CN110334663A - Age recognition methods and device, storage medium and terminal based on image - Google Patents
Age recognition methods and device, storage medium and terminal based on image Download PDFInfo
- Publication number
- CN110334663A CN110334663A CN201910615158.2A CN201910615158A CN110334663A CN 110334663 A CN110334663 A CN 110334663A CN 201910615158 A CN201910615158 A CN 201910615158A CN 110334663 A CN110334663 A CN 110334663A
- Authority
- CN
- China
- Prior art keywords
- age
- face image
- user
- candidate face
- initial pictures
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video 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
-
- 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
- G06V40/178—Human faces, e.g. facial parts, sketches or expressions estimating age from face image; using age information for improving recognition
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The disclosure provides a kind of age recognition methods based on image and device, storage medium and terminal.This method comprises: obtaining initial pictures, it include face in the initial pictures, then, recognition of face processing is carried out to the initial pictures, obtains at least one candidate face image collection, the candidate face image in each candidate face image collection belongs to a user, to, for one or more candidate face image collections, age identifying processing is carried out respectively, obtains the age recognition result of each user.Disclosed method improves the accuracy of age prediction result.
Description
Technical field
This disclosure relates to which computer technology more particularly to a kind of age recognition methods and device based on image, storage are situated between
Matter and terminal.
Background technique
With the development of internet technology, low age user is easy in using terminal or application program by the bad of network
It influences, therefore, how to identify low age user and corresponding low age safeguard measure is taken it to become the skill that this field is paid close attention to
Art problem.
Currently, being generally based on neural network model realizes low age user identification.Specifically, by obtaining user's hair
The image or video of table get several frame image datas, include face in image data, then, directly by these picture numbers
According to being directly inputted into the neural network model of age of user for identification, inputted face is predicted by neural network model
Minimum age, thus, realize be directed to low age user identification.
But when it is aforementioned be input in the image data of neural network model include a more than face when, due to not
Corresponding to different user with face, this brings biggish noise to prediction result, so that predicted by existing scheme
Minimum age is the age prediction result for being mixed with multiple users, and the accuracy of prediction result is lower.
Summary of the invention
The disclosure provides a kind of age recognition methods based on image and device, storage medium and terminal, to improve year
The accuracy of age prediction result.
In a first aspect, the disclosure provides a kind of age recognition methods based on image, comprising:
Initial pictures are obtained, include face in the initial pictures；
Recognition of face processing is carried out to the initial pictures, obtains at least one candidate face image collection, each candidate
Candidate face image in face image set belongs to a user；
For one or more candidate face image collections, age identifying processing is carried out respectively, obtains each user's
Age recognition result.
Second aspect, the disclosure provide a kind of age identification device based on image, comprising:
Module is obtained, includes face in the initial pictures for obtaining initial pictures；
First processing module obtains at least one candidate face for carrying out recognition of face processing to the initial pictures
Image collection, the candidate face image in each candidate face image collection belong to a user；
Second processing module, for carrying out age identification respectively for one or more candidate face image collections
Processing, obtains the age recognition result of each user.
The third aspect, the disclosure provide a kind of age identification device based on image, comprising:
Memory；
Processor；And
Computer program；
Wherein, the computer program stores in the memory, and is configured as being executed by the processor with reality
Now method as described in relation to the first aspect.
Fourth aspect, the disclosure provide a kind of terminal, comprising:
Age identification device based on image, for realizing method as described in relation to the first aspect；
Terminal body.
5th aspect, the disclosure provide a kind of computer readable storage medium, are stored thereon with computer program,
The computer program is executed by processor to realize method as described in relation to the first aspect.
A kind of age recognition methods and device, storage medium and terminal based on image that the disclosure provides, is getting
After initial pictures, by way of recognition of face, each candidate face image for including in initial pictures is split, is obtained each
The corresponding candidate face image collection of user, and target facial image is determined wherein, thus, by target face
Image carries out age identification, to obtain the age recognition result of target user.In the process, the embodiment of the present disclosure is by face figure
As distinguishing prediction, different faces image is avoided to the noise jamming of age prediction result, improves age recognition result
Recognition accuracy.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.
