第一面部特征生成第一人脸模板数据;
第一储存模块305,用于将第一人脸模板数据储存在储存设备中;
第一判断模块306,用于判断人脸旋转后的第二位置相对于第一位置的旋转角度是否达到预设值;
第二采集模块307,用于若达到预设值时采集人脸处于所述第二位置时的第二面部特征;
第二生成模块308,用于根据第二面部特征生成第二人脸模板数据;
第二储存模块309,用于将第二人脸模板数据储存在储存设备中。
为了便于理解,下面结合图4进行说明,本发明实施例的一种人脸登记设备,包括:
接收模块401,用于接收登记消息,登记消息指示人脸登记设备进行人脸登记;
第一检测模块402,用于在预置时间内检测是否存在用户的人脸;
第一提示模块403,用于发出第一提示信息,所述第一提示信息指示所述用户进行人脸登记。
第一采集模块404,用于若存在用户的人脸时采集人脸处于第一位置时的第一面部特征,第一位置为人脸登记设备检测到用户的人脸时人脸所处的位置;
第一生成模块405,用于根据第一面部特征生成第一人脸模板数据;
第一储存模块406,用于将第一人脸模板数据储存在储存设备中。
第二检测模块407,用于检测用户的面部移动轨迹;
第二判断模块408,用于判断面部移动轨迹是否偏离登记范围;
第一确定模块409,用于在确定面部移动轨迹偏离登记范围时,退出人脸登记;
第二确定模块410,用于在确定面部移动轨迹未偏离登记范围,进行人脸登记。
第二提示模块411,用于发出第二提示信息,第二提示信息指示用户控制面部进行旋转。
第一判断模块412,用于判断人脸旋转后的第二位置相对于第一位置的旋转角度是否达到预设值;
第二采集模块413,用于若达到预设值时采集人脸处于所述第二位置时的第二面部特征;
第二生成模块414,用于根据第二面部特征生成第二人脸模板数据;
第二储存模块415,用于第二人脸模板数据储存于储存设备中;
第三采集模块416,用于当人脸登记设备在检测范围内检测到用户人脸时,采集检测到用户人脸的第三面部特征;
第四生成模块417,用于根据第三面部特征生成第三人脸模板数据;
第三判断模块418,用于判断所述第三人脸模板数据与所述第一模板数据以及第二人脸模板数据中的任一模板数据的数据相似度是否达到预设的阈值;
第三确定模块419,用于在第三人脸模板数据与第一人脸模板数据以及第二模板数据中的任一模板数据的数据相似度达到预设的阈值时,则确定检测到用户的人脸验证通过;
第四确定模块420,用于在第三人脸模板数据与第一人脸模板数据以及第二模板数据中的任一模板数据的数据相似度未达到预设的阈值时,确定检测到用户的人脸验证未通过。
其中,第一判断模块412可以进一步包括:
第一判断单元4121,用于在当第一位置为用户的人脸正对人脸登记设备的位置时,判断人脸向左旋转后的位置相对于第一位置的旋转角度是否达到预
设值;
第一确定单元4122,用于在人脸向左旋转后的位置相对于第一位置的旋转角度是否达到预设值时,确定人脸向左旋转后的位置为第二位置;
和/或,
第二判断单元4123,用于在当第一位置为所述用户的人脸正对人脸登记设备的位置时,判断人脸向右旋转后的位置相对于第一位置的旋转角度是否达到预设值;
第二确定单元4124,用于在当人脸向右旋转后的位置相对于第一位置的旋转角度达到预设值时,确定所述人脸向右旋转后的位置为所述第二位置;
和/或,
第三判断单元4125,用于在所述第一位置为用户的人脸正对人脸登记设备的位置时,判断所述人脸向上旋转后的位置相对于第一位置的旋转角度是否达到预设值;
第三确定单元4126,用于在人脸向上旋转后的位置相对于第一位置的旋转角度达到预设值时,确定人脸向上旋转后的位置为第二位置;
和/或,
第四判断单元4127,用于在当第一位置为所述用户的人脸正对人脸登记设备的位置时,判断人脸向下旋转后的位置相对于第一位置的旋转角度是否达到预设值;
第四确定单元4128,用于在人脸向下旋转后的位置相对于第一位置的旋转角度达到预设值时,确定人脸向下旋转后的位置为第二位置。
本实施例中的人脸登记设备的各模块与单元之间的交互方式如前述图3所示实施例中的描述,具体此处不再赘述。
需要说明的时,在登记过程中,第二检测模块407会实时去检测用户的面部轨迹,如果用户在登记过程中人脸的面部运动轨迹偏离登记范围时,人脸登记设备会退出登记界面。
本实施例中,人脸登记设备通过接收模块401接收需要验证的用户的登记消息,第一检测模块402在预置的时间内检测是否存在人脸,若存在人脸,则第一采集模块404采集人脸处于第一位置时的第一面部特征,第一生成模块
405根据第一面部特征生成第一人脸模板数据,第一储存模块406将第一人脸模板数据储存在储存设备中,第二检测模块407会实时的进行检测人脸是否偏离登记范围,第一判断模块412判断人脸旋转后的第二位置相对于第一位置的旋转角度是否达到预设值,若达到,则第二采集模块413采集人脸处于第二位置时的第二面部特征,第二生成模块413根据第二面部特征生成第二人脸模板数据,第二储存模块415将第二人脸模板数据与第一人脸模板数据关联的储存在储存设备中。由此可以看出,人脸登记设备尽可能多的采集用户的人脸的面部特征,且根据采集到达到人脸的面部特征生成一系列携带有用户ID的模板数据,取到更加完善的人脸模板,有利于比对时对用户姿势的限制更小,提高验证通过的成功率。
