WO2020113571A1 - 人脸识别数据处理方法、装置、移动设备和计算机可读存储介质 - Google Patents

人脸识别数据处理方法、装置、移动设备和计算机可读存储介质 Download PDF

Info

Publication number
WO2020113571A1
WO2020113571A1 PCT/CN2018/119883 CN2018119883W WO2020113571A1 WO 2020113571 A1 WO2020113571 A1 WO 2020113571A1 CN 2018119883 W CN2018119883 W CN 2018119883W WO 2020113571 A1 WO2020113571 A1 WO 2020113571A1
Authority
WO
WIPO (PCT)
Prior art keywords
recognition
feedback information
face recognition
error
face
Prior art date
Application number
PCT/CN2018/119883
Other languages
English (en)
French (fr)
Inventor
梁俊豪
Original Assignee
深圳市欢太科技有限公司
Oppo广东移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市欢太科技有限公司, Oppo广东移动通信有限公司 filed Critical 深圳市欢太科技有限公司
Priority to CN201880098811.6A priority Critical patent/CN112889062B/zh
Priority to PCT/CN2018/119883 priority patent/WO2020113571A1/zh
Publication of WO2020113571A1 publication Critical patent/WO2020113571A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition

Definitions

  • the present application relates to the field of image recognition, and in particular, to a face recognition data processing method, device, mobile device, and non-volatile computer-readable storage medium.
  • Face recognition is affected by the environment and the state of the face, such as lights, hair occlusion, etc., there may be cases of false recognition.
  • Embodiments of the present application provide a face recognition data processing method, apparatus, mobile device, and non-volatile computer-readable storage medium, which can automatically adjust the face recognition algorithm control threshold through feedback information to improve the accuracy of face recognition.
  • a face recognition data processing method including:
  • the mobile device receives feedback information for face recognition errors
  • Face recognition is performed according to the current face recognition algorithm after adjusting the recognition control threshold.
  • a face recognition data processing device including:
  • the receiving module is used to receive feedback information of face recognition errors through the mobile device
  • An adjustment module configured to adjust the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information
  • the recognition module is used for face recognition according to the current face recognition algorithm after adjusting the recognition control threshold.
  • a mobile device includes a memory and a processor.
  • the memory stores computer-readable instructions.
  • the processor causes the processor to perform the following steps:
  • Face recognition is performed according to the current face recognition algorithm after adjusting the recognition control threshold.
  • Face recognition is performed according to the current face recognition algorithm after adjusting the recognition control threshold.
  • the above face recognition data processing method, apparatus, mobile device, and non-volatile computer-readable storage medium receive the face recognition error feedback information through the mobile device, and adjust the current face recognition algorithm according to the face recognition error feedback information.
  • Recognition control threshold, and face recognition according to the current face recognition algorithm after adjusting the recognition control threshold.
  • face recognition is wrong, the face recognition algorithm control threshold is automatically and intelligently adjusted through face recognition feedback information, thereby reducing the next error
  • the recognition phenomenon improves the accuracy of face recognition.
  • FIG. 1 is a schematic diagram of an application environment of a face recognition data processing method in an embodiment.
  • FIG. 2 is a flowchart of a face recognition data processing method in an embodiment.
  • FIG. 3 is a schematic diagram of a lock screen interface in an embodiment.
  • FIG. 4 is a schematic diagram of a desktop after successful unlocking in an embodiment.
  • FIG. 5 is a flowchart of a face recognition data processing method in an embodiment.
  • FIG. 6 is a flowchart of face recognition in an embodiment.
  • FIG. 7 is a flowchart of adjusting the recognition control threshold in an embodiment.
  • FIG. 8 is a flowchart of a face recognition data processing method in a specific embodiment.
  • FIG. 9 is a structural block diagram of a face recognition data processing device in an embodiment.
  • FIG. 10 is a schematic diagram of an internal structure of a mobile device in an embodiment.
  • FIG. 11 is a block diagram of a partial structure of a mobile device related to this embodiment in one embodiment.
  • first and second used in the embodiments of the present application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.
  • the first control may be called a second control. Both the first control and the second control are controls, but they are not the same control.
  • FIG. 1 is an application environment diagram of a face recognition data processing method in an embodiment.
  • the application environment includes a mobile device 110.
  • the mobile device 110 is provided with a camera, which can collect facial images, and perform face recognition on the collected facial images.
  • the mobile device 110 is provided with a feedback information receiving button, which can be a physical button or a virtual button, and receives the button through the feedback information. Receive the face recognition error feedback information, adjust the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information, and perform face recognition according to the current face recognition algorithm after adjusting the recognition control threshold.
  • the mobile device 110 may be a smart phone, a tablet computer, a wearable device, a personal digital assistant, or the like.
  • FIG. 2 is a flowchart of a face recognition data processing method in an embodiment. As shown in FIG. 2, a face recognition data processing method is described by taking the mobile device in FIG. 1 as an example, and specifically includes:
  • the mobile device receives face recognition error feedback information.
  • the face recognition error feedback information is feedback information used to describe the face recognition error.
  • the face recognition error includes multiple types of errors, such as wrong acceptance recognition and wrong rejection recognition.
  • False acceptance recognition means that the wrong face is recognized as a preset face image after algorithm recognition.
  • the preset face image recorded in the mobile device is the face image of user A.
  • the algorithm mistakenly recognizes the face image of user B as the face image of user A, which leads to a successful face verification.
  • False rejection means that the correct face is recognized as a non-preset face image after algorithm recognition.
  • the preset face image recorded in the mobile device is the face image of user A.
  • the algorithm mistakenly recognized the face image of user A as the face image of another user, resulting in unsuccessful face verification.
  • the face recognition error feedback information is received through the operation on the mobile device, where the operation can be an operation that directly or indirectly acts on the mobile device, and the operation that directly acts on the mobile device can be a physical button or a virtual button on the screen Operation.
  • the operations indirectly acting on the mobile device may be gesture operations, eyeball operations, and so on. Different types of face recognition error feedback information can be fed back through different buttons, or different types of face recognition error feedback information can be fed back through different gestures and eye movements.
  • the feedback information of face recognition errors can be notified to the system layer through the UI interface.
  • the corresponding recognition control threshold in the current face recognition algorithm is adjusted according to the face recognition error feedback information.
  • the recognition control threshold is a threshold used to control the accuracy of face recognition, and different types of recognition control thresholds corresponding to different face recognition algorithms.
  • the recognition control threshold may include a comparison threshold, a prosthesis threshold, and so on.
  • the recognition control threshold may be a parameter of the network model.
  • the recognition control threshold may be a recognition control threshold in a 2D face recognition algorithm or a 3D face recognition algorithm.
  • the recognition control threshold needs to be adjusted through different adjustment methods.
  • the specific adjustment algorithm can be determined according to the content of the face recognition error feedback information and the meaning of the recognition control threshold.
  • the adjustment principle is that when the face recognition error feedback information is wrongly accepted recognition feedback information, the face recognition algorithm needs to be made stricter to avoid mistakenly recognizing the faces of other users as the preset user's faces.
  • the face recognition error feedback information is error rejection recognition feedback information, it is necessary to make the face recognition algorithm more relaxed to avoid that the correct face of the preset user cannot be recognized successfully under the influence of the environment or other factors, resulting in face Unable to verify success.
  • Operation 206 performing face recognition according to the current face recognition algorithm after adjusting the recognition control threshold.
  • the system confirms that the current face recognition is false recognition according to the error feedback information of face recognition, so as to adjust the threshold to reduce the false acceptance recognition and/or false rejection recognition next time or later.
  • the results of face recognition can be used to unlock, pay, start or close applications, permission settings and other operations that require face recognition verification.
  • Performing face recognition according to the current face recognition algorithm after adjusting the recognition control threshold improves the face Recognition accuracy, so as to ensure the effectiveness of the authority of each operation based on the face recognition results.
  • the above face recognition data processing method receives the face recognition error feedback information through the mobile device, adjusts the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information, and adjusts the current face recognition after adjusting the recognition control threshold
  • the algorithm performs face recognition. Face recognition is performed according to the current face recognition algorithm after adjusting the recognition control threshold.
  • the face recognition algorithm can be automatically adjusted dynamically and intelligently when face recognition is wrong, which can avoid the wrong recognition next time and improve The accuracy of face recognition.
  • the mobile device receiving face recognition error feedback information includes receiving face recognition error feedback information through controls displayed on the mobile device screen, and different controls transmit different face recognition error feedback information.
  • the controls on the mobile device screen receive the user's operation, and different controls correspond to different trigger events.
  • the trigger event corresponding to the first control is to pass the error to accept the recognition feedback information
  • the second control corresponds to The trigger event is to pass error and refuse to recognize feedback information.
  • the display position of each control may be different.
  • a control that transmits error recognition failure information such as unlock failure can be displayed on the screen lock interface, which is convenient for operating the control to transmit error rejection recognition feedback information when unlocking fails.
  • error acceptance recognition feedback information it can be displayed on the interface after successful unlocking, so that the control can be quickly operated when the error unlock success is successful, thereby transmitting error acceptance recognition feedback information.
  • FIG. 3 it is a schematic diagram of the first control displayed on the screen lock interface in one embodiment.
  • the first control 208 is used to transmit error rejection recognition feedback information to the face recognition algorithm module when the unlocking fails.
  • FIG. 4 it is a schematic diagram of a second control displayed on the desktop after successful unlocking in one embodiment.
  • the second control 210 is used to pass error acceptance recognition feedback information to the face recognition algorithm module when the unlocking is not successful.
  • the second control is in a hidden state, and the display of the second control can be triggered by a pull-down operation on the desktop.
  • different face recognition error feedback information is transmitted to the system through different controls displayed on the UI interface, and the face recognition error feedback information is quickly fed back through the operation of the control, thereby improving the face recognition error feedback information feedback Convenience and timeliness.
  • receiving face recognition error feedback information through the controls displayed on the mobile device screen includes: generating error acceptance recognition feedback information in response to the operation of the first control displayed on the mobile device screen, and accepting the error recognition The feedback information is used as face recognition error feedback information, and receives error acceptance recognition feedback information transmitted by the first control.
  • the operation of the first control displayed on the screen of the mobile device may be any custom operation such as clicking, the touch duration exceeds a preset threshold, and sliding operation.
  • the first control may be a control specifically used for feedback error acceptance recognition feedback information, or a control that also has other functions.
  • the error acceptance recognition feedback information is triggered to be generated by a preset operation corresponding to the received face recognition error feedback information. If swiping left on the first control indicates that a face error has occurred to accept recognition, then when the first control recognizes the operation of swiping left, error acceptance recognition feedback information is generated.
  • the first control corresponds to a preset feedback value.
  • the preset feedback value indicates that the identification feedback information is incorrectly accepted. If the value 1 is passed, it indicates that the identification feedback information is incorrectly accepted.
  • a value of 1 is generated.
  • the feedback value is transmitted to the face recognition system.
  • the face recognition system determines the adjustment direction of the recognition control threshold according to the magnitude of the feedback value.
  • generating error acceptance recognition feedback information includes: the mobile device obtains the current face image, and when the current face image and the preset unlock the person The face images actually do not match, and the current face recognition algorithm misrecognizes the current face image as the preset unlocked face, and triggers the successful unlocking of the screen of the mobile device, and generates recognition error feedback information in response to the operation on the first control.
  • the current face recognition algorithm misrecognizes the current face image as a preset unlock face, which triggers the successful unlocking of the screen of the mobile device, indicating that this unlocking is due to the face recognition algorithm mistakenly recognizing the faces of other users as mobile
  • the preset face unlocked by the device unlocks the face, resulting in false unlocking. For example, if the face of user A is entered, the mobile phone is unlocked by user B. Then, user A can operate the first control, and the first control generates error acceptance recognition feedback information according to the operation of user A.
  • the first control when the unlock error is successful, can quickly feedback the error acceptance recognition feedback information.
  • the application scenario may not be limited to unlocking. If it is another application scenario, such as payment through a face, entering the application through a face, etc., it can correspond to different controls that match the application scenario. Quick feedback error acceptance recognition feedback information .
  • receiving the face recognition error feedback information through the control displayed on the mobile device screen includes: generating an error rejection recognition feedback information in response to the operation of the second control displayed on the mobile device screen, converting the error Rejection recognition feedback information is used as face recognition error feedback information, and receives error rejection recognition feedback information transmitted by the second control.
  • the operation of the second control displayed on the screen of the mobile device may be any custom operation such as clicking, the touch duration exceeds a preset threshold, and sliding operation.
  • the second control may be a control specifically used for feedback error refusal to identify feedback information, or a control that also has other functions.
  • the error rejection recognition feedback information is triggered to be generated by a preset operation corresponding to the received face recognition error rejection recognition feedback information. If swiping right on the second control indicates that a face error has refused recognition, when the second control recognizes the operation of swiping right, error rejection recognition feedback information is generated.
  • the feedback error rejection recognition feedback information and the error acceptance recognition feedback information control are the same control, and different types of information are fed back through different operations of this control. If you slide to the left, the feedback error accepts the recognition feedback information, and if you slide to the right, the feedback error rejects the recognition feedback information. By unifying controls, you can save the generation of controls and save resources.
  • the second control corresponds to a preset feedback value.
  • the preset feedback value indicates that the error rejects the feedback information. For example, when the value 2 is passed, it indicates that the error rejects the feedback information.
  • the first control recognizes the preset operation, it generates a value of 2.
  • the feedback value is transmitted to the face recognition system.
  • the face recognition system determines the adjustment direction of the recognition control threshold according to the magnitude of the feedback value.
  • generating error rejection recognition feedback information includes: the mobile device obtains the current face image, and when the current face image and the preset unlock the face image For actual matching, the current face recognition algorithm misrecognizes the current face image as a non-preset unlock face, so that when the screen unlocking of the mobile device fails, an error rejection feedback message is generated in response to the operation acting on the second control.
  • the current face recognition algorithm misrecognizes the current face image as a non-preset unlocked face, which makes the screen unlocking of the mobile device fail, indicating that this unlocking is because the face recognition algorithm cannot recognize the preset unlock entered by the mobile device
  • the face fails to unlock due to the face. If the face of user A is entered, user A cannot unlock it. Then, user A can operate the second control, and the second control generates error acceptance rejection feedback information according to the operation of user A.
