WO2017181842A1 - 电子镜图像处理方法、装置和存储介质 - Google Patents

电子镜图像处理方法、装置和存储介质 Download PDF

Info

Publication number
WO2017181842A1
WO2017181842A1 PCT/CN2017/079414 CN2017079414W WO2017181842A1 WO 2017181842 A1 WO2017181842 A1 WO 2017181842A1 CN 2017079414 W CN2017079414 W CN 2017079414W WO 2017181842 A1 WO2017181842 A1 WO 2017181842A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
face
face image
user
electronic mirror
Prior art date
Application number
PCT/CN2017/079414
Other languages
English (en)
French (fr)
Inventor
万昌洵
Original Assignee
腾讯科技（深圳）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 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Publication of WO2017181842A1 publication Critical patent/WO2017181842A1/zh
Priority to US15/948,743 priority Critical patent/US10846514B2/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques
    • 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
    • G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification
    • G06V40/173—Classification, e.g. identification face re-identification, e.g. recognising unknown faces across different face tracks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • 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
    • 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
    • G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships

Definitions

  • the present invention relates to the field of Internet technologies, and in particular, to the field of Internet of Things technologies, and in particular, to an electronic mirror image processing method and apparatus.
  • Mirrors are a common type of life and business. In the application scenes related to people's clothing or makeup, mirrors are often used, such as fitting mirrors in fitting rooms or makeup mirrors in barbershops.
  • the current mirror function is limited by the use of the scene, usually only in a specific geographical location to achieve its function, the mirror resources are not fully utilized.
  • the fitting mirror in the fitting room is generally only used to show the user's dress image through light reflection when trying on the clothes in the fitting room.
  • Some mirrors also have the function of virtual dressing, but these functions can only be used when the user faces the mirror. The mirror cannot be used when the user is away from the mirror.
  • An embodiment of the present application provides an electronic mirror image processing method, where the method includes:
  • An embodiment of the present application further provides an electronic mirror image processing apparatus, where the apparatus includes:
  • a memory coupled to the processor
  • the memory stores machine readable instructions, and the processor executes the machine readable instructions to perform the following operations:
  • An embodiment of the present application further provides an electronic mirror image processing method, where the method includes:
  • the electronic mirror sent by the receiving server collects a notification with a face image
  • the embodiment of the present application further provides a non-transitory computer readable storage medium storing machine readable instructions, the machine readable instructions being executable by a processor to perform the following operations:
  • the electronic mirror image processing method and device after acquiring the face image collected by the electronic mirror, can automatically recognize the face image to determine the user with the face image, thereby The responding user sends a face image.
  • the electronic mirror is no longer limited by the use scene.
  • the electronic mirror captures the face image, the user can receive the corresponding notification through the user terminal, which expands the application scenario of the electronic mirror, so that the electronic mirror resources are fully utilized, and Convenient.
  • FIG. 1 is an application environment diagram of an electronic mirror interaction system in some embodiments of the present application.
  • FIG. 2 is a schematic structural diagram of an electronic mirror in some embodiments of the present application.
  • FIG. 3 is a schematic structural diagram of an object-opening platform in some embodiments of the present application.
  • FIG. 4 is a schematic structural diagram of a user terminal in some embodiments of the present application.
  • FIG. 5 is a schematic flowchart diagram of an electronic mirror image processing method according to some embodiments of the present application.
  • FIG. 6 is a schematic flowchart of steps of associating a user identifier and a face model in some embodiments of the present application
  • FIG. 7 is a schematic flowchart of steps of performing de-privacy processing on a face image according to some embodiments of the present application.
  • FIG. 8 is a schematic flow chart of steps for authorizing an electronic mirror authorization request in some embodiments of the present application.
  • FIG. 9 is a schematic flowchart diagram of an electronic mirror image processing method according to some embodiments of the present application.
  • FIG. 10 is a schematic flow chart of steps of capturing a user face image to construct a face model in some embodiments of the present application
  • FIG. 11 is a sequence diagram of interaction between a user and an object-opening platform and an electronic mirror through a user terminal in some specific application scenarios;
  • FIG. 12 is a schematic diagram of an interface when an electronic mirror collects a face image according to some embodiments of the present application.
  • FIG. 13 is a schematic diagram of an interface of a user terminal displaying identification information of a face image and an electronic mirror according to a notification according to some embodiments of the present application;
  • FIG. 14 is a structural block diagram of an electronic mirror image processing apparatus in some embodiments of the present application.
  • 15 is another structural block diagram of an electronic mirror image processing apparatus in some embodiments of the present application.
  • 16 is a structural block diagram of an electronic mirror image processing apparatus according to some embodiments of the present application.
  • Figure 17 is a block diagram showing another structure of an electronic mirror image processing apparatus in some embodiments of the present application.
  • an electronic mirror interaction system including: an electronic mirror 102 , an object open platform 104 , and a user terminal 106 .
  • the electronic mirror 102 and the user terminal 106 access the open platform 104 through the network.
  • the IoT open platform 104 includes one or more physical servers, provides access services to the electronic mirror 102, and provides an interactive interface to the user terminal 106.
  • the electronic mirror also referred to as a smart mirror, has data processing capabilities and the ability to interact with the user in addition to presenting images of people or objects located in front of the mirror.
  • the electronic mirror can receive an operation instruction of the user, collect an image of the user in front of the mirror through the camera, and send it to the open platform of the object for storage or processing.
  • the IoT open platform is an open access platform, and an IoT device (such as an electronic mirror in the embodiment of the present application) can access the IoT open platform to implement data storage and management.
  • an IoT device such as an electronic mirror in the embodiment of the present application
  • an electronic mirror 102 including a processor 201 connected by a system bus, a non-volatile storage medium 202, an internal memory 203, a network interface 204, and a camera 205. Also included is at least one of a mirror 206 and a display 207.
  • the processor 201 has a computing function and a function of controlling the operation of the electronic mirror 102.
  • the non-volatile storage medium 202 stores an operating system 2022 and an application 2024 for buffering data of the application and buffering intermediate data processed by the processor.
  • the network interface 204 is for connecting to the IoT open platform 104 and interacting with the IoT open platform 104.
  • Camera 205 is used to capture images.
  • the mirror 206 is used to present a person or object in front of the mirror by the principle of light reflection, and the display screen 207 can be used to achieve an effect similar to that of the mirror 206 by displaying an image captured in real time by the camera 205.
  • an object open platform 104 including a processor 301, a non-volatile storage medium 302, an internal memory 303, and a network interface 304 connected by a system bus.
  • the processor 301 has a computing function and a function of controlling the operation of the object open platform 104, and the processor 301 is configured to perform an electronic mirror image processing method.
  • the non-volatile storage medium 302 includes magnetic storage media, optical storage media, and flash storage media.
  • the non-volatile storage medium stores an operating system 3022 and an electronic mirror image processing device 3024 for implementing an electronic mirror image processing method.
  • a user terminal 106 including a processor 401 connected through a system bus, a non-volatile storage medium 402, an internal memory 403, a network interface 404, and a display screen. 405. Camera 406 and input device 407.
  • the processor 401 has a computing function and a function of controlling the operation of the user terminal 106, and the processor 401 is configured to perform an image processing method.
  • the non-volatile storage medium 402 includes a magnetic storage medium, an optical storage medium, and a flash-type storage medium.
  • the non-volatile storage medium 402 stores an operating system 4022 and an electronic mirror image processing device 4024.
  • the internal memory 403 is used to provide a cache for the operating system 4022 and the electronic mirror image processing device 4024.
  • Display 405 can be a liquid crystal display or an electronic ink display.
  • the input device 407 can be a touchpad, a trackball, a mouse, or a touch layer that overlaps the display screen, wherein the display screen 405 and the touch layer constitute a touch screen.
  • Network interface 404 is then used to connect to the network.
  • the user terminal 104 includes a personal computer including a mobile phone, a tablet, a PDA (Personal Digital Assistant), and a mobile terminal. Internet e-readers, etc.
  • an electronic mirror image processing method is provided.
  • the present embodiment is applied to the object-opening platform 104 in FIG. 1 and FIG.
  • the method specifically includes the following steps:
  • Step 502 Receive a face image collected and uploaded by an electronic mirror.
  • the electronic mirror is a mirror that can collect images and send the collected images through the network, such as an electronic fitting mirror or an electronic makeup mirror.
  • the face image refers to an image with at least one face, and the face image may also include other parts of the human body, such as a half-length photograph with a face or a full-body photograph with a face.
  • the electronic mirror can be configured as a public device, and images can be acquired and provided to the Internet of Things open platform according to public operation instructions, and the authority to read or modify data from the Internet of Things open platform can be restricted.
  • the electronic mirror can acquire the user's manipulation command, thereby collecting the face image according to the manipulation command, wherein the manipulation command can be triggered by pressing a physical button, a touch operation or a voice control.
  • the electronic mirror can also collect images in real time and perform face detection on the collected images. If a human face is detected, the image of the detected face is uploaded to the open platform of the object as a face image.
  • the electronic mirror can also detect the specified limb movements through the images acquired in real time, so as to collect the face image when the specified limb movement is detected, and specify the limb movements such as blinking the eyes, opening and closing the mouth, raising the hand or nodding.
  • the electronic mirror may further edit the collected face image and upload it to the Internet of Things open platform.
  • the electronic mirror can provide various virtual dressing image options, select a virtual dressing image according to the user operation and fuse the selected virtual dressing image into the face image, so that the face image presents the effect of wearing a virtual dress image.
  • Step 504 Find a face model matching the face image in the pre-stored face model.
  • the face model is associated with the user identifier and is data for describing the facial features of the user corresponding to the user identifier.
  • the IoT open platform can obtain the user with the user ID in advance.
  • the face image thereby extracting features from the user face image, forming a face model, and associating and storing the face model with the user identifier of the user to which the face model belongs.
  • a user face image refers to an image of a face with a specific user.
  • the extracted features may use texture features, shape features, or visual word bag features, and the like.
  • the image with the user's face can be selected from the user's web album, or the user can be notified to upload it actively, and can also be pulled from the third-party identity authentication platform.
  • the object-opening platform can extract a face model with a face image, and calculate the similarity between the extracted face model and the pre-stored face models, and then the face model with the highest similarity. Determined to be a face model that matches the face image.
  • the similarity can be Euclidean distance or Mahalanobis distance.
  • the face image may be directly deleted or restricted.
  • the access authority with the face image, or the electronic mirror is deleted to delete the corresponding face image stored on the electronic mirror, and the face image may be maliciously acquired when the user with the face image cannot be determined.
  • Step 506 Acquire a user identifier associated with the matched face model.
  • the Internet of Things Open Platform associates the user identifier of the user with the face model of the user in advance, and each user identifier can be associated with multiple face models of the same user. After determining the face model matching the face image, the IOT open platform acquires the user identifier associated with the face model that matches the face image.
  • the user identifier is used to uniquely identify the corresponding user and may be composed of at least one of a number, a letter, and a special symbol.
  • the object-opening platform may perform a face recognition model according to a pre-existing face model of a known associated user identifier, thereby using a face recognition model after receiving the face image.
  • a user identifier corresponding to the face image is identified.
  • the face recognition model can adopt an artificial neural network model.
  • Step 508 Send an electronic mirror to the user terminal corresponding to the associated user identifier. Notification with a face image.
  • the object-opening platform can obtain the network address of the user terminal corresponding to the associated user identifier, and send a notification to the user terminal according to the network address, where the network address can be reported by the user terminal when the login is successful according to the user identifier.
  • the notification carries the image identification with the face image and the identification information of the electronic mirror, and can also carry the thumbnail image with the face image or the face image.
  • the identification information of the electronic mirror can identify which electronic mirror collects the face image.
  • the identification information is used to identify the corresponding electronic mirror, such as the number, name, location information, and/or owner information of the electronic mirror.
  • the face in the face image may be multiple, the corresponding matching face model may also be multiple, and the associated user identifier is also multiple, and finally the user with the face image is received.
  • the face image processing method After the electronic mirror image processing method obtains the face image collected by the electronic mirror, the face image can be automatically recognized to determine the user with the face image, thereby transmitting the face image to the corresponding user.
  • the electronic mirror is no longer limited by the use scene.
  • the electronic mirror captures the face image the user can receive the corresponding notification through the user terminal, which expands the application scenario of the electronic mirror, so that the electronic mirror resources are fully utilized, and Convenient.
  • the method before step 502, the method further includes the steps of associating the user identifier and the face model, and specifically includes the following steps:
  • Step 602 Receive a user face image and a corresponding user identifier that are sent by the user terminal after acquiring the image of the user face.