请参阅图5,图5是本发明实施例提供的一种人脸登记设备的结构示意图,该人脸登记设备500可因配置或性能不同而产生比较大的差异,可以包括一个或一个以***处理器(central processing units,CPU)522(例如,一个或一个以上处理器)和存储器32,一个或一个以上存储应用程序542或数据544的存储介质530(例如一个或一个以上海量存储设备)。其中,存储器532和存储介质530可以是短暂存储或持久存储。存储在存储介质530的程序可以包括一个或一个以上模块(图示没标出),每个模块可以包括对服务器中的一系列指令操作。更进一步地,中央处理器522可以设置为与存储介质530通信,在人脸登记设备500上执行存储介质530中的一系列指令操作。
人脸登记设备500还可以包括一个或一个以上电源526,一个或一个以上有线或无线网络接口550,一个或一个以上输入输出接口558,和/或,一个或一个以上操作***541,例如Windows ServerTM,Mac OS XTM,UnixTM,LinuxTM,FreeBSDTM等等。
上述实施例中由人脸登记设备所执行的步骤可以基于该图5所示的人脸登记设备结构。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的***,装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中,应该理解到,所揭露的***,装置和方
法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个***,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,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)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,以上实施例仅用以说明本发明的技术方案,而非对其限制;尽管参照前述实施例对本发明进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。
Claims (13)
- 一种人脸登记方法,其特征在于,包括:人脸登记设备接收登记消息,所述登记消息指示所述人脸登记设备进行人脸登记;所述人脸登记设备在预置时间内检测是否存在用户的人脸,若存在,则所述人脸登记设备采集所述人脸处于第一位置时的第一面部特征,所述第一位置为所述人脸登记设备检测到所述用户的人脸时所述人脸所处的位置;所述人脸登记设备根据所述第一面部特征生成第一人脸模板数据;所述人脸登记设备将所述第一人脸模板数据储存在储存设备中;所述人脸登记设备判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值,若达到,则采集所述人脸处于所述第二位置时的第二面部特征;所述人脸登记设备根据所述第二面部特征得到第二人脸模板数据;所述人脸登记设备将所述第二人脸模板数据储存在储存设备中。
- 根据权利要求1所述的人脸登记方法,其特征在于,所述第一位置为所述用户的人脸正对所述人脸登记设备的位置。
- 根据权利要求2所述的人脸登记方法,其特征在于,所述人脸登记设备判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值包括:当所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,所述人脸登记设备判断所述人脸向左旋转后的位置相对于所述第一位置的旋转角度是否达到预设值,若是,则确定所述人脸向左旋转后的位置为所述第二位置;和/或,当所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,所述人脸登记设备判断所述人脸向右旋转后的位置相对于所述第一位置的旋转角度是否达到预设值,若是,则确定所述人脸向右旋转后的位置为所述第二位置;和/或,当所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,所述人脸登记设备判断所述人脸向上旋转后的位置相对于所述第一位置的旋转角 度是否达到预设值,若是,则确定所述人脸向上旋转后的位置为所述第二位置;和/或,当所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,所述人脸登记设备判断所述人脸向下旋转后的位置相对于所述第一位置的旋转角度是否达到预设值,若是,则确定所述人脸向下旋转后的位置为所述第二位置。