  • the second control when the unlocking of the correct face fails, the second control can be used to quickly feedback the error and refuse to recognize the feedback information.
  • the application scenario may not be limited to unlocking. If it is another application scenario, such as payment through a face, entering the application through a face, etc., it may correspond to different controls that match the application scenario. Quick feedback error rejection rejection feedback information .
  • the current face recognition algorithm is used to recognize the current face image as the preset duration of unlocking the face, and the target control is determined according to the duration, responding The action on the target control generates an error refusing to identify feedback information.
  • the current face image actually matches the preset unlocked face image, for example, the face of user A is entered, the current face image is the face of user A, and the current face recognition algorithm compares the current person with the first time
  • the face image recognition is not preset to unlock the face, which means that the current face recognition algorithm cannot recognize the face of user A within the first time period.
  • the current face image is recognized as the preset unlocked face, then the second moment
  • the difference from the starting time is the length of time that the current face recognition algorithm uses to recognize the current face image as the preset face unlock. The shorter the duration, the more sensitive the current face recognition algorithm and the higher the recognition degree.
  • the target control is determined according to the recognition time. Different target controls are used to instruct the current face recognition algorithm to adjust the recognition control threshold by different amplitudes.
  • the recognition control threshold is adjusted differently by the recognition time, which further improves the accuracy of the recognition control threshold adjustment.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; operation 204 includes: when the face recognition error feedback information is error acceptance recognition feedback information, raising the comparison threshold and/or Lower the threshold of the prosthesis.
  • the 2D face recognition algorithm mainly includes two thresholds, a comparison threshold and a prosthesis threshold.
  • the comparison threshold is used to control the threshold of the similarity between the entered face and the face to be verified. The lower the threshold, the easier the verification is. If the face is verified for unlocking, the lower the threshold, the easier it is to unlock.
  • the face recognition error feedback information is wrongly accepted recognition feedback information, it means that the current face recognition algorithm is easy to recognize other faces as preset faces, indicating that the similarity threshold is low, and the comparison threshold needs to be increased.
  • the comparison threshold is used as a threshold for determining the degree of authenticity of the face attempted to unlock. The higher the threshold, the easier it is to identify the fake face as a real face.
  • the comparison threshold can be increased and the prosthesis threshold can be lowered, or both can be adjusted, and the adjustment range can be matched with the current face recognition algorithm or customized.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; operation 204 includes: when the face recognition error feedback information is error rejection recognition feedback information, lowering the comparison threshold and/or Raise the prosthesis threshold.
  • the face recognition error feedback information is error rejection recognition feedback information
  • the current face recognition algorithm is easy to enter the correct face recognition failure, such as the mobile device enters the face of user A, but when user A is exposed to light
  • the face recognition error feedback information is false rejection recognition feedback information
  • the current face recognition algorithm is easy to be mistakenly recognized as a fake face even if it is a real face. It can be understood that lowering the comparison threshold and raising the prosthesis threshold can be adjusted either or both, and the amplitude of the adjustment can be matched with the current face recognition algorithm or customized.
  • operation 204 before operation 204, as shown in FIG. 5, it also includes:
  • Operation 302 Obtain the current recognition scene corresponding to the current face recognition, and obtain the feedback recognition scene corresponding to the face recognition error feedback information.
  • operation 304 it is judged that when the current recognition scene matches the feedback recognition scene, if yes, the operation proceeds to operation 204, and if it does not match, no adjustment is made.
  • the background pattern, illuminance, brightness, occlusion, distance, face clarity, etc. can be used as judgment factors for different scenes, and the current scene judgment factors corresponding to the current face recognition can be obtained by analyzing the currently collected images, Compared with the feedback scene judgment factors corresponding to the feedback recognition scene, it is judged whether the current recognition scene matches the feedback recognition scene, and a specific matching algorithm can be defined. In one embodiment, when the similarity of the judgment factors exceeding the preset number exceeds the preset threshold, it is determined that the current recognition scene matches the feedback recognition scene.
  • the picture framed by the current camera and the picture framed by the feedback recognition scene can also be directly judged by the image analysis algorithm to determine whether they are the same or similar scenes. If they are the same or similar scenes, the current recognition scene and the feedback recognition scene match.
  • the recognition control threshold is adjusted only if the current recognition scene matches the feedback recognition scene, which further improves the accuracy of the recognition control threshold adjustment and avoids wrong adjustment.
  • the method further includes:
  • Operation 402 obtaining a feedback recognition scene corresponding to the face recognition error feedback information, and establishing a matching relationship between the feedback recognition scene and the adjusted recognition control threshold.
  • different scenes corresponding to the recognition error feedback information fed back by the user are obtained, and the adjusted recognition control thresholds under different scenes are obtained, so as to establish a matching relationship between different feedback recognition scenes and the adjusted recognition control thresholds.
  • the adjusted recognition control threshold in scenario A when face recognition is performed under scene A and scene B, both face recognition error feedback information is received, so that scene A and scene B are different feedback recognition scenes.
  • the adjusted target first recognition control threshold is obtained, and in the second adjustment of the recognition control threshold in scenario B, the adjusted target second recognition control threshold is obtained, and established
  • the matching relationship between the scene A and the target first recognition control threshold establishes the matching relationship between the scene B and the target second recognition control threshold.
  • the current recognition scene corresponding to the current face recognition is obtained, and a target feedback recognition scene matching the current recognition scene is determined.
  • the current recognition scene corresponding to the current face recognition is obtained by analyzing the currently collected images, and compared with each feedback recognition scene in the matching relationship to obtain the target feedback recognition scene matching the current recognition scene, and the target feedback recognition scene
  • the specific matching algorithm can be customized.
  • the target recognition control threshold corresponding to the target feedback recognition scene is obtained according to the matching relationship.
  • the recognition control thresholds corresponding to each feedback recognition scene are adjusted thresholds that match the scene, different recognition control thresholds are used in different scenes, and the accuracy of face recognition in different scenes is improved. .
  • the target recognition control threshold corresponding to the target feedback recognition scene since the target feedback recognition scene is a scene matching the current recognition scene, the target recognition control threshold is also more applicable to the current recognition scene.
  • face recognition is performed according to the current face recognition algorithm corresponding to the target recognition control threshold.
  • the current face recognition algorithm is obtained according to the target recognition control threshold, and face recognition is performed to ensure that the previously adjusted control threshold matching the current recognition scene is used in the current recognition scene, thereby improving the face in the current recognition scene Recognition accuracy.
  • operation 204 includes:
  • an initial recognition control threshold corresponding to the current face recognition algorithm is obtained.
  • the initial recognition control threshold is a recognition control threshold that has not been adjusted, such as a recognition control threshold corresponding to an initial model corresponding to the current face recognition algorithm, and a customizable recognition control threshold.
  • Operation 204b obtaining the recognition control threshold adjustment range, and adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information, so that the gap between the adjusted current recognition control threshold and the initial recognition control threshold is adjusted at the recognition control threshold Within range.
  • the recognition control threshold adjustment range controls the recognition control threshold within a certain range. If the initial recognition control threshold is 10 in one embodiment, the recognition control threshold adjustment range is [-10-10], which indicates that the recognition control threshold The amplitude of 10 can only be adjusted above and below 10. If the recognition control threshold exceeds the adjustment range of the recognition control threshold after adjustment, it means that the adjustment is excessive. Because the threshold adjustment must have a certain range, not too wide, otherwise it will seriously affect the false acceptance rate and false rejection rate. The false acceptance rate is the probability of successful verification of the wrong face after algorithm recognition, and the false rejection rate is the correct face Probability of verification failure after algorithm identification.
  • the adjustment of the threshold value allows slight fluctuations, but it needs to be restored under appropriate circumstances, because from the point of view of big data, the default threshold false acceptance rate and false rejection rate belong to a relatively balanced state. Breaking this balance for a long time has New problems may be introduced in other scenarios.
  • the difference between the adjusted current recognition control threshold and the initial recognition control threshold is within the recognition control threshold adjustment range, which ensures the stability of face recognition.
  • the mobile device displays different feedback controls on different interfaces.
  • the mobile device is in the lock screen state, and the first control is displayed on the interface corresponding to the lock screen state.
  • the character “I can’t unlock it?” is displayed on the first control to prompt the user
  • I cannot unlock it I operate the first control to feedback the error and refuse to recognize the feedback information.
  • the mobile device receives feedback error rejection recognition feedback information through the first control, and the user notifies the system rejection of the error rejection recognition feedback information through the UI interface. For example, if the face of user A is entered and user A cannot unlock it, the first control receives the click operation of user A, and the first control passes the error rejection recognition feedback information to the face recognition algorithm module of the system layer.
  • Operation 504 when the system layer receives the error rejection recognition feedback information, lower the comparison threshold, increase the prosthesis threshold, and ensure that the adjusted comparison threshold and the prosthesis threshold are within the corresponding adjustment range, respectively.
  • Operation 506 when the user B unlocks successfully, the second control is displayed on the interface where the unlock is successful, and the character "not unlocked by yourself?" is displayed on the first control to prompt the user to operate the first control when not unlocked by himself.
  • Feedback error accepts identification feedback information.
  • the mobile device receives the feedback error through the second control and accepts the identification feedback information, for example, the face of user A is entered, the face of user B passes the verification, and the screen is unlocked successfully.
  • Operation 508 when the system layer receives the error acceptance recognition feedback information, raise the comparison threshold, lower the prosthesis threshold, and ensure that the adjusted comparison threshold and the prosthesis threshold are within corresponding adjustment ranges, respectively.
  • control algorithm threshold can be adjusted by operating the control to adjust the recognition control threshold, and the threshold can be automatically adjusted through the feedback information, and the same user can use it again in the same scenario next time. Face unlocking can reduce the phenomenon of false recognition and improve the accuracy of face recognition. And the entire adjustment process does not need to update the software version, reducing development costs.
  • steps in the flowcharts of FIG. 2 and FIG. 5 to FIG. 8 are displayed in order according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless clearly stated in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least some of the steps in FIGS. 2 and 5 to 8 may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but may be executed at different times. These The execution order of the sub-steps or stages is not necessarily sequential, but may be executed in turn or alternately with at least a part of other steps or sub-steps or stages of other steps.
  • An embodiment of the present application also provides a mobile device.
  • the mobile device includes a memory and a processor.
  • the memory stores computer-readable instructions.
  • the processor causes the processor to perform the following steps: receiving face recognition error feedback information through the mobile device Adjust the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information; perform face recognition according to the current face recognition algorithm after adjusting the recognition control threshold.
  • the mobile device receiving face recognition error feedback information includes receiving face recognition error feedback information through controls displayed on the mobile device screen, and different controls transmit different face recognition error feedback information.
  • receiving face recognition error feedback information through the controls displayed on the mobile device screen includes: generating error acceptance recognition feedback information in response to the operation of the first control displayed on the mobile device screen, and accepting the error recognition The feedback information is used as face recognition error feedback information; receiving error feedback information received by the first control and receiving error recognition.
  • generating error acceptance recognition feedback information includes: the mobile device obtains the current face image; when the current face image and the preset unlock face image Actual mismatch, the current face recognition algorithm misrecognizes the current face image as the preset unlocked face, triggers the successful unlocking of the screen of the mobile device, generates an error to accept the recognition feedback information in response to the operation on the first control .
  • receiving face recognition error feedback information through the controls displayed on the mobile device screen includes: in response to the operation of the second control displayed on the mobile device screen, generating error rejection recognition feedback information, rejecting the error recognition
  • the feedback information is used as face recognition error feedback information; the error received by the second control refuses to recognize the feedback information.
  • generating error rejection recognition feedback information includes: acquiring the current face image through the mobile device; when the current face image and the preset unlock the face The images actually match, and the current face recognition algorithm misrecognizes the current face image as a non-preset unlocked face, so that when the screen unlocking of the mobile device fails, an error rejection feedback message is generated in response to the operation on the second control.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information includes: When the identification error feedback information is that the identification feedback information is received in error, the comparison threshold is increased and/or the prosthesis threshold is lowered.
  • receiving the face recognition error feedback information through the control displayed on the mobile device screen includes: generating an error rejection recognition feedback information in response to the operation of the second control displayed on the mobile device screen, converting the error Reject recognition feedback information as face recognition error feedback information; receive error rejection recognition feedback information transmitted by the second control.
  • generating error rejection recognition feedback information includes: acquiring the current face image through the mobile device; when the current face image is unlocked from the preset Face images actually match, and the current face recognition algorithm misidentifies the current face image as a non-preset unlocked face, so that when the screen unlocking of the mobile device fails, it responds to the action on the second control The operation generates the error rejection identification feedback information.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information, including: when face recognition error When the feedback information is wrongly accepted recognition feedback information, increase the comparison threshold and/or lower the prosthesis threshold.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information, including: when face recognition error When the feedback information is an error rejection of the feedback information, the comparison threshold is lowered and/or the prosthesis threshold is increased.
  • the processor performs the following steps: obtaining the current recognition scene corresponding to the current face recognition, obtaining the feedback recognition scene corresponding to the face recognition error feedback information; when the current recognition scene matches the feedback recognition scene, enter according to the person The step of adjusting the corresponding recognition control threshold in the current face recognition algorithm based on the feedback information of face recognition error.