  • the user terminal acquires a shooting instruction, and takes a user face image according to the shooting instruction, and sends the user face image together with the user identifier used for logging in to the object connection open platform, and is received by the Internet of Things open platform.
  • the shooting instruction can be performed by pressing a physical button, Trigger by touch operation or voice control.
  • the user face image refers to an image of a face of a user having a user terminal.
  • the number of images with user faces may be plural, and the number may be specified in advance or determined by the user.
  • the plurality of user face images may be face images with different angles of the user.
  • Step 604 extracting a face model according to the user face image.
  • the object-opening platform may extract features from the user's face image to form a face model, and the extracted features may adopt texture features, shape features, or visual word bag features. If there are a plurality of user face images, the face models are extracted from each of the user face images.
  • Step 606 Associate the extracted face model with the received user identifier and store it.
  • the object-opening platform associates the face model, records the relationship between the face model and the user identifier, and stores the face model. If there are multiple user face images, the face models extracted from the user face images are associated with the received user ID.
  • the user can actively take a user face image and upload it through the user terminal, construct a face model and associate with the user identifier, and expand the face model library, and the subsequent object-opening platform receives the electronic mirror collection. After uploading the face image, the notification of the face image can be accurately pushed to the user terminal.
  • the electronic mirror image processing method further includes the step of performing a privacy process on the face image, and specifically includes the following steps:
  • Step 702 Receive a privacy removal processing command fed back by the user terminal.
  • the user terminal displays a prompt that the user's face image is collected by the electronic mirror according to the notification, and specifically displays a thumbnail image with a face image or a face image, and displays the band.
  • the identification information of the electronic mirror of the face image is displayed by the electronic mirror according to the notification.
  • the user terminal can display the face image processing in the toolbar according to the notification display toolbar.
  • Icon detecting an operation instruction for the face image processing icon, generating a processing command for the face image according to the operation instruction, sending the processing command to the Internet of Things open platform, and responding to the processing command by the Internet of Things open platform .
  • the processing command includes a de-privacy processing command.
  • the de-privacy processing command is a command for processing a face image to remove private information in a face image.
  • Step 704 Determine a privacy area in the face image according to the de-privacy processing command.
  • the privacy zone refers to an area with a private image that contains privacy information and needs to be weakened or cleared.
  • the privacy area such as the area where the face is located, the area where the specific text is located, or other areas that are not suitable for exposure in the face image.
  • the de-privacy processing command may specify a location of the privacy zone in the face image, such that the object-opening platform may determine the privacy zone in the face image based on the location information in the privacy-removal processing command.
  • the user terminal may select a privacy area that needs to be subjected to privacy processing according to the user operation instruction, thereby carrying the location information of the privacy area in the privacy removal processing command.
  • Location information can be represented by horizontal and vertical coordinates.
  • the Internet of Things Open Platform may locate the privacy zone in the face image by a pattern recognition manner according to the de-privacy processing command.
  • the IoT open platform can train a classifier for classifying the privacy zone and the non-privacy zone in the face image, so as to respond after receiving the de-private processing command, and the classifier is used to bring the person Position the privacy area in the face image.
  • Step 706 performing privacy processing on the privacy zone.
  • the de-privacy process refers to weakening or eliminating the privacy information in the privacy zone.
  • the privacy region may be Gaussian blurring with a blur radius larger than a preset value, and the preset image or the random image may be covered on the privacy region, and the privacy region may be segmented and then randomly disordered.
  • Step 708 when receiving a request to access a face image, providing a past privacy removal Rational face image.
  • the object-opening platform can feed back the face image subjected to the privacy process according to the request when receiving any user-triggered request for accessing the face image.
  • the object-opening platform may provide a face image with de-privacy processing when receiving a public-triggered request to access a face image of a non-associated user identity; When a user identification triggers a request to access a face image, it provides a face image that has not been processed in the past.
  • the Internet of Things Open Platform can also perform the operation specified by the authorization request with the face image processed by the de-privacy process, or actively push the face image with the de-privacy process as the recommendable resource.
  • the electronic mirror is used as a public device, and is usually placed in a public place.
  • the face image collected by the electronic mirror usually has private information, and the privacy process for the face image can prevent the user from leaking privacy and protect the user from using the public. The safety of the equipment.
  • the user can know that his face is collected by the electronic mirror in time, and the user can actively perform privacy processing on the face image to prevent privacy accidents.
  • the face image after de-privacy processing can be publicly spread due to the weakening or elimination of the privacy information, facilitating image-based interaction.
  • the electronic mirror image processing method further includes: step 510, receiving a non-private image editing command fed back by the user terminal, according to the non-private image editing command pair Face images are edited.
  • the non-private image editing command is an image editing command different from the de-privacy processing command, and is used for controlling the image editing of the non-de-private processing of the face image by the object-opening platform, such as compressing the image, modifying the size, and adding a preset. Modify the picture elements or change the color matching of the image. Among them, preset decorative picture elements such as image borders, virtual dress images, and the like are preset.
  • the user can further edit the face image related to the face collected by the electronic mirror, so that the electronic face is collected and the face is stored by the object open platform.
  • the image meets the user's needs.
  • the face image can be processed in time, and the face image can be avoided to be used incorrectly in a manner unacceptable to the user to which the face belongs.
  • the electronic mirror image processing method further includes: step 512, receiving an image disclosure range configuration command fed back by the user terminal, and configuring the person according to the image disclosure scope configuration command.
  • the scope of the face image is not limited to:
  • the user may trigger an image disclosure scope configuration command through the user terminal to actively configure the disclosure scope of the face image, such as only being visible by itself, only friends, or public.
  • the user can perform a public scope configuration on the face image related to the face of the user to prevent the privacy of the user from being maliciously leaked.
  • the electronic mirror image processing method further includes: step 514, receiving an image sharing command fed back by the user terminal, and sharing the face image according to the image sharing command. .
  • the Internet of Things open platform may extract the identifier of the user who needs to accept the sharing from the image sharing command, thereby sharing the face image according to the extracted user's identifier.
  • the IoT open platform can also share the face image to the user home page corresponding to the user ID.
  • the user can actively share the face image collected by the electronic mirror, thereby increasing the interaction mode between the users.
  • the Internet of Things Open Platform may be initiated according to the user terminal before processing the face image according to the de-privacy processing command, the non-private image editing command, the image disclosure scope configuration command, or the image sharing command. After the payment request completes the payment operation, the face image is processed accordingly.
  • the electronic mirror image processing method further includes the step of authorizing the electronic mirror authorization request, and specifically includes the following steps:
  • Step 802 Receive an electronic mirror or a terminal sent by a terminal associated with the electronic mirror Like an authorization request.
  • the owner of the electronic mirror can initiate an authorization request for the face image through the electronic mirror to the object open platform, and the authorization request is used to request the user who belongs to the face in the face image to target the owner of the electronic mirror. Further operations with face images are authorized.
  • the owner of the electronic mirror is a user who associates the electronic mirror with his/her own user identification through a binding operation, and the terminal associated with the electronic mirror is the terminal to which the owner of the electronic mirror is registered.
  • Step 804 forwarding an authorization request to the user terminal.
  • Step 806 Receive an authorization result fed back by the user terminal and notify the electronic mirror, so that the electronic mirror performs an operation specified by the authorization request with the face image when the authorization result is the permission authorization.
  • the user terminal may provide an option of granting authorization and denying authorization, and selecting an option according to the user selection instruction, thereby generating an authorization result indicating permission authorization or denial authorization according to the selected option, and sending the authorization result.
  • the IoT open platform sends the authorization result to the electronic mirror. If the electronic mirror receives the authorization result indicating the authorization, the operation specified by the authorization request with the face image, for example, the face image is at the electronic mirror end or The operation specified by the authorization request over the network. If the electronic mirror receives the authorization result indicating that the authorization is denied, the electronic mirror may prompt to reject the authorization, and may further delete the face image saved locally by the electronic mirror.
  • the IoT open platform may further subtract the first value from the first value account corresponding to the owner of the electronic mirror according to the authorization result after receiving the authorization result indicating the authorization authorization, and in the associated
  • the second value is added to the second numerical account corresponding to the user identifier.
  • the second value is less than or equal to the first value, and the difference between the first value and the second value may be added to the preset third account.
  • the first value and the second value may be specified by an authorization request or pre-set by the Internet of Things Open Platform.
  • the first value account and the second value account may be a bank fund account, a third party payment platform fund account or a point account.
  • the electronic image may be requested to be in the face image.
  • the user to which the face belongs belongs to the owner of the electronic mirror for the authorization of the further operation of the face image, so that although the electronic mirror is a public device, there is no problem that the user privacy leaks due to the random processing of the face image.
  • an electronic mirror image processing method is further provided.
  • the present embodiment is applied to the user terminal 106 in FIG. 1 and FIG. 4 as an example for illustration.
  • the method specifically includes the following steps:
  • Step 902 The electronic mirror sent by the receiving server collects a notification with a face image, and notifies that the specified face image matches the face model associated with the user identifier corresponding to the local end.
  • the server such as the IoT open platform, can receive the face image collected and uploaded by the electronic mirror, find the face model matching the face image, obtain the user identifier associated with the matched face model, and associate with The user terminal corresponding to the user identifier sends a notification that the electronic mirror collects the image with the face.
  • the electronic mirror is a mirror that can collect images and send the collected images through the network, such as an electronic fitting mirror or an electronic makeup mirror.
  • An image with a face refers to an image with a human face, and may also include other parts of the human body, such as a half-length photograph with a human face or a full-body photograph with a human face.
  • the electronic mirror can be configured as a public device, and images can be acquired and provided to the Internet of Things open platform according to public operation instructions, and the authority to read or modify data from the Internet of Things open platform can be restricted.
  • the electronic mirror can acquire the user's manipulation command, thereby collecting the face image according to the manipulation command, wherein the manipulation command can be triggered by pressing a physical button, a touch operation or a voice control.
  • the electronic mirror can also collect images in real time and perform face detection on the captured images. If a human face is detected, the currently acquired image is uploaded to the open platform of the object as a face image.
  • the electronic mirror can also detect the specified limb movements through the images acquired in real time, thereby detecting the specified limb movements. When collecting images with faces, specify limb movements such as blinking eyes, opening and closing the mouth, raising your hand or nodding.
  • the electronic mirror may further edit the collected face image and upload it to the Internet of Things open platform.
  • the electronic mirror can provide various virtual dressing image options, select a virtual dressing image according to the user operation and merge the selected virtual dressing image into the face image.
  • the face model is associated with the user identifier and is data for describing the facial features of the user corresponding to the user identifier.
  • the object-opening platform can obtain the user face image corresponding to the user identifier in advance, thereby extracting features from the user face image, forming a face model, and associating and storing the face model with the user identifier.
  • the extracted features may use texture features, shape features, or visual word bag features, and the like.
  • the image with the user's face can be selected from the user's web album, or the user can be notified to upload it actively, and can also be pulled from the third-party identity authentication platform.
  • the object-opening platform can extract a face model with a face image, and calculate the similarity between the extracted face model and the pre-stored face models, and then the face model with the highest similarity. Determined to be a face model that matches the face image.
  • the similarity can be Euclidean distance or Mahalanobis distance.
  • the face image may be directly deleted or restricted.
  • the access authority with the face image, or the electronic mirror is deleted to delete the corresponding face image stored on the electronic mirror, and the face image may be maliciously acquired when the user with the face image cannot be determined.
  • the Internet of Things Open Platform associates the user's user ID with the user's face model in advance, and each user ID can be associated with multiple face models of the same user. After determining the face model matching the face image, the IOT open platform acquires the user identifier associated with the face model that matches the face image.
  • User ID is used to uniquely identify the corresponding user, which can be digits, words At least one of a mother and a special symbol.
  • the Internet of Things open platform can obtain the network address of the user terminal corresponding to the associated user identifier, and send a notification to the user terminal according to the network address.