- 根据权利要求1所述的人脸登记方法,其特征在于,所述人脸登记设备在预置时间内检测是否存在用户的人脸之后,所述人脸登记设备采集所述人脸处于第一位置时的第一面部特征之前,所述方法还包括:所述人脸登记设备发出第一提示信息,所述第一提示信息指示所述用户进行人脸登记。
- 根据权利要求1所述的人脸登记方法,其特征在于,所述人脸登记设备根据所述第一面部特征生成第一人脸模板数据之后,所述人脸登记设备判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值之前,所述方法还包括:所述人脸登记设备发出第二提示信息,所述第二提示信息指示所述用户进行旋转。
- 根据权利要求1所述的人脸登记方法,其特征在于,所述人脸登记设备接收登记消息之后,所述方法还包括:所述人脸登记设备检测所述用户的面部移动轨迹;所述人脸登记设备判断所述面部移动轨迹是否偏离登记范围;若是,则所述人脸登记设备退出人脸登记;若否,则所述人脸登记设备进行人脸登记。
- 一种人脸登记设备,其特征在于,包括:接收模块,用于接收登记消息,所述登记消息指示所述人脸登记设备进行人脸登记;第一检测模块,用于在预置时间内检测是否存在用户的人脸;第一采集模块,用于若存在所述用户的人脸时采集所述人脸处于第一位置时的第一面部特征,所述第一位置为所述人脸登记设备检测到所述用户的人脸时所述人脸所处的位置;第一生成模块,用于根据所述第一面部特征生成第一人脸模板数据;第一储存模块,用于将所述第一人脸模板数据储存在储存设备中;第一判断模块,用于判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值;第二采集模块,用于若达到所述预设值时采集所述人脸处于所述第二位置时的第二面部特征;第二生成模块,用于根据所述第二面部特征生成第二人脸模板数据;第二储存模块,用于将所述第二人脸模板数据储存在储存设备中。
- 根据权利要求7所述的人脸登记设备,其特征在于,所述第一位置为所述用户的人脸正对所述人脸登记设备的位置。
- 根据权利要求7所述的人脸登记设备,其特征在于,所述第一判断模块包括:第一判断单元,用于在当所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,判断所述人脸向左旋转后的位置相对于所述第一位置的旋转角度是否达到预设值;第一确定单元,用于在若所述人脸向左旋转后的位置相对于所述第一位置的旋转角度是否达到预设值时,确定所述人脸向左旋转后的位置为所述第二位置;和/或,第二判断单元,用于在当所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,判断所述人脸向右旋转后的位置相对于所述第一位置的旋转角度是否达到预设值;第二确定单元,用于在当所述人脸向右旋转后的位置相对于所述第一位置的旋转角度达到预设值时,确定所述人脸向右旋转后的位置为所述第二位置;和/或,第三判断单元,用于在所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,判断所述人脸向上旋转后的位置相对于所述第一位置的旋转角度是否达到预设值;第三确定单元,用于在所述人脸向上旋转后的位置相对于所述第一位置的 旋转角度达到预设值时,确定所述人脸向上旋转后的位置为所述第二位置;和/或,第四判断单元,用于在当所述第一位置为所述用户的人脸正对所述人脸登记设备的位置时,判断所述人脸向下旋转后的位置相对于所述第一位置的旋转角度是否达到预设值;第四确定单元,用于在所述人脸向下旋转后的位置相对于所述第一位置的旋转角度达到预设值时,确定所述人脸向下旋转后的位置为所述第二位置。
- 根据权利要求7所述的人脸登记设备,其特征在于,所述人脸登记设备法还包括:第一提示模块,用于发出第一提示信息,所述第一提示信息指示所述用户进行人脸登记。
- 根据权利要求7所述的人脸登记设备,其特征在于,所述人脸登记设备还包括:第二提示模块,用于发出第二提示信息,所述第二提示信息指示所述用户控制面部进行旋转。
- 根据权利要求7所述的人脸登记设备,其特征在于,所述人脸登记设备还包括:第二检测模块,用于检测所述用户的面部移动轨迹;第二判断模块,用于判断所述面部移动轨迹是否偏离登记范围;第一确定模块,用于在确定所述面部移动轨迹偏离登记范围时,退出人脸登记;第二确定模块,用于在确定所述面部移动轨迹未偏离登记范围,进行人脸登记。