  • the processor performs the following steps: obtaining a feedback recognition scene corresponding to the face recognition error feedback information, establishing a matching relationship between the feedback recognition scene and the adjusted recognition control threshold; obtaining the current corresponding to the current face recognition Identify the scene, determine the target feedback recognition scene that matches the current recognition scene; obtain the target recognition control threshold corresponding to the target feedback recognition scene according to the matching relationship; perform face recognition according to the current face recognition algorithm corresponding to the target recognition control threshold.
  • adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information includes: obtaining the initial recognition control threshold corresponding to the current face recognition algorithm; obtaining the recognition control threshold adjustment range; according to the person The face recognition error feedback information adjusts the corresponding recognition control threshold in the current face recognition algorithm so that the difference between the adjusted current recognition control threshold and the initial recognition control threshold is within the recognition control threshold adjustment range.
  • a face recognition data processing device includes a receiving module 602, an adjustment module 604, and a recognition module 606. among them:
  • the receiving module 602 is used to receive face recognition error feedback information through the mobile device.
  • the adjustment module 604 is used to adjust the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information.
  • the recognition module 606 is used for face recognition according to the current face recognition algorithm after adjusting the recognition control threshold.
  • the receiving module 602 is further configured to receive face recognition error feedback information through controls displayed on the mobile device screen, and different controls transmit different face recognition error feedback information.
  • the receiving module 602 is further configured to generate error acceptance recognition feedback information in response to the operation of the first control displayed on the screen of the mobile device, use the error acceptance recognition feedback information as face recognition error feedback information, and receive the first An error passed by a control accepts identification feedback information.
  • the receiving module 602 is also used to obtain the current face image through the mobile device.
  • the current face recognition algorithm misrecognizes the current face image as It is preset to unlock the human face, and when the screen unlocking of the mobile device is triggered successfully, an error acceptance recognition feedback information is generated in response to an operation acting on the first control.
  • the receiving module 602 is further configured to generate error rejection recognition feedback information in response to the operation of the second control displayed on the mobile device screen, and use the error rejection recognition feedback information as face recognition error feedback information; Receive the error passed by the second control and refuse to identify the feedback information.
  • the receiving module 602 is also used to obtain the current face image through the mobile device.
  • the current face recognition algorithm misrecognizes the current face image as a non
  • the preset unlocks the human face, so that when the screen unlocking of the mobile device fails, an error rejection feedback message is generated in response to the operation on the second control.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; the adjustment module 604 is further configured to increase the comparison threshold and/or when the face recognition error feedback information is falsely accepted recognition feedback information Or lower the prosthesis threshold.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; the adjustment module 604 is further configured to reduce the ratio when the face recognition error feedback information is error rejection recognition feedback information Threshold and/or increase the prosthesis threshold.
  • the device further includes:
  • the scene matching module is used to obtain the current recognition scene corresponding to the current face recognition and the feedback recognition scene corresponding to the face recognition error feedback information.
  • the adjustment module 604 is entered.
  • the device further includes:
  • the matching relationship module is used to obtain a feedback recognition scene corresponding to the face recognition error feedback information, and establish a matching relationship between the feedback recognition scene and the adjusted recognition control threshold.
  • the scene adjustment module is used to obtain the current recognition scene corresponding to the current face recognition, determine the target feedback recognition scene matching the current recognition scene, and obtain the target recognition control threshold corresponding to the target feedback recognition scene according to the matching relationship.
  • the recognition module 606 is also used to perform face recognition according to the current face recognition algorithm corresponding to the target recognition control threshold.
  • the adjustment module 604 is further used to obtain the initial recognition control threshold corresponding to the current face recognition algorithm; obtain the recognition control threshold adjustment range; and adjust the corresponding face recognition algorithm according to the face recognition error feedback information.
  • the recognition control threshold is such that the difference between the adjusted current recognition control threshold and the initial recognition control threshold is within the recognition control threshold adjustment range.
  • each module in the face recognition data processing apparatus may be in the form of computer-readable instructions.
  • the computer readable instructions can run on the terminal or server.
  • the program module composed of the computer-readable instructions may be stored in the memory of the terminal or the server.
  • the mobile device includes a processor and a memory connected by a system bus.
  • the processor is used to provide computing and control capabilities to support the operation of the entire electronic device.
  • the memory may include a non-volatile storage medium and internal memory.
  • the non-volatile storage medium stores an operating system and computer-readable instructions.
  • the computer readable instructions can be executed by the processor to implement a face recognition data processing method provided by the following embodiments.
  • the internal memory provides a cached operating environment for operating system computer-readable instructions in the non-volatile storage medium.
  • the mobile device may be a mobile phone, a tablet computer, a personal digital assistant or a wearable device.
  • An embodiment of the present application also provides a mobile device.
  • the mobile device can be any terminal device including a mobile phone, tablet computer, PDA (Personal Digital Assistant), POS (Point of Sales), in-vehicle computer, wearable device, etc. Taking the mobile device as a mobile phone for example .
  • An embodiment of the present application also provides a mobile device.
  • the above mobile device includes an image processing circuit.
  • the image processing circuit may be implemented using hardware and/or software components, and may include various processing units that define an ISP (Image Signal Processing) image signal processing pipeline.
  • ISP Image Signal Processing
  • 11 is a schematic diagram of an image processing circuit in an embodiment. As shown in FIG. 11, for ease of explanation, only various aspects of the image processing technology related to the embodiments of the present application are shown.
  • the image processing circuit includes an ISP processor 740 and a control logic 750.
  • the image data captured by the imaging device 710 is first processed by the ISP processor 740, which analyzes the image data to capture image statistical information that can be used to determine and/or one or more control parameters of the imaging device 710.
  • the imaging device 710 may include a camera having one or more lenses 712 and an image sensor 714.
  • the image sensor 714 may include a color filter array (such as a Bayer filter), and the image sensor 714 may acquire light intensity and wavelength information captured by each imaging pixel of the image sensor 714 and provide a set of raw data that can be processed by the ISP processor 740 Image data.
  • the sensor 720 may provide the acquired image processing parameters (such as anti-shake parameters) to the ISP processor 740 based on the sensor 720 interface type.
  • the sensor 720 interface may utilize SMIA (Standard Mobile Imaging Architecture, standard mobile imaging architecture) interface, other serial or parallel camera interfaces, or a combination of the above interfaces.
  • SMIA Standard Mobile Imaging Architecture, standard mobile imaging architecture
  • the image sensor 714 may also send the raw image data to the sensor 720, and the sensor 720 may provide the raw image data to the ISP processor 740 based on the sensor 720 interface type, or the sensor 720 stores the raw image data in the image memory 730.
  • the ISP processor 740 processes the original image data pixel by pixel in various formats.
  • each image pixel may have a bit depth of 8, 10, 12, or 14 bits, and the ISP processor 740 may perform one or more image processing operations on the original image data to collect statistical information about the image data.
  • the image processing operation can be performed with the same or different bit depth accuracy.
  • the ISP processor 740 may also receive image data from the image memory 730.
  • the sensor 720 interface sends the original image data to the image memory 730, and the original image data in the image memory 730 is provided to the ISP processor 740 for processing.
  • the image memory 730 may be a part of a memory device, a storage device, or an independent dedicated memory in an electronic device, and may include a DMA (Direct Memory Access) feature.
  • DMA Direct Memory Access
  • the ISP processor 740 may perform one or more image processing operations, such as time-domain filtering.
  • the processed image data can be sent to the image memory 730 for additional processing before being displayed.
  • the ISP processor 740 receives the processed data from the image memory 730, and performs image data processing in the original domain and in the RGB and YCbCr color spaces on the processed data.
  • the image data processed by the ISP processor 740 may be output to the display 770 for viewing by the user and/or further processed by a graphics engine or GPU (Graphics Processing Unit).
  • the output of the ISP processor 740 may also be sent to the image memory 730, and the display 770 may read image data from the image memory 730.
  • the image memory 730 may be configured to implement one or more frame buffers.
  • the output of the ISP processor 740 may be sent to the encoder/decoder 760 in order to encode/decode image data.
  • the encoded image data can be saved and decompressed before being displayed on the display 770 device.
  • the encoder/decoder 760 may be implemented by a CPU or GPU or coprocessor.
  • the statistical data determined by the ISP processor 740 may be sent to the control logic 750 unit.
  • the statistical data may include image sensor 714 statistical information such as auto exposure, auto white balance, auto focus, flicker detection, black level compensation, and lens 712 shading correction.
  • the control logic 750 may include a processor and/or a microcontroller that executes one or more routines (such as firmware). The one or more routines may determine the control parameters and ISP processing of the imaging device 710 based on the received statistical data Control parameters of the controller 740.
  • control parameters of the imaging device 710 may include sensor 720 control parameters (such as gain, integration time of exposure control, anti-shake parameters, etc.), camera flash control parameters, lens 712 control parameters (such as focal length for focusing or zooming), or these The combination of parameters.
  • the ISP control parameters may include gain levels and color correction matrices used for automatic white balance and color adjustment (eg, during RGB processing), and lens 712 shading correction parameters.
  • the ISP processor 740 included in the mobile device implements the steps of the face recognition data processing method when executing computer-readable instructions stored on the memory.
  • the embodiments of the present application also provide a computer-readable storage medium.
  • One or more non-volatile computer-readable storage media containing computer-readable instructions.
  • the computer-readable instructions When the computer-readable instructions are executed by the processor, the following steps are implemented: receiving face recognition error feedback information through a mobile device; based on face recognition errors The feedback information adjusts the corresponding recognition control threshold in the current face recognition algorithm; and performs face recognition according to the current face recognition algorithm after adjusting the recognition control threshold.
  • the mobile device receiving face recognition error feedback information includes receiving face recognition error feedback information through controls displayed on the mobile device screen, and different controls transmit different face recognition error feedback information.
  • receiving face recognition error feedback information through the controls displayed on the mobile device screen includes: generating error acceptance recognition feedback information in response to the operation of the first control displayed on the mobile device screen, and accepting the error recognition The feedback information is used as face recognition error feedback information; receiving error feedback information received by the first control and receiving error recognition.
  • generating error acceptance recognition feedback information includes: the mobile device obtains the current face image; when the current face image and the preset unlock face image Actual mismatch, the current face recognition algorithm misrecognizes the current face image as the preset unlocked face, triggers the successful unlocking of the screen of the mobile device, generates an error to accept the recognition feedback information in response to the operation on the first control .
  • receiving face recognition error feedback information through the controls displayed on the mobile device screen includes: in response to the operation of the second control displayed on the mobile device screen, generating error rejection recognition feedback information, rejecting the error recognition
  • the feedback information is used as face recognition error feedback information; the error received by the second control refuses to recognize the feedback information.
  • generating error rejection recognition feedback information includes: acquiring the current face image through the mobile device; when the current face image and the preset unlock the face The images actually match, and the current face recognition algorithm misrecognizes the current face image as a non-preset unlocked face, so that when the screen unlocking of the mobile device fails, an error rejection feedback message is generated in response to the operation on the second control.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information includes: When the identification error feedback information is that the identification feedback information is received in error, the comparison threshold is increased and/or the prosthesis threshold is lowered.
  • receiving the face recognition error feedback information through the controls displayed on the mobile device screen includes: in response to the operation of the second control displayed on the mobile device screen, generating an error rejecting the recognition feedback information, and converting the error Reject recognition feedback information as face recognition error feedback information; receive error rejection recognition feedback information transmitted by the second control.
  • generating error rejection recognition feedback information includes: acquiring the current face image through the mobile device; when the current face image is unlocked from the preset Face images actually match, and the current face recognition algorithm misidentifies the current face image as a non-preset unlocked face, so that when the screen unlocking of the mobile device fails, it responds to the action on the second control The operation generates the error rejection identification feedback information.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information, including: when face recognition error When the feedback information is wrongly accepted recognition feedback information, increase the comparison threshold and/or lower the prosthesis threshold.
  • the recognition control threshold includes at least one of a comparison threshold and a prosthesis threshold; adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information, including: when face recognition error When the feedback information is an error rejection of the feedback information, the comparison threshold is lowered and/or the prosthesis threshold is increased.
  • the computer readable instructions are executed by the processor: obtain the current recognition scene corresponding to the current face recognition, obtain the feedback recognition scene corresponding to the face recognition error feedback information; when the current recognition scene matches the feedback recognition scene Enter the step of adjusting the corresponding recognition control threshold in the current face recognition algorithm based on the face recognition error feedback information.
  • the computer readable instructions are executed by the processor: acquiring a feedback recognition scene corresponding to the face recognition error feedback information, establishing a matching relationship between the feedback recognition scene and the adjusted recognition control threshold; obtaining the current face Identify the corresponding current recognition scene and determine the target feedback recognition scene matching the current recognition scene; obtain the target recognition control threshold corresponding to the target feedback recognition scene according to the matching relationship; perform face recognition according to the current face recognition algorithm corresponding to the target recognition control threshold .
  • adjusting the corresponding recognition control threshold in the current face recognition algorithm according to the face recognition error feedback information includes: obtaining the initial recognition control threshold corresponding to the current face recognition algorithm; obtaining the recognition control threshold adjustment range; according to the person The face recognition error feedback information adjusts the corresponding recognition control threshold in the current face recognition algorithm so that the difference between the adjusted current recognition control threshold and the initial recognition control threshold is within the recognition control threshold adjustment range.
  • Embodiments of the present application also provide a computer-readable instruction product.
  • a computer-readable instruction product containing instructions, which when executed on a computer, causes the computer to perform a face recognition data processing method.
  • Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory can include random access memory (RAM), which acts as external cache memory.
  • RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR) SDRAM, enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
  • SRAM static RAM
  • DRAM dynamic RAM
  • SDRAM synchronous DRAM
  • DDR double data rate SDRAM
  • ESDRAM enhanced SDRAM
  • SLDRAM synchronous Link (Synchlink) DRAM
  • Rambus direct RAM
  • DRAM direct memory bus dynamic RAM
  • RDRAM memory bus dynamic RAM