  • the notification carries the image identification with the face image and the identification information of the electronic mirror, and can also carry the thumbnail image with the face image or the face image.
  • the identification information of the electronic mirror can identify which electronic mirror collects the face image.
  • the identification information is used to identify the corresponding electronic mirror, such as the number, name, location information, and/or owner information of the electronic mirror.
  • the face in the face image may be multiple, the corresponding matching face model may also be multiple, and the associated user identifier is also multiple, and finally the user with the face image is received.
  • Step 904 acquiring and displaying the identification information with the face image and the electronic mirror according to the notification.
  • the user terminal may extract the identification information with the face image and the electronic mirror from the notification, and may also extract the identification information with the face image and the electronic mirror from the Internet of Things open platform according to the notification.
  • the user terminal displays the captured face image or the thumbnail image with the face image and the identification information of the electronic mirror, so that the user recognizes the face image and collects the scene with the face image.
  • Step 906 Acquire an operation instruction, and generate a processing command for the face image according to the operation instruction.
  • the user terminal acquires an operation instruction for displaying the face image, thereby generating a corresponding processing command according to the operation instruction.
  • the operation command can be triggered by pressing a physical button, a touch operation, or a voice control.
  • Step 908 sending a processing command to the server.
  • the user terminal sends a processing command to a server, such as an object open platform, such that The server performs corresponding processing on the face image according to the processing command.
  • a server such as an object open platform
  • the processing command includes at least one of a privacy removal processing command, a non-private image editing command, an image disclosure scope configuration command, an image deletion command, and an image sharing command.
  • the Internet of Things open platform may determine a privacy area in the face image according to the de-privacy processing command, perform privacy processing on the privacy area, and when receiving the request to access the face image, Provides a face image with de-privacy processing.
  • the object open platform edits the face image according to the non-private image editing command. If the processing command is an image disclosure scope configuration command, the Internet of Things Open Platform configures the disclosure scope of the face image according to the image disclosure scope configuration command. If the processing command is an image sharing command, the object-opening platform shares the image with the face according to the image sharing command.
  • the face image processing method After the electronic mirror image processing method obtains the face image collected by the electronic mirror, the face image can be automatically recognized to determine the user with the face image, thereby transmitting the face image to the corresponding user.
  • the electronic mirror is no longer limited by the use scene.
  • the electronic mirror captures the face image the user can receive the corresponding notification through the user terminal, which expands the application scenario of the electronic mirror, so that the electronic mirror resources are fully utilized, and Convenient.
  • the face image can be processed in time, and the face image can be avoided to be used incorrectly in a manner unacceptable to the user to which the face belongs.
  • the electronic mirror image processing method before step 902, further includes the steps of capturing a user face image to construct a face model, and specifically includes the following steps:
  • step 1002 a shooting instruction is obtained.
  • the shooting instruction is used to trigger an operation of capturing a user face image, and the shooting instruction It can be triggered by pressing physical buttons, touch operations or voice control.
  • step 1004 the user face image is taken according to the shooting instruction.
  • Step 1006 Send the user identifier and the user face image to the server, so that the server associates and stores the face model extracted according to the user face image with the user identifier.
  • the user face image refers to an image of a face of a user having a user terminal.
  • the number of images with user faces may be plural, and the number may be specified in advance or determined by the user.
  • the plurality of user face images may be face images with different angles of the user.
  • the IoT open platform can extract features from the user face image to form a face model, and the extracted features can adopt texture features, shape features or visual word bag features. If there are a plurality of user face images, the face models are extracted from each of the user face images.
  • the IoT open platform associates the face model, records the relationship between the face model and the user identity, and stores the face model. If there are multiple user face images, the face models extracted from the user face images are associated with the received user ID.
  • the user can actively take a user face image and upload it through the user terminal, construct a face model and associate with the user identifier, and expand the face model library, and the subsequent object-opening platform receives the electronic mirror collection. After uploading the face image, the notification of the face image can be accurately pushed to the user terminal.
  • the user can operate the user terminal to trigger a shooting instruction (step S1101), and the user terminal takes a face image with the user's face (step S1102).
  • the user terminal sends the user face image with the user face image and the user identifier used for the user terminal login to the Internet of Things open platform (step S1103), and the object connection open platform extracts the face model from the user face image, and the face model and the user identifier are Association (step S1104).
  • the user can perform virtual try-on before the electronic mirror, and select a virtual dressing image, such as hair 1201, upper body clothes 1202, bottom body clothes 1203, and shoes.
  • the electronic mirror captures the face image (step S1105) and fuses the virtual dress image into the face image, and then transmits the face image to the Internet of Things open platform (step S1106).
  • the IOT open platform finds a face model that matches the face image (step S1107), acquires a user identifier associated with the face model, and sends an electronic mirror to the user terminal corresponding to the user identifier to collect the face image. Notification (step S1108).
  • the user terminal may display the face image 1301 and the identification information 1302 of the electronic mirror according to the notification (step S1109), as shown in FIG. 13 .
  • the user triggers an operation command (step S1110), and the user terminal generates a processing command for the face image based on the operation command and transmits it to the Internet of Things open platform (step S1111).
  • the Internet of Things Open Platform processes the human face image according to the processing command (step S1112), such as de-privacy processing, image editing without non-privacy processing, configuration disclosure range, and image sharing.
  • the electronic mirror may also send an authorization request to the user terminal through the Internet of Things open platform (step S1113), and the user terminal returns an authorization result to the Internet of Things open platform (step S1114).
  • the IoT open platform deducts the corresponding fee from the fund account of the electronic mirror (step S1115), and feeds back the authorized authorization result to the electronic mirror (step S1116), and the electronic mirror can be based on the authorization result. Further operations are performed on the face image (step S1117).
  • an electronic mirror image processing apparatus 1400 including: a receiving module 1401, a face recognition module 1402, and a sending module 1403.
  • the receiving module 1401 is configured to receive a face image collected and uploaded by the electronic mirror.
  • the face recognition module 1402 is configured to find a face model that matches the face image.
  • the sending module 1403 is configured to acquire a user identifier associated with the matched face model.
  • the user terminal corresponding to the associated user identifier sends a notification that the electronic mirror collects the image with the face.
  • the electronic mirror image processing device 1400 can automatically recognize the face image to determine the user with the face image, thereby correspondingly The user sends a face image.
  • the electronic mirror is no longer limited by the use scene.
  • the user can receive the corresponding notification through the user terminal, which expands the application scenario of the electronic mirror, so that the electronic mirror resources are fully utilized, and Convenient.
  • the electronic mirror image processing apparatus 1500 shown in FIG. 15 further includes: a face model establishing module 1404, It is configured to receive a user face image and a corresponding user identifier that are sent after the user terminal captures the image with the user face; extract the face model according to the user face image; and extract the extracted face model and the received user identifier Link and store.
  • a face model establishing module 1404 It is configured to receive a user face image and a corresponding user identifier that are sent after the user terminal captures the image with the user face; extract the face model according to the user face image; and extract the extracted face model and the received user identifier Link and store.
  • the user can actively take a user face image and upload it through the user terminal, construct a face model and associate with the user identifier, and expand the face model library, and the subsequent object-opening platform receives the electronic mirror collection. After uploading the face image, the notification of the face image can be accurately pushed to the user terminal.
  • the electronic mirror image processing apparatus 1400 further includes a face image processing module 1405, configured to receive a de-privacy processing command fed back by the user terminal, and determine privacy in the face image according to the de-privacy processing command. Area; de-privacy processing of the privacy area; providing a face image with de-privacy processing when receiving a request to access the face image.
  • the electronic mirror is used as a public device, and is usually placed in a public place.
  • the face image collected by the electronic mirror usually has private information, and the privacy process for the face image can prevent the user from leaking privacy and protect the user from using the public. The safety of the equipment.
  • the user can know that his face is collected by the electronic mirror in time, and the user can actively perform privacy processing on the face image to prevent privacy accidents.
  • the face image after de-privacy processing can be publicly spread due to the weakening or elimination of the privacy information, facilitating image-based interaction.
  • the face image processing module 1405 is further configured to receive a user.
  • the non-private image editing command fed back by the terminal edits the face image according to the non-private image editing command.
  • the face image processing module 1405 is further configured to receive an image disclosure scope configuration command fed back by the user terminal, and configure a disclosure scope of the face image according to the image disclosure scope configuration command.
  • the face image processing module 1405 is further configured to receive an image sharing command fed back by the user terminal, and share the face image according to the image sharing command.
  • the electronic mirror image processing apparatus 1400 further includes: an authorization request processing module 1406, configured to receive an authorization request for the face image transmitted by the electronic mirror or the terminal associated with the electronic mirror; and forward to the user terminal Authorizing the request; receiving the authorization result fed back by the user terminal and notifying the electronic mirror, so that the electronic mirror performs the operation specified by the authorization request with the face image when the authorization result is the permission authorization.
  • an authorization request processing module 1406, configured to receive an authorization request for the face image transmitted by the electronic mirror or the terminal associated with the electronic mirror; and forward to the user terminal Authorizing the request; receiving the authorization result fed back by the user terminal and notifying the electronic mirror, so that the electronic mirror performs the operation specified by the authorization request with the face image when the authorization result is the permission authorization.
  • the electronic mirror may request the user who belongs to the face in the face image to authorize the owner of the electronic mirror for further operation of the face image, so that although the electronic mirror is A public device does not cause the user's privacy leakage due to the random processing of the face image, and improves the security of the face-image interaction method based on the electronic mirror.
  • an electronic mirror image processing apparatus 1600 including a notification receiving module 1601, a presentation module 1602, and a processing command processing module 1603.
  • the notification receiving module 1601 is configured to receive a notification that the electronic mirror sent by the server is collected to the face image, and notify the designated face image to match the face model associated with the user identifier corresponding to the local end.
  • the display module 1602 is configured to acquire and display the identification information with the face image and the electronic mirror according to the notification.
  • the processing command processing module 1603 is configured to acquire an operation instruction, generate a processing command for the face image according to the operation instruction, and send the processing command to the server.
  • the electronic mirror image processing device 1600 can automatically recognize the face image to determine the user to which the face image belongs, thereby transmitting the face image to the corresponding user.
  • the electronic mirror is no longer limited by the use scene.
  • the electronic mirror captures the face image
  • the user can receive the corresponding notification through the user terminal, which expands the application scenario of the electronic mirror, so that the electronic mirror resources are fully utilized, and Convenient.
  • the face image can be processed in time, and the face image can be avoided to be used incorrectly in a manner unacceptable to the user to which the face belongs.
  • the electronic mirror image processing apparatus 1700 illustrated in FIG. 17 further includes a face model building module. 1604, for acquiring a shooting instruction; capturing a user face image according to the shooting instruction; sending the user identification and the user face image to the server, so that the server associates the face model extracted according to the user face image with the user identifier and storage.
  • a face model building module. 1604 for acquiring a shooting instruction; capturing a user face image according to the shooting instruction; sending the user identification and the user face image to the server, so that the server associates the face model extracted according to the user face image with the user identifier and storage.
  • the user can actively take a user face image and upload it through the user terminal, construct a face model and associate with the user identifier, and expand the face model library, and the subsequent object-opening platform receives the electronic mirror collection. After uploading the face image, the notification of the face image can be accurately pushed to the user terminal.
  • the processing command includes at least one of a privacy removal processing command, a non-private image editing command, an image disclosure scope configuration command, an image deletion command, and an image sharing command.
  • the image delete command can trigger the server to delete the face image specified by the image delete command on the server.
  • the flow may be accomplished by a computer program instructing associated hardware, which may be stored in a computer readable storage medium, which, when executed, may include the flow of an embodiment of the methods described above.
  • the storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Library & Information Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Image Processing (AREA)