- 一种人脸登记设备,其特征在于,包括:中央处理器、存储器、存储介质、电源、无线网络接口以及输入输出接口;通过调用所述存储器或存储介质上存储的操作指令,所述中央处理器,用于执行如下步骤:接收登记消息,所述登记消息指示所述人脸登记设备进行人脸登记;在预置时间内检测是否存在用户的人脸,若存在,则采集所述人脸处于第 一位置时的第一面部特征,所述第一位置为所述人脸登记设备检测到所述用户的人脸时所述人脸所处的位置;根据所述第一面部特征生成第一人脸模板数据;将所述第一人脸模板数据储存在储存设备中;判断所述人脸旋转后的第二位置相对于所述第一位置的旋转角度是否达到预设值,若达到,则采集所述人脸处于所述第二位置时的第二面部特征;根据所述第二面部特征得到第二人脸模板数据;将所述第二人脸模板数据储存在储存设备中。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2016/101996 WO2018068261A1 (zh) | 2016-10-13 | 2016-10-13 | 一种人脸登记方法及人脸登记设备 |
CN201680001152.0A CN106489153A (zh) | 2016-10-13 | 2016-10-13 | 一种人脸登记方法及人脸登记设备 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2016/101996 WO2018068261A1 (zh) | 2016-10-13 | 2016-10-13 | 一种人脸登记方法及人脸登记设备 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018068261A1 true WO2018068261A1 (zh) | 2018-04-19 |
Family
ID=58285427
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2016/101996 WO2018068261A1 (zh) | 2016-10-13 | 2016-10-13 | 一种人脸登记方法及人脸登记设备 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106489153A (zh) |
WO (1) | WO2018068261A1 (zh) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108573218A (zh) * | 2018-03-21 | 2018-09-25 | 漳州立达信光电子科技有限公司 | 人脸数据采集方法及终端设备 |
CN110175572A (zh) * | 2019-05-28 | 2019-08-27 | 深圳市商汤科技有限公司 | 人脸图像处理方法及装置、电子设备及存储介质 |
CN112766086A (zh) * | 2021-01-04 | 2021-05-07 | 深圳阜时科技有限公司 | 一种识别模板注册方法及存储介质 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008227906A (ja) * | 2007-03-13 | 2008-09-25 | Aisin Seiki Co Ltd | 撮影範囲調節装置、撮影範囲調節方法及びコンピュータプログラム |
CN101408929A (zh) * | 2007-10-10 | 2009-04-15 | 三星电子株式会社 | 一种用于人脸识别***的多模板人脸注册方法和装置 |
CN102385692A (zh) * | 2010-08-31 | 2012-03-21 | 中国科学院深圳先进技术研究院 | 人脸偏转图像获取***及方法 |
CN102419819A (zh) * | 2010-10-25 | 2012-04-18 | 深圳市中控生物识别技术有限公司 | 人脸图像识别方法和*** |
CN103310179A (zh) * | 2012-03-06 | 2013-09-18 | 上海骏聿数码科技有限公司 | 一种基于人脸识别技术的最优姿态检测方法及*** |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103714347B (zh) * | 2013-12-30 | 2017-08-25 | 汉王科技股份有限公司 | 人脸识别方法及人脸识别装置 |
CN105488371B (zh) * | 2014-09-19 | 2021-04-20 | 中兴通讯股份有限公司 | 一种脸部识别方法和装置 |
-
2016
- 2016-10-13 WO PCT/CN2016/101996 patent/WO2018068261A1/zh active Application Filing
- 2016-10-13 CN CN201680001152.0A patent/CN106489153A/zh active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008227906A (ja) * | 2007-03-13 | 2008-09-25 | Aisin Seiki Co Ltd | 撮影範囲調節装置、撮影範囲調節方法及びコンピュータプログラム |