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Collating Specific Patterns (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种人脸识别数据处理方法，包括：移动设备接收人脸识别错误反馈信息，根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；及根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。

Description

人脸识别数据处理方法、装置、移动设备和计算机可读存储介质 技术领域
本申请涉及图像识别领域，特别是涉及一种人脸识别数据处理方法、装置、移动设备和非易失性计算机可读存储介质。
背景技术
随着图像识别技术的发展，越来越多的移动设备使用人脸识别完成认证，如解锁、付款验证等，给人们的生活带来便捷。人脸识别受到环境和人脸状态的影响，如灯光、头发遮挡等，可能存在误识别的情况。
传统的系统对误识别的情况不做任何处理，下次同一用户在同一场景下再使用人脸识别认证，很可能还会出现误识别的现象。
发明内容
本申请实施例提供一种人脸识别数据处理方法、装置、移动设备和非易失性计算机可读存储介质，可以通过反馈信息自动调整人脸识别算法控制阈值，提高人脸识别准确率。
一种人脸识别数据处理方法，包括：
移动设备接收人脸识别错误反馈信息；
根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；及
根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
一种人脸识别数据处理装置，包括：
接收模块，用于通过移动设备接收人脸识别错误反馈信息；
调整模块，用于根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；
识别模块，用于根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
一种移动设备，包括存储器和处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行以下步骤：
通过移动设备接收人脸识别错误反馈信息；
根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；及
根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
一种非易失性计算机可读存储介质，其上存储有计算机可读指令，所述计算机可读指令被处理器执行时，使得所述处理器执行以下步骤：
通过移动设备接收人脸识别错误反馈信息；
根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；及
根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
上述人脸识别数据处理方法、装置、移动设备和非易失性计算机可读存储介质，通过移动设备接收人脸识别错误反馈信息，根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，及根据调整识别控制阈值后的当前人脸识别算法进行人脸识别，当人脸识别错误时，通过人脸识别反馈信息自动智能调整人脸识别算法控制阈值，从而减少下次误识别的现象，提高人脸识别准确率。
附图说明
为了更好地描述和说明这里公开的那些申请的实施例和/或示例，可以参考一幅或多 幅附图。用于描述附图的附加细节或示例不应当被认为是对所公开的发明、目前描述的实施例和/或示例以及目前理解的这些申请的最佳模式中的任何一者的范围的限制。
图1为一个实施例中人脸识别数据处理方法的应用环境示意图。
图2为一个实施例中人脸识别数据处理方法的流程图。
图3为一个实施例中锁屏界面示意图。
图4为一个实施例中解锁成功后的桌面示意图。
图5为一个实施例中人脸识别数据处理方法的流程图。
图6为一个实施例中进行人脸识别的流程图。
图7为一个实施例中调整识别控制阈值的流程图。
图8为一个具体的实施例中人脸识别数据处理方法的流程图。
图9为一个实施例中人脸识别数据处理装置的结构框图。
图10为一个实施例中移动设备的内部结构示意图。
图11为一个实施例中与本实施例相关的移动设备的部分结构的框图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
可以理解，本申请实施例中所使用的术语“第一”、“第二”等可在本文中用于描述各种元件，但这些元件不受这些术语限制。这些术语仅用于将第一个元件与另一个元件区分。举例来说，在不脱离本申请的范围的情况下，可以将第一控件称为第二控件，第一控件和第二控件两者都是控件，但其不是同一控件。
图1为一个实施例中人脸识别数据处理方法的应用环境图。如图1所示，该应用环境包括移动设备110。在移动设备110上设置有摄像头，可采集人脸图像，对采集的人脸图像进行人脸识别，移动设备110上设置有反馈信息接收按键，可为实体按键或虚拟按键，通过反馈信息接收按键接收人脸识别错误反馈信息，根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。移动设备110可为智能手机、平板电脑、穿戴式设备、个人数字助理等。
图2为一个实施例中人脸识别数据处理方法的流程图。如图2所示，一种人脸识别数据处理方法，以应用于图1中的移动设备上为例进行说明，具体包括：
操作202，移动设备接收人脸识别错误反馈信息。
其中，人脸识别错误反馈信息是用于描述人脸识别错误的反馈信息，人脸识别错误包括多种类型的错误，如错误接受识别、错误拒绝识别。错误接受识别是指将错误的人脸进行算法识别后识别为预设人脸图像，如移动设备中录入的预设人脸图像为用户A人脸图像，当采集的人脸图像是用户B的人脸图像时，算法误将用户B的人脸图像识别为用户A人脸图像，导致人脸验证误成功。错误拒绝识别是指将正确的人脸进行算法识别后识别为非预设人脸图像，如移动设备中录入的预设人脸图像为用户A人脸图像，当采集的人脸图像是用户A的人脸图像时，算法误将用户A的人脸图像识别为其他用户的人脸图像，导致人脸验证不成功。
具体地，通过作用于移动设备的操作接收人脸识别错误反馈信息，其中操作可以为直接或间接作用于移动设备的操作，直接作用于移动设备的操作可以为对实体按键或屏幕的虚拟按键进行的操作。间接作用于移动设备的操作可以为手势操作，眼球操作等。可通过不同的按键反馈不同类型的人脸识别错误反馈信息，或通过不同的手势、眼球运动反馈不同类型的人脸识别错误反馈信息。可通过UI界面将人脸识别错误反馈信息通知到系统层。
操作204，根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值。
其中，识别控制阈值是用于控制人脸识别准确度的阈值，不同的人脸识别算法对应的不同类型的识别控制阈值。如对于2D人脸识别算法，识别控制阈值可包括比对阈值、假体阈值等。在一个实施例中，当人脸识别算法为网络模型算法时，识别控制阈值可以为网络模型的参数。识别控制阈值可以为2D人脸识别算法或3D人脸识别算法中的识别控制阈值。
针对不同的人脸识别错误反馈信息，需要通过不同的调整方式对识别控制阈值进行调整，具体的调整算法可根据人脸识别错误反馈信息的内容和识别控制阈值的含义确定。调整的原则是，当人脸识别错误反馈信息为错误接受识别反馈信息时，需要使得人脸识别算法更为严格，避免将其他用户的人脸误识别为预设用户的人脸。当人脸识别错误反馈信息为错误拒绝识别反馈信息时，需要使得人脸识别算法更为宽松，避免预设用户的正确的人脸在环境或其他因素的影响下也不能识别成功，导致人脸无法验证成功。
操作206，根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
其中，系统根据人脸识别错误反馈信息确认本次人脸识别为误识别，从而调整阀值来降低下次或以后的错误接受识别和/或错误拒绝识别。其中人脸识别的结果可用于解锁、支付、启动或关闭应用程序、权限设置等需要进行人脸识别验证的操作，根据调整识别控制阈值后的当前人脸识别算法进行人脸识别提高了人脸识别的准确率，从而保证根据人脸识别结果进行的各项操作的权限的有效性。
上述人脸识别数据处理方法，通过移动设备接收人脸识别错误反馈信息，根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，根据调整识别控制阈值后的当前人脸识别算法进行人脸识别根据调整识别控制阈值后的当前人脸识别算法进行人脸识别，可以在人脸识别错误时对人脸识别算法进行动态智能的自动调整，可以避免下次继续误识别，提高人脸识别的准确率。
在一个实施例中，移动设备接收人脸识别错误反馈信息，包括：通过显示在移动设备屏幕的控件接收人脸识别错误反馈信息，不同的控件传递不同的人脸识别错误反馈信息。
具体地，当人脸识别错误时，在移动设备屏幕的控件接收用户的操作，不同的控件对应不同的触发事件，如第一控件对应的触发事件为传递错误接受识别反馈信息，第二控件对应的触发事件为传递错误拒绝识别反馈信息。各个控件的显示位置可以不同，如对于传递解锁失败这种错误拒绝识别反馈信息的控件，可显示在屏幕锁定界面上，便于解锁失败的情况下也能操作控件传递错误拒绝识别反馈信息。对于传递错误解锁成功这种错误接受识别反馈信息的控件，可显示在解锁成功后的界面上，便于错误解锁成功的情况下快速对控件进行操作从而传递错误接受识别反馈信息。
如图3所示，为一个实施例中显示在屏幕锁定界面上的第一控件示意图，第一控件208用于在本人解锁失败时，向人脸识别算法模块传递错误拒绝识别反馈信息。如图4所示，为一个实施例中显示在解锁成功后的桌面上的第二控件示意图，第二控件210用于非本人解锁成功时，向人脸识别算法模块传递错误接受识别反馈信息。在一个实施例中，第二控件处于隐藏的状态，可通过作用于桌面的下拉操作触发第二控件的显示。
本实施例中，通过显示在UI界面的不同的控件传递不同的人脸识别错误反馈信息至系统，通过作用于控件的操作快速反馈人脸识别错误反馈信息，提高了人脸识别错误反馈信息反馈的便利性与及时性。
在一个实施例中，通过显示在移动设备屏幕的控件接收人脸识别错误反馈信息，包括：响应作用于显示在移动设备屏幕的第一控件的操作，生成错误接受识别反馈信息,将错误接受识别反馈信息作为人脸识别错误反馈信息，接收第一控件传递的错误接受识别反馈信息。
其中，作用于显示在移动设备屏幕的第一控件的操作可以是点击、触摸时长超过预设阈值、滑动操作等任意自定义的操作。其中第一控件可以是专门用于反馈错误接受识别反 馈信息的控件，也可以是还具有其它功能的控件。当第一控件具备多种功能时，通过与接收人脸识别错误反馈信息对应的预设操作触发生成错误接受识别反馈信息。如在第一控件上向左滑动表示发生人脸错误接受识别，则当第一控件识别到向左滑动的操作时，生成错误接受识别反馈信息。
第一控件对应预设的反馈值，预设的反馈值表示错误接受识别反馈信息，如传递值1时表示错误接受识别反馈信息，则第一控件识到预设操作时，生成值为1的反馈值，将反馈值传递至人脸识别系统，人脸识别系统通过反馈值的大小判断识别控制阈值的调节方向。
在一个实施例中，响应作用于显示在所述移动设备屏幕的第一控件的操作,生成错误接受识别反馈信息，包括：移动设备获取当前人脸图像，当当前人脸图像与预设解锁人脸图像实际不匹配，当前人脸识别算法将当前人脸图像误识别为预设解锁人脸，触发移动设备的屏幕解锁成功时，响应作用于第一控件的操作生成错误接受识别反馈信息。
具体地，当前人脸识别算法将当前人脸图像误识别为预设解锁人脸，触发移动设备的屏幕解锁成功，说明本次解锁是由于人脸识算法误将其他用户的人脸识别为移动设备录入的预设解锁人脸，导致的误解锁，如录入用户A的人脸，手机却被用户B解锁。则用户A可对第一控件进行操作,第一控件根据用户A的操作生成错误接受识别反馈信息。
本实施例中，当解锁误成功时，可通过第一控件快速反馈错误接受识别反馈信息。可以理解的是，应用场景可不限定于解锁，如果是其他的应用场景，如通过人脸支付，通过人脸进入应用等，可分别对应不同的与应用场景匹配的控件快速反馈错误接受识别反馈信息。
在一个实施例中，通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息，包括：响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，将错误拒绝识别反馈信息作为人脸识别错误反馈信息，接收所述第二控件传递的错误拒绝识别反馈信息。
具体地，其中，作用于显示在移动设备屏幕的第二控件的操作可以是点击、触摸时长超过预设阈值、滑动操作等任意自定义的操作。其中第二控件可以是专门用于反馈错误拒绝识别反馈信息的控件，也可以是还具有其它功能的控件。当第二控件具备多种功能时，通过与接收人脸识别错误拒绝识别反馈信息对应的预设操作触发生成错误拒绝识别反馈信息。如在第二控件上向右滑动表示发生人脸错误拒绝识别，则当第二控件识别到向右滑动的操作时，生成错误拒绝识别反馈信息。
在一个实施例中，反馈错误拒绝识别反馈信息和错误接受识别反馈信息的控件为同一个控件，通过对这个控件的不同操作反馈不同类型的信息。如向左滑动则反馈错误接受识别反馈信息，向右滑动则反馈错误拒绝识别反馈信息。通过将控件合一，可节省控件的生成，节省资源。
第二控件对应预设的反馈值，预设的反馈值表示错误拒绝识别反馈信息，如传递值2时表示错误拒绝识别反馈信息，则第一控件识到预设操作时，生成值为2的反馈值，将反馈值传递至人脸识别系统，人脸识别系统通过反馈值的大小判断识别控制阈值的调节方向。
在一个实施例中，响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，包括：移动设备获取当前人脸图像，当当前人脸图像与预设解锁人脸图像实际匹配，当前人脸识别算法将当前人脸图像误识别为非预设解锁人脸，使得移动设备的屏幕解锁失败时，响应作用于第二控件的操作生成错误拒绝识别反馈信息。
具体地，当前人脸识别算法将当前人脸图像误识别为非预设解锁人脸，使得移动设备的屏幕解锁失败，说明本次解锁是由于人脸识算法不能识别移动设备录入的预设解锁人脸，导致的解锁失败，如录入用户A的人脸，用户A却无法解锁。则用户A可对第二控件进行操作,第二控件根据用户A的操作生成错误接受拒绝反馈信息。
本实施例中，当正确人脸解锁失败时，可通过第二控件快速反馈错误拒绝识别反馈信 息。可以理解的是，应用场景可不限定于解锁，如果是其他的应用场景，如通过人脸支付，通过人脸进入应用等，可分别对应不同的与应用场景匹配的控件快速反馈错误拒绝识别反馈信息。