Abstract

本发明涉及一种电子镜图像处理方法和装置，该电子镜图像处理方法包括：接收电子镜采集到的带人脸图像；当查找到与所述带人脸图像匹配的人脸模型时，获取与匹配的人脸模型关联的用户标识；向与关联的用户标识对应的终端发送所述带人脸图像的通知。

Description

电子镜图像处理方法、装置和存储介质
本申请要求于2016年4月22日提交中国专利局、申请号为201610257054.5，发明名称为“电子镜图像处理方法和装置、图像处理方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及互联网技术领域，特别是涉及一种物联网技术领域，尤其涉及电子镜图像处理方法和装置。
发明背景
镜子是一种常见的生活、商业用具，在与人的衣着或妆容有关的应用场景下，通常会用到镜子，比如试衣间中的试衣镜或者理发店中的妆容镜等。然而，目前的镜子功能受限于使用场景，通常只能在特定的地理位置实现其功能，镜子资源没有得到充分利用。比如试衣间中的试衣镜一般仅供在试衣间试穿衣服时通过光反射呈现用户着装形象，有些镜子还具备虚拟穿衣的功能，但这些功能只能在用户面对镜子时使用，用户远离镜子时便无法使用镜子。
发明内容
基于此，有必要针对目前镜子的资源利用率低的问题，提供一种电子镜图像处理方法和装置。
本申请实施例提供了一种电子镜图像处理方法，所述方法包括：
接收电子镜采集到的带人脸图像；
当查找到与所述带人脸图像匹配的人脸模型时，获取与匹配的人脸模型关联的用户标识；
向所述关联的用户标识对应的终端发送所述带人脸图像的通知。
本申请实施例还提供了一种电子镜图像处理装置，所述装置包括：
处理器；
与所述处理器连接的存储器；
所述存储器中存储有机器可读指令，所述处理器执行所述机器可读指令以执行以下操作：
接收电子镜采集到的带人脸图像；
当查找到与所述带人脸图像匹配的人脸模型时，获取与匹配的人脸模型关联的用户标识；
向所述关联的用户标识对应的终端发送所述带人脸图像的通知。
本申请实施例还提供了一种电子镜图像处理方法，所述方法包括：
接收服务器发送的电子镜采集到带人脸图像的通知，
根据所述通知获取并展示所述带人脸图像和所述电子镜的识别信息；
根据获取到的操作指令对所述带人脸图像执行对应处理命令。
本申请实施例还提供了一种非易失性计算机可读存储介质，所述存储介质中存储有机器可读指令，所述机器可读指令可以由处理器执行以完成以下操作：
接收电子镜采集到的带人脸图像；
当查找到与所述带人脸图像匹配的人脸模型作为匹配的人脸模型时，获取与匹配的人脸模型关联的用户标识；
向所述关联的用户标识对应的终端发送所述带人脸图像的通知。
上述电子镜图像处理方法和装置，获取到电子镜采集到的带人脸图像后，可自动识别带人脸图像以确定带人脸图像所属的用户，从而向相 应的用户发送带人脸图像。电子镜不再受限于使用场景，当电子镜采集到带人脸图像后，用户就可以通过用户终端接收到相应的通知，扩展了电子镜的应用场景，使得电子镜资源得到充分利用，而且方便快捷。
附图简要说明
图1为本申请一些实施例中电子镜互动系统的应用环境图；
图2为本申请一些实施例中电子镜的组成结构示意图；
图3为本申请一些实施例中物联开放平台的组成结构示意图；
图4为本申请一些实施例中用户终端的组成结构示意图；
图5为本申请一些实施例中电子镜图像处理方法的流程示意图；
图6为本申请一些实施例中关联用户标识和人脸模型的步骤的流程示意图；
图7为本申请一些实施例中对带人脸图像进行去隐私处理的步骤的流程示意图；
图8为本申请一些实施例中应电子镜授权请求进行授权的步骤的流程示意图；
图9为本申请一些实施例中电子镜图像处理方法的流程示意图；
图10为本申请一些实施例中拍摄带用户人脸图像以构建人脸模型的步骤的流程示意图；
图11为本申请一些具体应用场景中用户通过用户终端与物联开放平台及电子镜进行互动的时序图；
图12为本申请一些实施例中电子镜采集带人脸图像时的界面示意图；
图13为本申请一些实施例中用户终端根据通知展示带人脸图像和电子镜的识别信息的界面示意图；
图14为本申请一些实施例中电子镜图像处理装置的结构框图；
图15为本申请一些实施例中电子镜图像处理装置的另一结构框图；
图16为本申请一些实施例中电子镜图像处理装置的结构框图；
图17为本申请一些实施例中电子镜图像处理装置的另一结构框图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明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发明，并不用于限定本发明。
如图1所示，在本申请一些实施例中，提供了一种电子镜互动系统，包括：电子镜102、物联开放平台104以及用户终端106。其中电子镜102和用户终端106通过网络接入物联开放平台104。物联开放平台104包括一个或多个物理服务器，向电子镜102提供接入服务，并为用户终端106提供交互接口。
在一些实施例中，所述电子镜，也称为智能镜，除了呈现出位于镜前的人或物的影像之外，还具有数据处理能力和与用户交互的能力。例如，所述电子镜可以接收用户的操作指令，通过摄像头采集镜前用户的图像并发送到物联开放平台进行存储或处理。
在一些实施例中，所述物联开放平台是一个开放的接入平台，物联网设备(例如本申请实施例中的电子镜)可以接入到该物联开放平台，实现其数据的存储和管理。
如图2所示，在本申请一些实施例中，提供了一种电子镜102，包括通过系统总线连接的处理器201、非易失性存储介质202、内存储器203、网络接口204和摄像头205，还包括反射镜206和显示屏207中的至少一种。其中处理器201具有计算功能和控制电子镜102工作的功能。 非易失性存储介质202中存储有操作系统2022和应用程序2024，内存储器203用于缓存应用程序的数据并缓存处理器处理得到的中间数据。网络接口204用于连接到物联开放平台104并与物联开放平台104交互。摄像头205用于采集图像。反射镜206用于通过光反射原理呈现位于反射镜前方的人或物，显示屏207可用于通过展示摄像头205实时采集的图像以实现与反射镜206近似的效果。
如图3所示，在本申请一些实施例中，提供了一种物联开放平台104，包括通过系统总线连接的处理器301、非易失性存储介质302、内存储器303和网络接口304。其中处理器301具有计算功能和控制物联开放平台104工作的功能，该处理器301被配置为执行一种电子镜图像处理方法。非易失性存储介质302包括磁存储介质、光存储介质以及闪存式存储介质。非易失性存储介质存储有操作系统3022和电子镜图像处理装置3024，该电子镜图像处理装置用于实现一种电子镜图像处理方法。
如图4所示，在本申请一些实施例中，提供了一种用户终端106，包括通过系统总线连接的处理器401、非易失性存储介质402、内存储器403、网络接口404、显示屏405、摄像头406及输入装置407。其中处理器401具有计算功能和控制用户终端106工作的功能，该处理器401被配置为执行一种图像处理方法。非易失性存储介质402包括磁存储介质、光存储介质以及闪存式存储介质。非易失性存储介质402存储有操作系统4022和电子镜图像处理装置4024。内存储器403用于为操作系统4022和电子镜图像处理装置4024提供高速缓存。显示屏405可以是液晶显示屏或者电子墨水显示屏。输入装置407可以是触控板、轨迹球、鼠标或者与显示屏重叠的触摸层，其中显示屏405和触摸层构成触控屏。网络接口404则用于连接到网络。用户终端104包括个人计算机和移动终端，移动终端包括手机、平板电脑、PDA(个人数字助理)以及可联 网电子阅读器等。
如图5所示，在本申请一些实施例中，提供了一种电子镜图像处理方法，本实施例以该方法应用于上述图1和图3中的物联开放平台104来举例说明。该方法具体包括如下步骤：
步骤502，接收电子镜采集并上传的带人脸图像。
其中，电子镜是一种可采集图像并通过网络发送采集的图像的镜子，比如电子试衣镜或者电子妆容镜等。带人脸图像是指带有至少一张人脸的图像，带人脸图像还可以包括人身体的其它部分，比如带有人脸的半身照片或者带有人脸的全身照片等。电子镜可被配置为公共设备，可根据公众的操作指令采集图像并向物联开放平台提供，而从物联开放平台读取或修改数据的权限可被限制。
电子镜可获取用户的操控指令，从而根据操控指令采集带人脸图像，其中操控指令可通过按压物理按键、触控操作或者声控等方式触发。电子镜也可以实时采集图像并对采集的图像进行人脸检测，若检测到人脸则将检测到人脸的图像作为带人脸图像上传到物联开放平台。电子镜也可以通过实时采集的图像检测指定肢体动作，从而在检测到指定肢体动作时采集带人脸图像，指定肢体动作比如眨眼睛、嘴部开合、举手或者点头等。
在本申请一些实施例中，电子镜还可以对采集的带人脸图像进行进一步编辑后上传到物联开放平台。比如电子镜可提供各种虚拟着装形象选项，根据用户操作选定虚拟着装形象并将选定的虚拟着装形象融合到带人脸图像中，使得带人脸图像呈现人穿着虚拟着装形象的效果。