CN101408929A (zh) * | 2007-10-10 | 2009-04-15 | 三星电子株式会社 | 一种用于人脸识别***的多模板人脸注册方法和装置 |
CN102385692A (zh) * | 2010-08-31 | 2012-03-21 | 中国科学院深圳先进技术研究院 | 人脸偏转图像获取***及方法 |
CN102419819A (zh) * | 2010-10-25 | 2012-04-18 | 深圳市中控生物识别技术有限公司 | 人脸图像识别方法和*** |
CN103310179A (zh) * | 2012-03-06 | 2013-09-18 | 上海骏聿数码科技有限公司 | 一种基于人脸识别技术的最优姿态检测方法及*** |
Also Published As
Publication number | Publication date |
---|---|
CN106489153A (zh) | 2017-03-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10242364B2 (en) | Image analysis for user authentication | |
CN107066983B (zh) | 一种身份验证方法及装置 | |
JP6820062B2 (ja) | アイデンティティ認証方法ならびに装置、端末及びサーバ | |
TWI759613B (zh) | 基於多視角的多人臉場景刷臉支付風險防控方法和設備 | |
US10346990B2 (en) | Detecting facial liveliness | |
JP6610906B2 (ja) | 活動検出方法及びデバイス、並びに本人認証方法及びデバイス | |
WO2017181769A1 (zh) | 一种人脸识别方法、装置和***、设备、存储介质 | |
JP2018160237A (ja) | 顔認証方法及び装置 | |
TW201629851A (zh) | 一種影像識別系統及方法 | |
CN105426723A (zh) | 基于声纹识别、人脸识别以及同步活体检测的身份认证方法及*** | |
WO2016177154A1 (zh) | 一种切换移动终端工作模式的方法及装置 | |
US11721087B2 (en) | Living body detection method and apparatus, electronic device, storage medium, and related system to which living body detection method is applied | |
US10217009B2 (en) | Methods and systems for enhancing user liveness detection | |
CN108141445B (zh) | 用于人员重新识别的***和方法 | |
TW201512880A (zh) | 電子裝置、身份驗證系統及方法 | |
US11245707B2 (en) | Communication terminal, communication system, communication control method, and recording medium | |
WO2022166090A1 (zh) | 自助打印方法、装置、设备、存储介质及程序产品 | |
WO2018068261A1 (zh) | 一种人脸登记方法及人脸登记设备 | |
CN110612530A (zh) | 用于选择脸部处理中使用的帧的方法 | |
US20200275271A1 (en) | Authentication of a user based on analyzing touch interactions with a device | |
WO2019062347A1 (zh) | 人脸识别方法及相关产品 | |
CA2886136A1 (en) | Authentication using a video signature | |
TWI734246B (zh) | 人臉辨識的方法及裝置 | |
TWI584146B (zh) | 基於人臉識別的整合登錄系統及方法 | |
WO2017101071A1 (en) | Facial gesture captcha |
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: 16918845 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: 16918845 Country of ref document: EP Kind code of ref document: A1 |