在一个实施例中，当当前人脸图像与预设解锁人脸图像实际匹配，获取当前人脸识别算法将当前人脸图像识别为预设解锁人脸使用的时长，根据时长确定目标控件，响应作用于目标控件的操作生成错误拒绝识别反馈信息。
具体地，当当前人脸图像与预设解锁人脸图像实际匹配，如录入用户A的人脸，当前人脸图像为用户A的人脸，当前人脸识别算法在第一时长内将当前人脸图像识别非预设解锁人脸，说明在第一时长内当前人脸识别算法无法识别用户A的人脸，在第二时刻将当前人脸图像识别为预设解锁人脸，则第二时刻与起始时刻的差值为当前人脸识别算法将当前人脸图像识别为预设解锁人脸使用的时长。时长越短，说明当前人脸识别算法越灵敏，识别度越高。根据识别时长确定目标控件，不同的目标控件用于指示当前人脸识别算法对识别控制阈值进行不同幅度的调整。
本实施例中，通过识别时长使得识别控制阈值进行不同的调整，进一步提高了识别控制阈值调整的精确度。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；操作204，包括：当人脸识别错误反馈信息为错误接受识别反馈信息时，提高比对阈值和/或降低假体阈值。
具体地，2D人脸识别算法主要包含两个阀值，比对阀值和假体阀值。比对阀值用来控制录入的人脸和尝试验证的人脸的相似度门槛，门槛越低越容易验证成功，如验证人脸是为了解锁，则门槛越低越容易解锁。当人脸识别错误反馈信息为错误接受识别反馈信息时，说明当前人脸识别算法容易将其他的人脸识别为预设人脸，说明相似度门槛偏低，需要提高比对阈值。比对阈值用来是判定尝试解锁的人脸的真假度的门槛，门槛越高越容易将假人脸识别为真人脸。当人脸识别错误反馈信息为错误接受识别反馈信息时，说明当前人脸识别算法容易将假人脸识别为真人脸，说明真假度的门槛偏高，需要降低假体阈值。可以理解的是，提高比对阈值和降低假体阈值可以择一调整或两者同时调整，调整的幅度可与当前人脸识别算法相匹配或自定义。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；操作204，包括：当人脸识别错误反馈信息为错误拒绝识别反馈信息时，降低比对阈值和/或提高假体阈值。
具体地，当人脸识别错误反馈信息为错误拒绝识别反馈信息时，说明当前人脸识别算法容易将正确的录入人脸识别失败，如移动设备录入A用户的人脸，但是当A用户受到光照变化、头发遮挡，人脸角度偏转等原因影响时，即使是A用户人脸，也无法识别，说明相似度门槛偏高，需要降低比对阈值。当人脸识别错误反馈信息为错误拒绝识别反馈信息时，说明当前人脸识别算法即使是真人脸也容易误识别为假人脸，说明真假度的门槛偏低，需要提高假体阈值。可以理解的是，降低比对阈值和提高假体阈值可以择一调整或两者同时调整，调整的幅度可与当前人脸识别算法相匹配或自定义。
在一个实施例中，操作204之前，如图5所示，还包括：
操作302，获取当前人脸识别对应的当前识别场景，获取人脸识别错误反馈信息对应的反馈识别场景。
操作304，判断当当前识别场景与反馈识别场景匹配时，如果是，则进入操作204，如果不匹配，则不作调整。
具体地，可以通过背景图案、光照度、亮度、遮挡度、远近度、人脸清晰度等作为不同场景的判断因素，通过对当前采集的图像进行分析得到当前人脸识别对应的当前场景判断因素，与反馈识别场景对应的反馈场景判断因素进行对比，判断当前识别场景与反馈识 别场景是否匹配，具体的匹配算法可定义。在一个实施例中，当超过预设数量的判断因素的相似度超过预设阈值时，判定当前识别场景与反馈识别场景匹配。
也可直接将当前相机出帧的图片与反馈识别场景出帧的图片通过图像分析算法进行判断识别是不是同一或者相似场景，如果是同一或者相似场景，则当前识别场景和反馈识别场景匹配。
本实施例中，只有当前识别场景和反馈识别场景匹配才调整识别控制阀值，进一步提高了识别控制阀值调整的精确性，避免错误的调节。
在一个实施例中，如图6所示，方法还包括：
操作402，获取人脸识别错误反馈信息对应的反馈识别场景，建立反馈识别场景与调整后的识别控制阈值的匹配关系。
具体地，获取用户反馈的识别错误反馈信息对应的不同场景，获取在不同场景下调整后的识别控制阈值，从而建立不同的反馈识别场景与调整后的识别控制阈值的匹配关系。如在场景A与场景B下进行人脸识别时，都接收到了人脸识别错误反馈信息，从而场景A与场景B为不同的反馈识别场景。在场景A下对识别控制阈值进行第一调整后得到调整后的目标第一识别控制阈值，在场景B下对识别控制阈值进行第二调整后得到调整后的目标第二识别控制阈值，分别建立场景A与目标第一识别控制阈值的匹配关系，建立场景B与目标第二识别控制阈值的匹配关系。
操作404，获取当前人脸识别对应的当前识别场景，确定与当前识别场景匹配的目标反馈识别场景。
具体地，通过对当前采集图像的分析得到当前人脸识别对应的当前识别场景，与匹配关系中的各个反馈识别场景进行比对，得到与当前识别场景匹配的目标反馈识别场景，目标反馈识别场景为与当前识别场景相同或相似的场景，具体的匹配算法可自定义。
操作406，根据匹配关系获取目标反馈识别场景对应的目标识别控制阈值。
具体地，由于各个反馈识别场景对应的识别控制阈值，都是经过调整后的与场景匹配的阈值，保证了在不同的场景下使用不同的识别控制阈值，提高不同场景下人脸识别的准确性。通过获取目标反馈识别场景对应的目标识别控制阈值，由于目标反馈识别场景是与当前识别场景匹配的场景，所以目标识别控制阈值也更适用于当前识别场景。
操作408，根据目标识别控制阈值对应的当前人脸识别算法进行人脸识别。
具体地，根据目标识别控制阈值得到当前人脸识别算法，进行人脸识别，保证了当前识别场景下使用的是之前调整过的与当前识别场景匹配的控制阈值，从而提高当前识别场景下人脸识别的准确性。
在一个实施例中，如图7所示，操作204包括：
操作204a，获取当前人脸识别算法对应的初始识别控制阈值。
具体地，初始识别控制阈值是未进行调整的识别控制阈值，如当前人脸识别算法对应的初始模型对应的识别控制阈值，可自定义的识别控制阈值。
操作204b,获取识别控制阈值调整范围，根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，使得调整后的当前识别控制阈值与初始识别控制阈值的差距在识别控制阈值调整范围内。
具体地，识别控制阈值调整范围将识别控制阈值控制在一定的范围内，如一个实施例中初始识别控制阈值为10，则识别控制阈值调整范围为[-10-10]，则说明识别控制阈值只能在10上下调整10的幅度，如果识别控制阈值在调整后超出了识别控制阈值调整范围，则说明调整过度。因为阀值调整要有一定范围，不能太广，否则会严重影响错误接受率和错误拒绝率，错误接受率即错误的人脸进行算法识别后验证成功的概率，错误拒绝率即正确的人脸进行算法识别后验证失败的概率。阀值的调整允许有轻微波动，但需要在合适的情况下还原，因为从大数据来看，默认阀值的错误接受率和错误拒绝率属于一个比较均衡 的状态，长时间打破这一平衡有可能在其他场景下又引入新的问题。
本实施例中，使得调整后的当前识别控制阈值与初始识别控制阈值的差距在识别控制阈值调整范围内，保证了人脸识别的稳定性。
如图8所示，下面结合一个具体的示例对人脸识别数据处理方法进行详细的描述。首先假设移动设备在不同的界面显示不同的反馈控件，移动设备处于锁屏状态，在锁屏状态对应的界面显示第一控件，第一控件上显示字符“本人无法解锁？”，用于提示用户在本人无法解锁时，对第一控件进行操作以反馈错误拒绝识别反馈信息。
操作502，移动设备通过第一控件接收反馈错误拒绝识别反馈信息，用户通过UI界面将错误拒绝识别反馈信息通知到系统层。比如，录入用户A的人脸，用户A无法解锁，则第一控件接收用户A的点击操作，第一控件把错误拒绝识别反馈信息传递到系统层的人脸识别算法模块。
操作504，当系统层接收到错误拒绝识别反馈信息时，降低比对阈值、提高假体阈值，并且保证调整后的比对阈值和假体阈值分别在对应的调整范围内。
操作506，当用户B解锁成功时，在解锁成功的界面显示第二控件，第一控件上显示字符“不是本人解锁？”，用于提示用户在不是本人解锁时，对第一控件进行操作以反馈错误接受识别反馈信息。移动设备通过第二控件接收反馈错误接受识别反馈信息，比如，录入用户A的人脸，用户B的人脸通过验证，屏幕解锁成功。
操作508，当系统层接收到错误接受识别反馈信息时，提高比对阈值、降低假体阈值，并且保证调整后的比对阈值和假体阈值分别在对应的调整范围内。
本实施例中，在人脸误识别后，可以通过对控件的操作给系统反馈算法的缺陷，从而调整识别控制阀值，通过反馈信息自动调整阀值，下次同一用户在同一场景下再使用人脸解锁，可减少误识别的现象，提高人脸识别的准确率。且整个调整过程无需更新软件版本，降低了开发成本。
应该理解的是，虽然图2、图5-图8的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图2、图5-图8中的至少一部分步骤可以包括多个子步骤或者多个阶段，这些子步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些子步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
本申请实施例还提供了一种移动设备。该移动设备，包括存储器和处理器，该存储器中储存有计算机可读指令，该计算机可读指令被该处理器执行时，使得该处理器执行以下步骤：通过移动设备接收人脸识别错误反馈信息；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
在一个实施例中，移动设备接收人脸识别错误反馈信息，包括：通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息，不同的控件传递不同的人脸识别错误反馈信息。
在一个实施例中，通过显示在移动设备屏幕的控件接收人脸识别错误反馈信息，包括：响应作用于显示在移动设备屏幕的第一控件的操作，生成错误接受识别反馈信息,将错误接受识别反馈信息作为人脸识别错误反馈信息；接收第一控件传递的错误接受识别反馈信息。
在一个实施例中，响应作用于显示在移动设备屏幕的第一控件的操作,生成错误接受识别反馈信息，包括：移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际不匹配，当前人脸识别算法将所述当前人脸图像误识别为预设解锁人脸，触发所述移动设备的屏幕解锁成功时，响应作用于第一控件的操作生成错误接受识别反馈信息。
在一个实施例中，通过显示在移动设备屏幕的控件接收人脸识别错误反馈信息，包括：响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，将错误拒绝识别反馈信息作为人脸识别错误反馈信息；接收第二控件传递的错误拒绝识别反馈信息。
在一个实施例中，响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，包括：通过移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际匹配，当前人脸识别算法将当前人脸图像误识别为非预设解锁人脸，使得移动设备的屏幕解锁失败时，响应作用于第二控件的操作生成错误拒绝识别反馈信息。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：当人脸识别错误反馈信息为错误接受识别反馈信息时，提高比对阈值和/或降低所述假体阈值。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际不匹配，当前人脸识别算法将所述当前人脸图像误识别为预设解锁人脸，触发移动设备的屏幕解锁成功时，响应作用于第一控件的操作生成错误接受识别反馈信息。
在一个实施例中，通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息，包括：响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，将错误拒绝识别反馈信息作为人脸识别错误反馈信息；接收第二控件传递的错误拒绝识别反馈信息。
在一个实施例中，响应作用于显示在所述移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，包括：通过移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际匹配，所述当前人脸识别算法将所述当前人脸图像误识别为非预设解锁人脸，使得所述移动设备的屏幕解锁失败时，响应作用于所述第二控件的操作生成所述错误拒绝识别反馈信息。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：当人脸识别错误反馈信息为错误接受识别反馈信息时，提高比对阈值和/或降低假体阈值。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：当人脸识别错误反馈信息为错误拒绝识别反馈信息时，降低比对阈值和/或提高所述假体阈值。
在一个实施例中，处理器执行以下步骤：获取当前人脸识别对应的当前识别场景，获取人脸识别错误反馈信息对应的反馈识别场景；当当前识别场景与反馈识别场景匹配时，进入根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值的步骤。
在一个实施例中，处理器执行以下步骤：获取人脸识别错误反馈信息对应的反馈识别场景，建立反馈识别场景与所述调整后的识别控制阈值的匹配关系；获取当前人脸识别对应的当前识别场景，确定与当前识别场景匹配的目标反馈识别场景；根据匹配关系获取目标反馈识别场景对应的目标识别控制阈值；根据目标识别控制阈值对应的当前人脸识别算法进行人脸识别。
在一个实施例中，根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：获取当前人脸识别算法对应的初始识别控制阈值；获取识别控制阈值调整 范围；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，使得调整后的当前识别控制阈值与初始识别控制阈值的差距在识别控制阈值调整范围内。
图7为一个实施例中人脸识别数据处理装置的结构框图。如图5所示，一种人脸识别数据处理装置，包括接收模块602、调整模块604、识别模块606。其中：
接收模块602用于通过移动设备接收人脸识别错误反馈信息。