步骤504，在预存的人脸模型中查找与带人脸图像匹配的人脸模型。
其中，人脸模型与用户标识关联，是用于描述用户标识对应的用户的脸部特征的数据。物联开放平台可预先获取与用户标识对应的带用户 人脸图像，从而从带用户人脸图像中提取特征，形成人脸模型，将人脸模型与该人脸模型所属用户的用户标识关联并存储。带用户人脸图像是指带有特定用户的人脸的图像。提取的特征可以采用纹理特征、形状特征或者视觉词袋特征等。带用户人脸图像可从用户网络相册中选取，也可以通知用户主动上传，还可以从第三方身份认证平台拉取。
在本申请一些实施例中，物联开放平台可提取带人脸图像的人脸模型，并计算提取的人脸模型与预存的各个人脸模型的相似度，进而将相似度最高的人脸模型确定为与带人脸图像匹配的人脸模型。相似度可采用欧氏距离或者马氏距离等。
在本申请一些实施例中，若相似度低于预设最小门限值，则判定未查找到与带人脸图像匹配的人脸模型，此时可直接将带人脸图像删除，或者限制对该带人脸图像的访问权限等，或者通知电子镜删除电子镜上存储的相应带人脸图像，可尽量避免在无法确定带人脸图像归属用户时带人脸图像被恶意获取。
步骤506，获取与匹配的人脸模型关联的用户标识。
具体地，物联开放平台预先将用户的用户标识与用户的人脸模型进行关联，每个用户标识可与相同用户的多个人脸模型关联。物联开放平台在确定与带人脸图像匹配的人脸模型后，获取与该带人脸图像匹配的人脸模型关联的用户标识。用户标识用于唯一标识出相应的用户，可由数字、字母以及特殊符号中的至少一种组成。
在本申请一些实施例中，物联开放平台可事先根据预存的已知关联的用户标识的人脸模型进行训练得到人脸识别模型，从而在接收到带人脸图像之后，利用人脸识别模型识别出带人脸图像对应的用户标识。人脸识别模型可采用人工神经网络模型。
步骤508，向与关联的用户标识对应的用户终端发送电子镜采集到 带人脸图像的通知。
具体地，物联开放平台可获取关联的用户标识对应的用户终端的网络地址，从而根据该网络地址向该用户终端发送通知，该网络地址可由用户终端在根据用户标识登录成功时上报。该通知携带带人脸图像的图像标识和电子镜的识别信息，还可以携带带人脸图像或者带人脸图像的缩略图。
通过通知所携带的图像标识、带人脸图像或者缩略图，可以确定采集到了带人脸图像，通过电子镜的识别信息可识别出具体哪个电子镜采集到带人脸图像。识别信息用来识别出相应的电子镜，比如电子镜的编号、名称、所在地信息和/或拥有者信息等。
在本申请一些实施例中，带人脸图像中的人脸可以是多个，相应的匹配的人脸模型也可以是多个，关联的用户标识也是多个，最终接收带人脸图像的用户终端也可是多个。
上述电子镜图像处理方法，获取到电子镜采集到的带人脸图像后，可自动识别带人脸图像以确定带人脸图像所属的用户，从而向相应的用户发送带人脸图像。电子镜不再受限于使用场景，当电子镜采集到带人脸图像后，用户就可以通过用户终端接收到相应的通知，扩展了电子镜的应用场景，使得电子镜资源得到充分利用，而且方便快捷。
如图6所示，在本申请一些实施例中，步骤502之前，该方法还包括关联用户标识和人脸模型的步骤，具体包括如下步骤：
步骤602，接收用户终端在拍摄获得带用户人脸图像后发送的带用户人脸图像及相应的用户标识。
具体地，用户终端获取拍摄指令，根据拍摄指令拍摄带用户人脸图像，并将带用户人脸图像连同用户终端登录所用的用户标识发送到物联开放平台，由物联开放平台接收。其中拍摄指令可通过按压物理按键、 触控操作或者声控等方式触发。带用户人脸图像是指具有用户终端的用户的人脸的图像。带用户人脸图像的数量可以是多个，该数量可以事先指定或者由用户确定。多个带用户人脸图像可以是用户不同角度的带人脸图像。
步骤604，根据带用户人脸图像提取人脸模型。
具体地，物联开放平台可从带用户人脸图像中提取特征以形成人脸模型，提取的特征可采用纹理特征、形状特征或者视觉词袋特征等。若是多个带用户人脸图像，则从每个带用户人脸图像分别提取人脸模型。
步骤606，将提取的人脸模型与接收到的用户标识关联并存储。
具体地，物联开放平台将人脸模型关联起来，记录该人脸模型和该用户标识之间的关联关系，并将人脸模型存储下来。若是多个带用户人脸图像，则将从这些带用户人脸图像中提取出的人脸模型均与接收到的用户标识关联。
本实施例中，用户可主动通过用户终端拍摄带用户人脸图像并上传，构建人脸模型并与用户标识进行关联，对人脸模型库进行扩充，后续物联开放平台在接收到电子镜采集并上传的带人脸图像后，便可以准确地向用户终端推送采集到带人脸图像的通知。
如图7所示，在本申请一些实施例中，步骤508之后，该电子镜图像处理方法还包括对带人脸图像进行去隐私处理的步骤，具体包括以下步骤：
步骤702，接收用户终端反馈的去隐私处理命令。
具体地，用户终端在接收到通知后，根据该通知展示用户的带人脸图像被电子镜采集到的提示，具体可展示带人脸图像或者带人脸图像的缩略图，并展示采集该带人脸图像的电子镜的识别信息。
用户终端可根据通知展示工具栏，在工具栏中展示带人脸图像处理 图标，检测对该带人脸图像处理图标的操作指令，根据该操作指令生成针对带人脸图像的处理命令，将处理命令发送到物联开放平台，由物联开放平台对该处理命令进行响应。该处理命令包括去隐私处理命令。去隐私处理命令是用于对带人脸图像进行处理以去除带人脸图像中的隐私信息的命令。
步骤704，根据去隐私处理命令，确定带人脸图像中的隐私区域。
其中，隐私区域是指带人脸图像中含有隐私信息并需要减弱或清除隐私信息的区域。隐私区域比如人脸所在区域、特定的文字所在区域或者带人脸图像中其它不宜暴露的区域等。
在本申请一些实施例中，去隐私处理命令可指定隐私区域在带人脸图像中的位置，从而物联开放平台可根据去隐私处理命令中的位置信息确定带人脸图像中的隐私区域。比如用户终端在展示带人脸图像时，可根据用户操作指令选择需要进行去隐私处理的隐私区域，从而将隐私区域的位置信息携带在去隐私处理命令中。位置信息可用横纵坐标来表示。
在本申请一些实施例中，物联开放平台可根据去隐私处理命令通过模式识别方式在带人脸图像中定位隐私区域。具体地，物联开放平台可训练出用于对带人脸图像中的隐私区域和非隐私区域进行分类的分类器，从而在接收到去隐私处理命令后作出响应，通过该分类器从带人脸图像中定位隐私区域。
步骤706，对隐私区域进行去隐私处理。
其中，去隐私处理是指将隐私区域中的隐私信息减弱或者消除。具体地，可对隐私区域进行模糊半径大于预设值的高斯模糊处理，可在隐私区域上遮盖预设图像或随机图像，可对隐私区域进行切分后进行随机乱序等。
步骤708，当接收到访问带人脸图像的请求时，提供经过去隐私处 理的带人脸图像。
具体地，物联开放平台可在接收到任意用户触发的访问带人脸图像的请求时，根据请求反馈经过去隐私处理的带人脸图像。在本申请一些实施例中，物联开放平台可在接收到非关联的用户标识的公众触发的访问带人脸图像的请求时，提供经过去隐私处理的带人脸图像；在接收到关联的用户标识触发的访问带人脸图像的请求时，提供未经过去隐私处理的带人脸图像。物联开放平台还可以将经过去隐私处理的带人脸图像进行授权请求所指定的操作，或者将经过去隐私处理的带人脸图像作为可推荐资源进行主动推送。
本实施例中，电子镜作为公共设备，通常放置于公共场所，电子镜采集的带人脸图像也通常具有隐私信息，对带人脸图像进行去隐私处理可以防止用户隐私泄露，保障用户使用公共设备的安全。而且通过向用户终端发送电子镜采集到带人脸图像的通知，可使得用户及时获知自己的脸部被电子镜采集到，用户便可以主动对带人脸图像进行去隐私处理，防止自己隐私意外泄露。而且经过去隐私处理的带人脸图像因隐私信息被减弱或者消除，便可以公开传播，便于基于图像的互动。
在本申请一些实施例中，如图5所示，步骤508之后，该电子镜图像处理方法还包括：步骤510，接收用户终端反馈的非去隐私图像编辑命令，根据非去隐私图像编辑命令对带人脸图像进行编辑。
其中，非去隐私图像编辑命令是与去隐私处理命令不同的图像编辑命令，用于控制物联开放平台对人脸图像进行非去隐私处理的图像编辑，比如压缩图像、修改尺寸、添加预设修饰性画面元素或者改变图像颜色搭配等。其中预设修饰性画面元素比如图像边框、虚拟着装形象等。
本实施例中，用户可以对由电子镜采集到的涉及自己人脸的带人脸图像进行进一步编辑，使得电子镜采集且由物联开放平台存储的带人脸 图像符合用户使用需求。用户在及时获知自己的脸部被电子镜采集到时，可及时地对带人脸图像进行处理，可尽量避免带人脸图像被以人脸所属用户不可接受的方式错误地使用。
在本申请一些实施例中，如图5所示，步骤508之后，该电子镜图像处理方法还包括：步骤512，接收用户终端反馈的图像公开范围配置命令，根据图像公开范围配置命令配置带人脸图像的公开范围。
具体地，用户可以通过用户终端触发图像公开范围配置命令，以主动地配置带人脸图像的公开范围，比如仅自己可见、仅好友可见或者公众可见。本实施例中，用户可以对涉及自己人脸的带人脸图像进行公开范围的配置，防止自己隐私被恶意泄露。