调整模块604用于根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值。
识别模块606用于根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
在一个实施例中，接收模块602还用于通过显示在移动设备屏幕的控件接收人脸识别错误反馈信息，不同的控件传递不同的人脸识别错误反馈信息。
在一个实施例中，接收模块602还用于响应作用于显示在移动设备屏幕的第一控件的操作，生成错误接受识别反馈信息,将错误接受识别反馈信息作为人脸识别错误反馈信息，接收第一控件传递的错误接受识别反馈信息。
在一个实施例中，接收模块602还用于通过移动设备获取当前人脸图像，当当前人脸图像与预设解锁人脸图像实际不匹配，当前人脸识别算法将当前人脸图像误识别为预设解锁人脸，触发移动设备的屏幕解锁成功时，响应作用于第一控件的操作生成错误接受识别反馈信息。
在一个实施例中，接收模块602还用于响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，将所述错误拒绝识别反馈信息作为人脸识别错误反馈信息；接收第二控件传递的错误拒绝识别反馈信息。
在一个实施例中，接收模块602还用于通过移动设备获取当前人脸图像，当当前人脸图像与预设解锁人脸图像实际匹配，当前人脸识别算法将当前人脸图像误识别为非预设解锁人脸，使得移动设备的屏幕解锁失败时，响应作用于第二控件的操作生成错误拒绝识别反馈信息。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；调整模块604还用于当人脸识别错误反馈信息为错误接受识别反馈信息时，提高比对阈值和/或降低假体阈值。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；调整模块604还用于当所述人脸识别错误反馈信息为错误拒绝识别反馈信息时，降低所述比对阈值和/或提高所述假体阈值。
在一个实施例中，装置还包括：
场景匹配模块，用于获取当前人脸识别对应的当前识别场景，获取人脸识别错误反馈信息对应的反馈识别场景，当当前识别场景与反馈识别场景匹配时，进入调整模块604。
在一个实施例中，装置还包括：
匹配关系模块，用于获取人脸识别错误反馈信息对应的反馈识别场景，建立反馈识别场景与调整后的识别控制阈值的匹配关系。
场景调整模块，用于获取当前人脸识别对应的当前识别场景，确定与当前识别场景匹配的目标反馈识别场景，根据匹配关系获取目标反馈识别场景对应的目标识别控制阈值。
识别模块606还用于根据目标识别控制阈值对应的当前人脸识别算法进行人脸识别。
在一个实施例中，调整模块604还用于获取所述当前人脸识别算法对应的初始识别控制阈值；获取识别控制阈值调整范围；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，使得调整后的当前识别控制阈值与初始识别控制阈值的差距在所述识别控制阈值调整范围内。
本申请实施例中提供的人脸识别数据处理装置中的各个模块的实现可为计算机可读指令的形式。该计算机可读指令可在终端或服务器上运行。该计算机可读指令构成的程序 模块可存储在终端或服务器的存储器上。该计算机可读指令被处理器执行时，实现本申请实施例中所描述方法的步骤。
图10为一个实施例中移动设备的内部结构示意图。如图10所示，该移动设备包括通过系统总线连接的处理器和存储器。其中，该处理器用于提供计算和控制能力，支撑整个电子设备的运行。存储器可包括非易失性存储介质及内存储器。非易失性存储介质存储有操作系统和计算机可读指令。该计算机可读指令可被处理器所执行，以用于实现以下各个实施例所提供的一种人脸识别数据处理方法。内存储器为非易失性存储介质中的操作系统计算机可读指令提供高速缓存的运行环境。该移动设备可以是手机、平板电脑或者个人数字助理或穿戴式设备等。
本申请实施例还提供了一种移动设备。该移动设备可以为包括手机、平板电脑、PDA(Personal Digital Assistant，个人数字助理)、POS(Point of Sales，销售终端)、车载电脑、穿戴式设备等任意终端设备，以移动设备为手机为例。
本申请实施例还提供一种移动设备。上述移动设备中包括图像处理电路，图像处理电路可以利用硬件和/或软件组件实现，可包括定义ISP(Image Signal Processing，图像信号处理)管线的各种处理单元。图11为一个实施例中图像处理电路的示意图。如图11所示，为便于说明，仅示出与本申请实施例相关的图像处理技术的各个方面。
如图11所示，图像处理电路包括ISP处理器740和控制逻辑器750。成像设备710捕捉的图像数据首先由ISP处理器740处理，ISP处理器740对图像数据进行分析以捕捉可用于确定和/或成像设备710的一个或多个控制参数的图像统计信息。成像设备710可包括具有一个或多个透镜712和图像传感器714的照相机。图像传感器714可包括色彩滤镜阵列(如Bayer滤镜)，图像传感器714可获取用图像传感器714的每个成像像素捕捉的光强度和波长信息，并提供可由ISP处理器740处理的一组原始图像数据。传感器720(如陀螺仪)可基于传感器720接口类型把采集的图像处理的参数(如防抖参数)提供给ISP处理器740。传感器720接口可以利用SMIA(Standard Mobile Imaging Architecture，标准移动成像架构)接口、其它串行或并行照相机接口或上述接口的组合。
此外，图像传感器714也可将原始图像数据发送给传感器720，传感器720可基于传感器720接口类型把原始图像数据提供给ISP处理器740，或者传感器720将原始图像数据存储到图像存储器730中。
ISP处理器740按多种格式逐个像素地处理原始图像数据。例如，每个图像像素可具有8、10、12或14比特的位深度，ISP处理器740可对原始图像数据进行一个或多个图像处理操作、收集关于图像数据的统计信息。其中，图像处理操作可按相同或不同的位深度精度进行。
ISP处理器740还可从图像存储器730接收图像数据。例如，传感器720接口将原始图像数据发送给图像存储器730，图像存储器730中的原始图像数据再提供给ISP处理器740以供处理。图像存储器730可为存储器装置的一部分、存储设备、或电子设备内的独立的专用存储器，并可包括DMA(Direct Memory Access，直接直接存储器存取)特征。
当接收到来自图像传感器714接口或来自传感器720接口或来自图像存储器730的原始图像数据时，ISP处理器740可进行一个或多个图像处理操作，如时域滤波。处理后的图像数据可发送给图像存储器730，以便在被显示之前进行另外的处理。ISP处理器740从图像存储器730接收处理数据，并对所述处理数据进行原始域中以及RGB和YCbCr颜色空间中的图像数据处理。ISP处理器740处理后的图像数据可输出给显示器770，以供用户观看和/或由图形引擎或GPU(Graphics Processing Unit，图形处理器)进一步处理。此外，ISP处理器740的输出还可发送给图像存储器730，且显示器770可从图像存储器730读取图像数据。在一个实施例中，图像存储器730可被配置为实现一个或多个帧缓冲器。此外，ISP处理器740的输出可发送给编码器/解码器760，以便编码/解码图像数据。 编码的图像数据可被保存，并在显示于显示器770设备上之前解压缩。编码器/解码器760可由CPU或GPU或协处理器实现。
ISP处理器740确定的统计数据可发送给控制逻辑器750单元。例如，统计数据可包括自动曝光、自动白平衡、自动聚焦、闪烁检测、黑电平补偿、透镜712阴影校正等图像传感器714统计信息。控制逻辑器750可包括执行一个或多个例程(如固件)的处理器和/或微控制器，一个或多个例程可根据接收的统计数据，确定成像设备710的控制参数及ISP处理器740的控制参数。例如，成像设备710的控制参数可包括传感器720控制参数(例如增益、曝光控制的积分时间、防抖参数等)、照相机闪光控制参数、透镜712控制参数(例如聚焦或变焦用焦距)、或这些参数的组合。ISP控制参数可包括用于自动白平衡和颜色调整(例如，在RGB处理期间)的增益水平和色彩校正矩阵，以及透镜712阴影校正参数。
在本申请实施例中，该移动设备所包括的ISP处理器740执行存储在存储器上的计算机可读指令时实现人脸识别数据处理方法的步骤。
本申请实施例还提供了一种计算机可读存储介质。一个或多个包含计算机可读指令的非易失性计算机可读存储介质，该计算机可读指令被处理器执行时实现以下步骤：通过移动设备接收人脸识别错误反馈信息；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
在一个实施例中，移动设备接收人脸识别错误反馈信息，包括：通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息，不同的控件传递不同的人脸识别错误反馈信息。
在一个实施例中，通过显示在移动设备屏幕的控件接收人脸识别错误反馈信息，包括：响应作用于显示在移动设备屏幕的第一控件的操作，生成错误接受识别反馈信息,将错误接受识别反馈信息作为人脸识别错误反馈信息；接收第一控件传递的错误接受识别反馈信息。
在一个实施例中，响应作用于显示在移动设备屏幕的第一控件的操作,生成错误接受识别反馈信息，包括：移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际不匹配，当前人脸识别算法将所述当前人脸图像误识别为预设解锁人脸，触发所述移动设备的屏幕解锁成功时，响应作用于第一控件的操作生成错误接受识别反馈信息。
在一个实施例中，通过显示在移动设备屏幕的控件接收人脸识别错误反馈信息，包括：响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，将错误拒绝识别反馈信息作为人脸识别错误反馈信息；接收第二控件传递的错误拒绝识别反馈信息。
在一个实施例中，响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，包括：通过移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际匹配，当前人脸识别算法将当前人脸图像误识别为非预设解锁人脸，使得移动设备的屏幕解锁失败时，响应作用于第二控件的操作生成错误拒绝识别反馈信息。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：当人脸识别错误反馈信息为错误接受识别反馈信息时，提高比对阈值和/或降低所述假体阈值。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际不匹配，当前人脸识别算法将所述当前人脸图像误识别为预设解锁人脸，触发移动设备的屏幕解锁成功时，响应作用于第一控件的操作生成错误接受识别反馈信息。
在一个实施例中，通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息， 包括：响应作用于显示在移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，将错误拒绝识别反馈信息作为人脸识别错误反馈信息；接收第二控件传递的错误拒绝识别反馈信息。
在一个实施例中，响应作用于显示在所述移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，包括：通过移动设备获取当前人脸图像；当当前人脸图像与预设解锁人脸图像实际匹配，所述当前人脸识别算法将所述当前人脸图像误识别为非预设解锁人脸，使得所述移动设备的屏幕解锁失败时，响应作用于所述第二控件的操作生成所述错误拒绝识别反馈信息。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：当人脸识别错误反馈信息为错误接受识别反馈信息时，提高比对阈值和/或降低假体阈值。
在一个实施例中，识别控制阈值包括比对阈值、假体阈值中的至少一种；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：当人脸识别错误反馈信息为错误拒绝识别反馈信息时，降低比对阈值和/或提高所述假体阈值。
在一个实施例中，该计算机可读指令被处理器执行：获取当前人脸识别对应的当前识别场景，获取人脸识别错误反馈信息对应的反馈识别场景；当当前识别场景与反馈识别场景匹配时，进入根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值的步骤。
在一个实施例中，该计算机可读指令被处理器执行：获取人脸识别错误反馈信息对应的反馈识别场景，建立反馈识别场景与所述调整后的识别控制阈值的匹配关系；获取当前人脸识别对应的当前识别场景，确定与当前识别场景匹配的目标反馈识别场景；根据匹配关系获取目标反馈识别场景对应的目标识别控制阈值；根据目标识别控制阈值对应的当前人脸识别算法进行人脸识别。
在一个实施例中，根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：获取当前人脸识别算法对应的初始识别控制阈值；获取识别控制阈值调整范围；根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，使得调整后的当前识别控制阈值与初始识别控制阈值的差距在识别控制阈值调整范围内。
本申请实施例还提供一种计算机可读指令产品。一种包含指令的计算机可读指令产品，当其在计算机上运行时，使得计算机执行人脸识别数据处理方法。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的程序可存储于一非易失性计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所使用的对存储器、存储、数据库或其它介质的任何引用可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)，它用作外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDR SDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对本申请专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (14)