在本申请一些实施例中，如图5所示，步骤508之后，该电子镜图像处理方法还包括：步骤514，接收用户终端反馈的图像分享命令，根据图像分享命令对带人脸图像进行分享。
具体地，物联开放平台可从图像分享命令提取需接受分享的用户的标识，从而根据提取的用户的标识分享带人脸图像。物联开放平台也可以将带人脸图像分享到用户标识对应的用户主页中。本实施例中，用户可以主动对电子镜采集到的带人脸图像进行分享，增加了用户之间的互动方式。
在本申请一些实施例中，物联开放平台在根据去隐私处理命令、非去隐私图像编辑命令、图像公开范围配置命令或者图像分享命令对带人脸图像进行处理前，可在根据用户终端发起的支付请求完成支付操作后，再对带人脸图像进行相应的处理。
如图8所示，在本申请一些实施例中，电子镜图像处理方法还包括应电子镜授权请求进行授权的步骤，具体包括如下步骤：
步骤802，接收电子镜或与电子镜关联的终端发送的针对带人脸图 像的授权请求。
具体地，电子镜的主人可通过电子镜向物联开放平台发起针对带人脸图像的授权请求，该授权请求用于请求带人脸图像中的人脸所属的用户对电子镜的主人针对该带人脸图像的进一步操作进行授权。电子镜的主人是事先经过绑定操作将电子镜与自己的用户标识进行关联的用户，与电子镜关联的终端是电子镜的主人所登录的终端。
步骤804，向用户终端转发授权请求。
步骤806，接收用户终端反馈的授权结果并通知电子镜，使得电子镜在授权结果为准许授权时对带人脸图像进行授权请求所指定的操作。
具体地，用户终端在接收到授权请求后，可提供准许授权和拒绝授权的选项，根据用户选择指令选择选项，从而根据选择的选项生成表示准许授权或者拒绝授权的授权结果，并将授权结果发送到物联开放平台。物联开放平台将授权结果发送到电子镜，电子镜若接收到表示准许授权的授权结果，则对带人脸图像进行授权请求所指定的操作，比如将带人脸图像在电子镜本端或者通过网络进行授权请求所指定的操作。电子镜若接收到表示拒绝授权的授权结果，则可提示拒绝授权，还可以进一步删除电子镜本地保存的带人脸图像。
在本申请一些实施例中，物联开放平台还可在接收到表示准许授权的授权结果后，根据授权结果从电子镜的主人对应的第一数值账户中减去第一数值，并在关联的用户标识对应的第二数值账户中增加第二数值。第二数值小于等于第一数值，第一数值与第二数值的差值可增加到预设的第三账户中。第一数值和第二数值可由授权请求指定或者由物联开放平台预先设定。第一数值账户和第二数值账户可以是银行资金账户、第三方支付平台资金账户或者积分账户。
本实施例中，电子镜在采集到带人脸图像后，可请求带人脸图像中 的人脸所属的用户对于电子镜的主人针对带人脸图像的进一步操作的授权，这样虽然电子镜是一种公共设备，也不会因为随意处理带人脸图像而导致用户隐私泄露的问题，提高了基于电子镜的带人脸图像互动方式的安全性。
如图9所示，在本申请一些实施例中，还提供了一种电子镜图像处理方法，本实施例以该方法应用于上述图1和图4中的用户终端106来举例说明。该方法具体包括如下步骤：
步骤902，接收服务器发送的电子镜采集到带人脸图像的通知，通知所指定的带人脸图像与本端对应的用户标识所关联的人脸模型相匹配。
具体地，服务器比如物联开放平台可接收电子镜采集并上传的带人脸图像，查找与带人脸图像匹配的人脸模型，获取与匹配的人脸模型关联的用户标识，并向与关联的用户标识对应的用户终端发送电子镜采集到带人脸图像的通知。
其中，电子镜是一种可采集图像并通过网络发送采集的图像的镜子，比如电子试衣镜或者电子妆容镜等。带人脸图像是指带有人脸的图像，还可以包括人身体的其它部分，比如带有人脸的半身照片或者带有人脸的全身照片等。电子镜可被配置为公共设备，可根据公众的操作指令采集图像并向物联开放平台提供，而从物联开放平台读取或修改数据的权限可被限制。
电子镜可获取用户的操控指令，从而根据操控指令采集带人脸图像，其中操控指令可通过按压物理按键、触控操作或者声控等方式触发。电子镜也可以实时采集图像并对采集的图像进行人脸检测，若检测到人脸则将当前采集的图像作为带人脸图像上传到物联开放平台。电子镜也可以通过实时采集的图像检测指定肢体动作，从而在检测到指定肢体动作 时采集带人脸图像，指定肢体动作比如眨眼睛、嘴部开合、举手或者点头等。
在本申请一些实施例中，电子镜还可以对采集的带人脸图像进行进一步编辑后上传到物联开放平台。比如电子镜可提供各种虚拟着装形象选项，根据用户操作选定虚拟着装形象并将选定的虚拟着装形象融合到带人脸图像中。
人脸模型与用户标识关联，是用于描述用户标识对应的用户的脸部特征的数据。物联开放平台可预先获取与用户标识对应的带用户人脸图像，从而从带用户人脸图像中提取特征，形成人脸模型，将人脸模型与用户标识关联并存储。提取的特征可以采用纹理特征、形状特征或者视觉词袋特征等。带用户人脸图像可从用户网络相册中选取，也可以通知用户主动上传，还可以从第三方身份认证平台拉取。
在本申请一些实施例中，物联开放平台可提取带人脸图像的人脸模型，并计算提取的人脸模型与预存的各个人脸模型的相似度，进而将相似度最高的人脸模型确定为与带人脸图像匹配的人脸模型。相似度可采用欧氏距离或者马氏距离等。
在本申请一些实施例中，若相似度低于预设最小门限值，则判定未查找到与带人脸图像匹配的人脸模型，此时可直接将带人脸图像删除，或者限制对该带人脸图像的访问权限等，或者通知电子镜删除电子镜上存储的相应带人脸图像，可尽量避免在无法确定带人脸图像归属用户时带人脸图像被恶意获取。
物联开放平台预先将用户的用户标识与用户的人脸模型进行关联，每个用户标识可与相同用户的多个人脸模型关联。物联开放平台在确定与带人脸图像匹配的人脸模型后，获取与该带人脸图像匹配的人脸模型关联的用户标识。用户标识用于唯一标识出相应的用户，可由数字、字 母以及特殊符号中的至少一种组成。
物联开放平台可获取关联的用户标识对应的用户终端的网络地址，从而根据该网络地址向该用户终端发送通知。该通知携带带人脸图像的图像标识和电子镜的识别信息，还可以携带带人脸图像或者带人脸图像的缩略图。
通过通知所携带的图像标识、带人脸图像或者缩略图，可以确定采集到了带人脸图像，通过电子镜的识别信息可识别出具体哪个电子镜采集到带人脸图像。识别信息用来识别出相应的电子镜，比如电子镜的编号、名称、所在地信息和/或拥有者信息等。
在本申请一些实施例中，带人脸图像中的人脸可以是多个，相应的匹配的人脸模型也可以是多个，关联的用户标识也是多个，最终接收带人脸图像的用户终端也可是多个。
步骤904，根据通知获取并展示带人脸图像和电子镜的识别信息。
具体地，用户终端可从通知中提取带人脸图像以及电子镜的识别信息，还可以根据通知从物联开放平台拉取带人脸图像和电子镜的识别信息。用户终端将获取到的带人脸图像或者带人脸图像的缩略图，以及电子镜的识别信息展示出来，以便用户辨识带人脸图像以及采集该带人脸图像的场景。
步骤906，获取操作指令，根据操作指令生成针对带人脸图像的处理命令。
具体地，用户终端获取针对展示的带人脸图像的操作指令，从而根据该操作指令生成相应的处理命令。操作指令可通过按压物理按键、触控操作或者声控等方式触发。
步骤908，向服务器发送处理命令。
具体地，用户终端向服务器比如物联开放平台发送处理命令，使得 服务器根据该处理命令对带人脸图像进行相应处理。
在本申请一些实施例中，处理命令包括：去隐私处理命令、非去隐私图像编辑命令、图像公开范围配置命令、图像删除命令和图像分享命令中的至少一种。
若处理命令为去隐私处理命令，则物联开放平台可根据去隐私处理命令确定带人脸图像中的隐私区域，对隐私区域进行去隐私处理，当接收到访问带人脸图像的请求时，提供经过去隐私处理的带人脸图像。
若处理命令为非去隐私图像编辑命令，则物联开放平台根据非去隐私图像编辑命令对带人脸图像进行编辑。若处理命令为图像公开范围配置命令，则物联开放平台根据图像公开范围配置命令配置带人脸图像的公开范围。若处理命令为图像分享命令，则物联开放平台根据图像分享命令对带人脸图像进行分享。
上述电子镜图像处理方法，获取到电子镜采集到的带人脸图像后，可自动识别带人脸图像以确定带人脸图像所属的用户，从而向相应的用户发送带人脸图像。电子镜不再受限于使用场景，当电子镜采集到带人脸图像后，用户就可以通过用户终端接收到相应的通知，扩展了电子镜的应用场景，使得电子镜资源得到充分利用，而且方便快捷。而且用户在及时获知自己的脸部被电子镜采集到时，可及时地对带人脸图像进行处理，可尽量避免带人脸图像被以人脸所属用户不可接受的方式错误地使用。
如图10所示，在本申请一些实施例中，步骤902之前，该电子镜图像处理方法还包括拍摄带用户人脸图像以构建人脸模型的步骤，具体包括以下步骤：
步骤1002，获取拍摄指令。
具体地，拍摄指令用于触发拍摄带用户人脸图像的操作，拍摄指令 可通过按压物理按键、触控操作或者声控等方式触发。
步骤1004，根据拍摄指令拍摄带用户人脸图像。
步骤1006，将用户标识和带用户人脸图像发送到服务器，使得服务器将根据带用户人脸图像提取的人脸模型与用户标识关联并存储。
带用户人脸图像是指具有用户终端的用户的人脸的图像。带用户人脸图像的数量可以是多个，该数量可以事先指定或者由用户确定。多个带用户人脸图像可以是用户不同角度的带人脸图像。
物联开放平台可从带用户人脸图像中提取特征以形成人脸模型，提取的特征可采用纹理特征、形状特征或者视觉词袋特征等。若是多个带用户人脸图像，则从每个带用户人脸图像分别提取人脸模型。
物联开放平台将人脸模型关联起来，记录该人脸模型和该用户标识之间的关联关系，并将人脸模型存储下来。若是多个带用户人脸图像，则将从这些带用户人脸图像中提取出的人脸模型均与接收到的用户标识关联。
本实施例中，用户可主动通过用户终端拍摄带用户人脸图像并上传，构建人脸模型并与用户标识进行关联，对人脸模型库进行扩充，后续物联开放平台在接收到电子镜采集并上传的带人脸图像后，便可以准确地向用户终端推送采集到带人脸图像的通知。
下面用一个具体应用场景来说明上述电子镜图像处理方法的原理。参照图11，用户可操作用户终端触发拍摄指令(步骤S1101)，由用户终端拍摄带用户人脸图像(步骤S1102)。用户终端将带用户人脸图像和用户终端登录所用的用户标识发送到物联开放平台(步骤S1103)，物联开放平台从带用户人脸图像提取人脸模型，并将人脸模型与用户标识关联(步骤S1104)。参照图12，用户在电子镜前可进行虚拟试穿，选择虚拟着装形象，比如头发1201、上身衣着1202、下身衣着1203以及鞋 1204等，还可以选择虚拟着装形象的颜色，如图12中1205所示。电子镜采集带人脸图像(步骤S1105)并将虚拟着装形象融合到带人脸图像中，然后将带人脸图像发送到物联开放平台(步骤S1106)。物联开放平台查找到与带人脸图像匹配的人脸模型(步骤S1107)，获取与该人脸模型关联的用户标识，从而向用户标识对应的用户终端发送电子镜采集到带人脸图像的通知(步骤S1108)。
进一步地，用户终端在接收到通知后，可根据通知展示带人脸图像1301以及电子镜的识别信息1302(步骤S1109)，具体如图13所示。进而用户触发操作指令(步骤S1110)，用户终端根据操作指令生成针对带人脸图像的处理命令并发送到物联开放平台(步骤S1111)。物联开放平台根据处理命令对带人脸图像进行处理(步骤S1112)，比如去隐私处理、非去隐私处理的图像编辑、配置公开范围以及进行图像分享等。电子镜还可以通过物联开放平台向用户终端发送授权请求(步骤S1113)，用户终端向物联开放平台返回授权结果(步骤S1114)。若授权结果为准许授权，则物联开放平台从电子镜的资金账户中扣除相应费用(步骤S1115)，并将准许授权的授权结果反馈给电子镜(步骤S1116)，电子镜根据该授权结果可对带人脸图像进行进一步操作(步骤S1117)。
如图14所示，在本申请一些实施例中，提供了一种电子镜图像处理装置1400，包括：接收模块1401、人脸识别模块1402和发送模块1403。
接收模块1401，用于接收电子镜采集并上传的带人脸图像。
人脸识别模块1402，用于查找与带人脸图像匹配的人脸模型。
发送模块1403，用于获取与匹配的人脸模型关联的用户标识。向与关联的用户标识对应的用户终端发送电子镜采集到带人脸图像的通知。
上述电子镜图像处理装置1400，获取到电子镜采集到的带人脸图像后，可自动识别带人脸图像以确定带人脸图像所属的用户，从而向相应 的用户发送带人脸图像。电子镜不再受限于使用场景，当电子镜采集到带人脸图像后，用户就可以通过用户终端接收到相应的通知，扩展了电子镜的应用场景，使得电子镜资源得到充分利用，而且方便快捷。
如图15所示，在本申请一些实施例中，除了接收模块1401、人脸识别模块1402和发送模块1403，图15所示的电子镜图像处理装置1500还包括：人脸模型建立模块1404，用于接收用户终端在拍摄获得带用户人脸图像后发送的带用户人脸图像及相应的用户标识；根据带用户人脸图像提取人脸模型；将提取的人脸模型与接收到的用户标识关联并存储。
本实施例中，用户可主动通过用户终端拍摄带用户人脸图像并上传，构建人脸模型并与用户标识进行关联，对人脸模型库进行扩充，后续物联开放平台在接收到电子镜采集并上传的带人脸图像后，便可以准确地向用户终端推送采集到带人脸图像的通知。
在本申请一些实施例中，电子镜图像处理装置1400还包括带人脸图像处理模块1405，用于接收用户终端反馈的去隐私处理命令；根据去隐私处理命令，确定带人脸图像中的隐私区域；对隐私区域进行去隐私处理；当接收到访问带人脸图像的请求时，提供经过去隐私处理的带人脸图像。
本实施例中，电子镜作为公共设备，通常放置于公共场所，电子镜采集的带人脸图像也通常具有隐私信息，对带人脸图像进行去隐私处理可以防止用户隐私泄露，保障用户使用公共设备的安全。而且通过向用户终端发送电子镜采集到带人脸图像的通知，可使得用户及时获知自己的脸部被电子镜采集到，用户便可以主动对带人脸图像进行去隐私处理，防止自己隐私意外泄露。而且经过去隐私处理的带人脸图像因隐私信息被减弱或者消除，便可以公开传播，便于基于图像的互动。
在本申请一些实施例中，带人脸图像处理模块1405还用于接收用户 终端反馈的非去隐私图像编辑命令，根据非去隐私图像编辑命令对带人脸图像进行编辑。
在本申请一些实施例中，带人脸图像处理模块1405还用于接收用户终端反馈的图像公开范围配置命令，根据图像公开范围配置命令配置带人脸图像的公开范围。
在本申请一些实施例中，带人脸图像处理模块1405还用于接收用户终端反馈的图像分享命令，根据图像分享命令对带人脸图像进行分享。
在本申请一些实施例中，电子镜图像处理装置1400还包括：授权请求处理模块1406，用于接收电子镜或与电子镜关联的终端发送的针对带人脸图像的授权请求；向用户终端转发授权请求；接收用户终端反馈的授权结果并通知电子镜，使得电子镜在授权结果为准许授权时对带人脸图像进行授权请求所指定的操作。
本实施例中，电子镜在采集到带人脸图像后，可请求带人脸图像中的人脸所属的用户对于电子镜的主人针对带人脸图像的进一步操作的授权，这样虽然电子镜是一种公共设备，也不会因为随意处理带人脸图像而导致用户隐私泄露的问题，提高了基于电子镜的带人脸图像互动方式的安全性。
如图16所示，在本申请一些实施例中，提供了一种电子镜图像处理装置1600，包括通知接收模块1601、展示模块1602和处理命令处理模块1603。
通知接收模块1601，用于接收服务器发送的电子镜采集到带人脸图像的通知，通知所指定的带人脸图像与本端对应的用户标识所关联的人脸模型相匹配。
展示模块1602，用于根据通知获取并展示带人脸图像和电子镜的识别信息。
处理命令处理模块1603，用于获取操作指令，根据操作指令生成针对带人脸图像的处理命令；向服务器发送处理命令。
上述电子镜图像处理装置1600，获取到电子镜采集到的带人脸图像后，可自动识别带人脸图像以确定带人脸图像所属的用户，从而向相应的用户发送带人脸图像。电子镜不再受限于使用场景，当电子镜采集到带人脸图像后，用户就可以通过用户终端接收到相应的通知，扩展了电子镜的应用场景，使得电子镜资源得到充分利用，而且方便快捷。而且用户在及时获知自己的脸部被电子镜采集到时，可及时地对带人脸图像进行处理，可尽量避免带人脸图像被以人脸所属用户不可接受的方式错误地使用。
如图17所示，在本申请一些实施例中，除了上述的通知接收模块1601、展示模块1602和处理命令处理模块1603，图17所示的电子镜图像处理装置1700还包括人脸模型构建模块1604，用于获取拍摄指令；根据拍摄指令拍摄带用户人脸图像；将用户标识和带用户人脸图像发送到服务器，使得服务器将根据带用户人脸图像提取的人脸模型与用户标识关联并存储。
本实施例中，用户可主动通过用户终端拍摄带用户人脸图像并上传，构建人脸模型并与用户标识进行关联，对人脸模型库进行扩充，后续物联开放平台在接收到电子镜采集并上传的带人脸图像后，便可以准确地向用户终端推送采集到带人脸图像的通知。
在本申请一些实施例中，处理命令包括：去隐私处理命令、非去隐私图像编辑命令、图像公开范围配置命令、图像删除命令和图像分享命令中的至少一种。图像删除命令可触发服务器删除服务器上由图像删除命令指定的带人脸图像。
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，是可以通过计算机程序来指令相关的硬件来完成，该计算机程序可存储于一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，前述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)等非易失性存储介质，或随机存储记忆体(Random Access Memory，RAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上实施例仅表达了本发明的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。