  1. 一种人脸识别数据处理方法，其特征在于，包括：
    移动设备接收人脸识别错误反馈信息；
    根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；及
    根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
  2. 根据权利要求1所述的方法，其特征在于，所述移动设备接收人脸识别错误反馈信息，包括：
    通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息，不同的控件传递不同的人脸识别错误反馈信息。
  3. 根据权利要求2所述的方法，其特征在于，所述通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息，包括：
    响应作用于显示在所述移动设备屏幕的第一控件的操作，生成错误接受识别反馈信息,将所述错误接受识别反馈信息作为所述人脸识别错误反馈信息；
    接收所述第一控件传递的所述错误接受识别反馈信息。
  4. 根据权利要求3所述的方法，其特征在于，所述响应作用于显示在所述移动设备屏幕的第一控件的操作,生成错误接受识别反馈信息，包括：
    所述移动设备获取当前人脸图像；
    当所述当前人脸图像与预设解锁人脸图像实际不匹配，所述当前人脸识别算法将所述当前人脸图像误识别为预设解锁人脸，触发所述移动设备的屏幕解锁成功时，响应作用于所述第一控件的操作生成所述错误接受识别反馈信息。
  5. 根据权利要求2所述的方法，其特征在于，所述通过显示在所述移动设备屏幕的控件接收人脸识别错误反馈信息，包括：
    响应作用于显示在所述移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，将所述错误拒绝识别反馈信息作为所述人脸识别错误反馈信息；
    接收所述第二控件传递的所述错误拒绝识别反馈信息。
  6. 根据权利要求5所述的方法，其特征在于，所述响应作用于显示在所述移动设备屏幕的第二控件的操作,生成错误拒绝识别反馈信息，包括：
    所述移动设备获取当前人脸图像；
    当所述当前人脸图像与预设解锁人脸图像实际匹配，所述当前人脸识别算法将所述当前人脸图像误识别为非预设解锁人脸，使得所述移动设备的屏幕解锁失败时，响应作用于所述第二控件的操作生成所述错误拒绝识别反馈信息。
  7. 根据权利要求1所述的方法，其特征在于，所述识别控制阈值包括比对阈值、假体阈值中的至少一种；所述根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：
    当所述人脸识别错误反馈信息为错误接受识别反馈信息时，提高所述比对阈值和/或降低所述假体阈值。
  8. 根据权利要求1所述的方法，其特征在于，所述识别控制阈值包括比对阈值、假体阈值中的至少一种；所述根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：
    当所述人脸识别错误反馈信息为错误拒绝识别反馈信息时，降低所述比对阈值和/或提高所述假体阈值。
  9. 根据权利要求1所述的方法，其特征在于，所述根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值之前，还包括：
    获取当前人脸识别对应的当前识别场景，获取所述人脸识别错误反馈信息对应的反馈 识别场景；
    当所述当前识别场景与所述反馈识别场景匹配时，进入所述根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值的步骤。
  10. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    获取所述人脸识别错误反馈信息对应的反馈识别场景，建立所述反馈识别场景与所述调整后的识别控制阈值的匹配关系；
    获取当前人脸识别对应的当前识别场景，确定与所述当前识别场景匹配的目标反馈识别场景；
    根据所述匹配关系获取所述目标反馈识别场景对应的目标识别控制阈值；
    根据所述目标识别控制阈值对应的当前人脸识别算法进行人脸识别。
  11. 根据权利要求1所述的方法，其特征在于，所述根据人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，包括：
    获取所述当前人脸识别算法对应的初始识别控制阈值；
    获取识别控制阈值调整范围；
    根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值，使得调整后的当前识别控制阈值与所述初始识别控制阈值的差距在所述识别控制阈值调整范围内。
  12. 一种人脸识别数据处理装置，其特征在于，包括：
    接收模块，用于通过移动设备接收人脸识别错误反馈信息；
    调整模块，用于根据所述人脸识别错误反馈信息调整当前人脸识别算法中对应的识别控制阈值；
    识别模块，用于根据调整识别控制阈值后的当前人脸识别算法进行人脸识别。
  13. 一种移动设备，其特征在于，包括存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行权利要求1至11中任一项权利要求所述方法的步骤。
  14. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有计算机可读指令，所述计算机可读指令被处理器执行时，使得所述处理器执行权利要求1至11中任一项权利要求所述方法的步骤。
PCT/CN2018/119883 2018-12-07 2018-12-07 人脸识别数据处理方法、装置、移动设备和计算机可读存储介质 WO2020113571A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201880098811.6A CN112889062B (zh) 2018-12-07 2018-12-07 人脸识别数据处理方法、装置、移动设备和计算机可读存储介质
PCT/CN2018/119883 WO2020113571A1 (zh) 2018-12-07 2018-12-07 人脸识别数据处理方法、装置、移动设备和计算机可读存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/119883 WO2020113571A1 (zh) 2018-12-07 2018-12-07 人脸识别数据处理方法、装置、移动设备和计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2020113571A1 true WO2020113571A1 (zh) 2020-06-11