Claims (10)

  1. 一种电子镜图像处理方法，所述方法包括：
    接收电子镜采集到的带人脸的图像；
    当查找到与所述带人脸图像匹配的人脸模型时，获取与匹配的人脸模型关联的用户标识；
    向所述关联的用户标识对应的终端发送所述带人脸图像的通知。
  2. 根据权利要求1所述的方法，其特征在于，所述接收电子镜采集并上传的带人脸图像之前，预先接收与用户标识相关联的用户人脸图像，并将该接收到的用户人脸图像提取为与用户标识相关联的人脸模型进行存储。
  3. 根据权利要求1所述的方法，其特征在于，所述向所述关联的用户标识对应的终端发送所述带人脸图像的通知之后，还包括：
    根据接收到的用户标识对应的终端反馈的命令，对所述带人脸图像进行相应处理；
    当接收到访问所述带人脸图像的请求时，提供经过处理的带人脸图像。
  4. 根据权利要求3所述的方法，其特征在于，所述命令包括：
    去隐私处理命令、图像编辑命令、图像公开范围配置命令、图像分享命令中的至少一种。
  5. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    当接收到所述用户标识对应终端发送的授权请求后，在获得授权的终端上执行所述授权请求指定的操作。
  6. 一种电子镜图像处理方法，所述方法包括：
    接收服务器发送的电子镜采集到带人脸图像的通知；
    根据所述通知获取并展示所述带人脸图像和所述电子镜的识别信息；
    根据获取到的操作指令对图像执行对应处理命令。
  7. 根据权利要求6所述的方法，其特征在于，所述接收服务器发送的电子镜采集到带人脸图像的通知之前，还包括：
    预先根据获取到的拍摄指令拍摄带人脸图像；
    将所述带人脸图像与用户标识相关联发送到服务器以生成与所述用户标识相关联的人脸模型并进行存储。
  8. 根据权利要求5所述的方法，其特征在于，所述处理命令包括：去隐私处理命令、非去隐私图像编辑命令、图像公开范围配置命令、图像删除命令和图像分享命令中的至少一种。
  9. 一种电子镜图像处理装置，其特征在于，所述装置包括：
    处理器；
    与所述处理器连接的存储器；
    所述存储器中存储有机器可读指令，所述处理器执行所述机器可读指令以执行以下操作：
    接收电子镜采集到的带人脸图像；
    当查找到与所述带人脸图像匹配的人脸模型时，获取与匹配的人脸模型关联的用户标识；
    向所述关联的用户标识对应的终端发送所述带人脸图像的通知。
  10. 一种非易失性计算机可读存储介质，所述存储介质中存储有机器可读指令，所述机器可读指令可以由处理器执行以完成以下操作：
    接收电子镜采集到的带人脸图像；
    当查找到与所述带人脸图像匹配的人脸模型时，获取与匹配的人脸模型关联的用户标识；
    向所述关联的用户标识对应的终端发送所述带人脸图像的通知。
PCT/CN2017/079414 2016-04-22 2017-04-05 电子镜图像处理方法、装置和存储介质 WO2017181842A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/948,743 US10846514B2 (en) 2016-04-22 2018-04-09 Processing images from an electronic mirror

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610257054.5 2016-04-22
CN201610257054.5A CN105956022B (zh) 2016-04-22 2016-04-22 电子镜图像处理方法和装置、图像处理方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/948,743 Continuation US10846514B2 (en) 2016-04-22 2018-04-09 Processing images from an electronic mirror

Publications (1)

Publication Number Publication Date
WO2017181842A1 true WO2017181842A1 (zh) 2017-10-26

Family

ID=56915669

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/079414 WO2017181842A1 (zh) 2016-04-22 2017-04-05 电子镜图像处理方法、装置和存储介质

Country Status (3)

Country Link
US (1) US10846514B2 (zh)
CN (1) CN105956022B (zh)
WO (1) WO2017181842A1 (zh)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105956022B (zh) * 2016-04-22 2021-04-16 腾讯科技（深圳）有限公司 电子镜图像处理方法和装置、图像处理方法和装置
KR102455056B1 (ko) * 2018-02-01 2022-10-17 삼성전자주식회사 컨텍스트에 따라 이벤트의 출력 정보를 제공하는 전자 장치 및 이의 제어 방법
CN110309691B (zh) * 2018-03-27 2022-12-27 腾讯科技（深圳）有限公司 一种人脸识别方法、装置、服务器及存储介质
CN108881728B (zh) * 2018-07-26 2021-03-30 北京京东尚科信息技术有限公司 一种线下跨设备影像拍摄的方法、系统和拍摄设备
JP7412348B2 (ja) * 2018-10-29 2024-01-12 サターン ライセンシング エルエルシー 表示装置、及び表示制御方法
WO2021150880A1 (en) * 2020-01-22 2021-07-29 Stayhealthy, Inc. Augmented reality custom face filter
CN114499903B (zh) * 2020-11-11 2024-01-30 腾讯科技（深圳）有限公司 一种人脸识别场景中的数据传输方法以及相关装置
CN112966737A (zh) * 2021-03-04 2021-06-15 支付宝(杭州)信息技术有限公司 图像处理、图像识别模型的训练、图像识别的方法和系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090060289A1 (en) * 2005-09-28 2009-03-05 Alex Shah Digital Image Search System And Method
CN103945001A (zh) * 2014-05-05 2014-07-23 百度在线网络技术（北京）有限公司 一种图片分享方法及装置
CN104572732A (zh) * 2013-10-22 2015-04-29 腾讯科技（深圳）有限公司 查询用户标识的方法及装置、获取用户标识的方法及装置
CN105956022A (zh) * 2016-04-22 2016-09-21 腾讯科技（深圳）有限公司 电子镜图像处理方法和装置、图像处理方法和装置

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100771137B1 (ko) * 2005-02-21 2007-10-30 삼성전자주식회사 프라이버시영역을 마스크 처리하는 감시시스템 및 마스크영역 설정방법
US10109315B2 (en) * 2005-03-01 2018-10-23 Eyesmatch Ltd Devices, systems and methods for auto-delay video presentation
CN101090482B (zh) * 2006-06-13 2010-09-08 唐琎 一种基于图象处理和信息融合技术的驾驶员疲劳监测系统及方法
JP5018404B2 (ja) * 2007-11-01 2012-09-05 ソニー株式会社 画像識別装置および画像識別方法、並びに、プログラム
JP5088161B2 (ja) * 2008-02-15 2012-12-05 ソニー株式会社 画像処理装置、カメラ装置、通信システム、画像処理方法、およびプログラム
US20090324022A1 (en) * 2008-06-25 2009-12-31 Sony Ericsson Mobile Communications Ab Method and Apparatus for Tagging Images and Providing Notifications When Images are Tagged
CN102521619A (zh) * 2011-11-22 2012-06-27 汉王科技股份有限公司 基于人脸识别的信息呈现方法及装置和人脸识别打印系统
CN102724192B (zh) * 2012-06-14 2016-04-06 北京奇乐客科技有限公司 基于协同编辑的网络收藏方法
US9727781B2 (en) * 2012-10-31 2017-08-08 Entrust Datacard Corporation Machine vision verification
JP6143469B2 (ja) * 2013-01-17 2017-06-07 キヤノン株式会社 情報処理装置、情報処理方法及びプログラム
WO2015045713A1 (ja) * 2013-09-24 2015-04-02 シャープ株式会社 画像処理装置及び画像表示装置
CN103605928B (zh) * 2013-11-18 2016-03-30 清华大学 一种图像隐私保护方法和系统
CN104199542A (zh) * 2014-08-20 2014-12-10 百度在线网络技术（北京）有限公司 一种智能镜实现方法、装置和智能镜
CA2902093C (en) * 2014-08-28 2023-03-07 Kevin Alan Tussy Facial recognition authentication system including path parameters
CN105187810B (zh) * 2014-11-11 2017-06-06 怀效宁 一种基于人脸色彩特征的自动白平衡方法及电子媒体装置
CN105354334B (zh) * 2015-11-27 2019-04-26 广州视源电子科技股份有限公司 一种基于智能镜子的信息发布方法和智能镜子

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090060289A1 (en) * 2005-09-28 2009-03-05 Alex Shah Digital Image Search System And Method
CN104572732A (zh) * 2013-10-22 2015-04-29 腾讯科技（深圳）有限公司 查询用户标识的方法及装置、获取用户标识的方法及装置
CN103945001A (zh) * 2014-05-05 2014-07-23 百度在线网络技术（北京）有限公司 一种图片分享方法及装置
CN105956022A (zh) * 2016-04-22 2016-09-21 腾讯科技（深圳）有限公司 电子镜图像处理方法和装置、图像处理方法和装置

Also Published As

Publication number Publication date
CN105956022A (zh) 2016-09-21
CN105956022B (zh) 2021-04-16
US10846514B2 (en) 2020-11-24
US20180225505A1 (en) 2018-08-09

Similar Documents

Publication Publication Date Title
WO2017181842A1 (zh) 电子镜图像处理方法、装置和存储介质
KR102350507B1 (ko) 출입 제어 방법, 출입 제어 장치, 시스템 및 저장매체
CN105528573B (zh) 用户终端设备及其虹膜识别方法
CA3111097C (en) Bounding box doubling as redaction boundary
US20150009359A1 (en) Method and apparatus for collaborative digital imaging
US20140380446A1 (en) Method and apparatus for protecting browser private information
CN106603928B (zh) 一种拍摄方法及移动终端
JP2017531950A (ja) 撮影テンプレートデータベースを構築し、且つ撮影推薦情報を提供するための方法及び装置
CN105912912B (zh) 一种终端用户身份登录方法和系统
US9792421B1 (en) Secure storage of fingerprint related elements
CN110688914A (zh) 一种手势识别的方法、智能设备、存储介质和电子设备
AU2014339827B2 (en) Generating image compositions
CN104778416A (zh) 一种信息隐藏方法及终端
WO2022048352A1 (zh) 基于面部表情的解锁方法、装置、计算机设备和存储介质
US9965612B2 (en) Method and system for visual authentication
CN107992811A (zh) 人脸识别方法及装置
WO2018092378A1 (ja) 情報処理装置、情報処理方法、及びプログラム
JP2019074938A (ja) 通信中継装置、システム、方法及びプログラム
WO2020115910A1 (ja) 情報処理システム、情報処理装置、情報処理方法、およびプログラム
CN112734657B (zh) 基于人工智能和三维模型的云合影方法、装置及存储介质
CN108304563A (zh) 图片处理方法、装置及设备
JP2020091876A (ja) 情報処理装置、情報処理方法、プログラムならびに情報処理サーバ
JP2020091731A (ja) 情報処理装置、情報処理方法、プログラムならびに情報処理サーバ
JP2019186914A (ja) 通信端末、通信システム、通信方法、及びプログラム
TWI729679B (zh) 認證系統、認證裝置、及認證方法

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17785327

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17785327

Country of ref document: EP

Kind code of ref document: A1