Family

ID=70973465

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/119883 WO2020113571A1 (zh) 2018-12-07 2018-12-07 人脸识别数据处理方法、装置、移动设备和计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN112889062B (zh)
WO (1) WO2020113571A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112667984A (zh) * 2020-12-31 2021-04-16 上海商汤临港智能科技有限公司 一种身份认证方法及装置、电子设备和存储介质
CN113963392A (zh) * 2020-07-03 2022-01-21 北京君正集成电路股份有限公司 一种基于动态调节阈值的人脸识别方法
CN116778614A (zh) * 2023-06-16 2023-09-19 泰安鸿瑞信息科技有限公司 一种电动车开关锁智能控制系统
CN117421729A (zh) * 2023-12-18 2024-01-19 湖南森鹰科技有限公司 一种自动化程序攻击检测方法、装置、系统及介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113628376A (zh) * 2021-07-19 2021-11-09 杭州奥声物业有限公司 一种基于无接触识别的物业管理系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101021899A (zh) * 2007-03-16 2007-08-22 南京搜拍信息技术有限公司 综合利用人脸及人体辅助信息的交互式人脸识别系统及方法
US20100232658A1 (en) * 2009-03-13 2010-09-16 Omron Corporation Vehicle operation control device and method, as well as, program
CN102567712A (zh) * 2010-12-08 2012-07-11 现代自动车株式会社 用于面部识别的系统和方法
CN105700737A (zh) * 2014-12-11 2016-06-22 本田技研工业株式会社 触摸板模块、远程输入系统、和控制远程输入系统的方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2503509A4 (en) * 2009-11-17 2018-01-17 Hitachi Industry & Control Solutions, Ltd. Authentication system using organism information, and authentication device
CN105574500B (zh) * 2015-12-15 2019-09-27 北京眼神智能科技有限公司 提高人脸识别通过率的方法和装置
CN105550877A (zh) * 2015-12-21 2016-05-04 北京智付融汇科技有限公司 支付方法及装置
CN106096539B (zh) * 2016-06-07 2018-03-27 广东欧珀移动通信有限公司 一种进行生物识别的方法及终端
CN107818308A (zh) * 2017-10-31 2018-03-20 平安科技（深圳）有限公司 一种人脸识别智能比对方法、电子装置及计算机可读存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101021899A (zh) * 2007-03-16 2007-08-22 南京搜拍信息技术有限公司 综合利用人脸及人体辅助信息的交互式人脸识别系统及方法
US20100232658A1 (en) * 2009-03-13 2010-09-16 Omron Corporation Vehicle operation control device and method, as well as, program
CN102567712A (zh) * 2010-12-08 2012-07-11 现代自动车株式会社 用于面部识别的系统和方法
CN105700737A (zh) * 2014-12-11 2016-06-22 本田技研工业株式会社 触摸板模块、远程输入系统、和控制远程输入系统的方法

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113963392A (zh) * 2020-07-03 2022-01-21 北京君正集成电路股份有限公司 一种基于动态调节阈值的人脸识别方法
CN113963392B (zh) * 2020-07-03 2024-05-03 北京君正集成电路股份有限公司 一种基于动态调节阈值的人脸识别方法
CN112667984A (zh) * 2020-12-31 2021-04-16 上海商汤临港智能科技有限公司 一种身份认证方法及装置、电子设备和存储介质
WO2022142330A1 (zh) * 2020-12-31 2022-07-07 上海商汤临港智能科技有限公司 一种身份认证方法及装置、电子设备和存储介质
JP7482326B2 (ja) 2020-12-31 2024-05-13 シャンハイ リンガン ジュエイン インテリジェント テクノロジー カンパニー リミテッド 身元認証方法および装置、電子機器並びに記憶媒体
CN116778614A (zh) * 2023-06-16 2023-09-19 泰安鸿瑞信息科技有限公司 一种电动车开关锁智能控制系统
CN116778614B (zh) * 2023-06-16 2023-12-08 泰安鸿瑞信息科技有限公司 一种电动车开关锁智能控制系统
CN117421729A (zh) * 2023-12-18 2024-01-19 湖南森鹰科技有限公司 一种自动化程序攻击检测方法、装置、系统及介质
CN117421729B (zh) * 2023-12-18 2024-04-26 湖南森鹰科技有限公司 一种自动化程序攻击检测方法、装置、系统及介质

Also Published As

Publication number Publication date
CN112889062B (zh) 2024-09-20
CN112889062A (zh) 2021-06-01

Similar Documents

Publication Publication Date Title
WO2020113571A1 (zh) 人脸识别数据处理方法、装置、移动设备和计算机可读存储介质
US11093731B2 (en) Analysis of reflections of projected light in varying colors, brightness, patterns, and sequences for liveness detection in biometric systems
US20220172510A1 (en) Face verification method and apparatus
JP5502837B2 (ja) 電子機器、電子機器においてオペレーションを実行する段階を備える方法、及びプログラム
JP7159509B1 (ja) 画像に基づくなりすまし検出のために複数の焦点スタックを使用するシステムおよび方法
WO2019192216A1 (zh) 图像处理、身份验证方法、装置、电子设备和存储介质
US8422746B2 (en) Face authentication system and authentication method thereof
US20230252120A1 (en) Method and apparatus with selective combined authentication
US10956553B2 (en) Method of unlocking an electronic device, unlocking device and system and storage medium
US11314966B2 (en) Facial anti-spoofing method using variances in image properties
US20230290178A1 (en) Detection system, detection method, and computer-readable recording medium
JP2007148968A (ja) 顔認証装置、セキュリティ強度変更方法およびプログラム
US9928399B2 (en) Non-contact palmprint authentication method, device and mobile terminal
TWI752105B (zh) 特徵圖像的獲取方法及獲取裝置、使用者認證方法
WO2020022034A1 (ja) 情報処理装置、情報処理方法および情報処理プログラム
KR20210047848A (ko) 사용자 얼굴 판별 장치 및 사용자 얼굴 판별 방법
CN112417998A (zh) 活体人脸图像的获取方法和装置、介质、设备
EP3724815B1 (en) Anti-spoofing face id sensing based on retro-reflection
US20190303652A1 (en) Multi-view face recognition system and recognition and learning method therefor
KR20200130042A (ko) 사용자 얼굴 판별 장치 및 사용자 얼굴 판별 방법
KR102636195B1 (ko) 눈깜빡임 패턴을 이용한 십진 패스워드 입력 장치 및 그 방법
WO2023119560A1 (ja) 認証装置、認証方法、及び、記録媒体
RU2815689C1 (ru) Способ, терминал и система для биометрической идентификации
KR20240078145A (ko) 금융 거래를 위한 인증을 위한 전자 장치, 방법, 및 컴퓨터 판독가능 저장 매체
JP2023132475A (ja) 管理装置、管理方法および管理プログラム

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: 18942455

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

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

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 15/10/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18942455

Country of ref document: EP

Kind code of ref document: A1