Fig. 1 is a kind of flow diagram of the age recognition methods based on image provided by the embodiment of the present disclosure；
Fig. 2 is the flow diagram of age recognition methods of the another kind based on image provided by the embodiment of the present disclosure；
Fig. 3 is the configuration diagram of age identification model provided by the embodiment of the present disclosure；
Fig. 4 is the flow diagram of age recognition methods of the another kind based on image provided by the embodiment of the present disclosure；
Fig. 5 is the flow diagram of age recognition methods of the another kind based on image provided by the embodiment of the present disclosure；
Fig. 6 is a kind of functional block diagram of the age identification device based on image provided by the embodiment of the present disclosure；
Fig. 7 is a kind of entity structure schematic diagram of the age identification device based on image provided by the embodiment of the present disclosure；
Fig. 8 is the entity structure signal of age identification device of the another kind based on image provided by the embodiment of the present disclosure
Figure；
Fig. 9 is a kind of configuration diagram of terminal provided by the embodiment of the present disclosure.
Through the above attached drawings, it has been shown that the specific embodiment of the disclosure will be hereinafter described in more detail.These attached drawings
It is not intended to limit the scope of this disclosure concept by any means with verbal description, but is by referring to specific embodiments
Those skilled in the art illustrate the concept of the disclosure.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all implementations consistent with this disclosure.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the disclosure.
The specific application scenarios of the disclosure are as follows: the scene of age identification is carried out to user.It further, can also be further
Specifically: according to age of user, the scene of personalized recommendation is carried out for user.Alternatively, can be with specifically: according to user year
Age carries out the application scenarios of rights management such as low age user to special user.For example, delivering this permission of content for user
Rights management scene, the rights management scene etc. when being directed to user's browsing content (such as video, information).
As previously mentioned, existing do not handled the initial pictures got for age knowledge method for distinguishing, but
It is directly inputted age prediction model and carries out age prediction, the age prediction result that age prediction model obtains as a result,
It may be the age prediction result for having mixed multiple faces.In other words, it is carried out in age identification process for user's face, just
Other faces for including in beginning image cause biggish interference noise to the recognition result of user's face, have seriously affected the age
The accuracy and reliability of prediction result.
Technical solution provided by the present disclosure, it is intended to solve the technical problem as above of the prior art.
How the technical solution of the disclosure and the technical solution of the application are solved with specifically embodiment below above-mentioned
Technical problem is described in detail.These specific embodiments can be combined with each other below, for the same or similar concept
Or process may repeat no more in certain embodiments.Below in conjunction with attached drawing, embodiment of the disclosure is described.
Embodiment one
The embodiment of the present disclosure provides a kind of age recognition methods based on image.Referring to FIG. 1, this method includes as follows
Step:
S102 obtains initial pictures, includes face in the initial pictures.
This programme is for realizing the identification to age of user, in the process, the behavioral data of available target user,
And an at least frame image data is obtained in the behavioral data, to carry out follow-up process as initial pictures.
Wherein, behavioral data involved by the embodiment of the present disclosure can include but is not limited to: deliver data, storing data
With acquisition at least one of data.And the behavioral data can be what target user currently acquire or get in real time
Behavioral data, and/or, the historical behavior data of target user.
Wherein, data are delivered and refer to the data that target user delivers out in the application.For example, target user is answering
With at least one of the picture, video and lteral data delivered in program.And storing data refers to and is stored in terminal or application
Data in program servers (or application memory), these data may not delivered externally.For example, target user exists
It uploads onto the server in application program or memory, but is only used for collecting or store, but the data that do not deliver externally.Acquire number
The collected data of equipment are acquired using application call according to referring to.For example, calling terminal camera to adopt in the application
Image or video data for collecting etc..
Specifically, primary data is the image comprising face, it can be an image, as camera is shot
Photo etc. comprising face, alternatively, can also be at least frame image in multiple image, such as the frame or more in one section of video
Frame image, wherein include facial image in an at least frame image.
In addition, if getting target user's behavioral data is that video data can be in video counts when obtaining initial pictures
According to a middle extraction at least frame image using as initial pictures.Wherein, the mode of extraction can be with self-definition design, for example, can be with
Machine extracts, or can extract according to preset interval or period, or can extract a few frame images of intermediate beginning, ending.
In addition, it should be noted that, technical solution provided by the embodiment of the present disclosure be based on to the face in image into
The row age identification and realize, the embodiment of the present disclosure is illustrated only for the case where including face in initial pictures as a result,.From
And in initial pictures accessed by the step, existing in an at least frame (or at least one) image includes facial image.Example
Such as, initial pictures can be the multiple image in one section of video data, wrap in an at least frame image at this point, existing in multiple image
Containing face, it is specifically limited whether the embodiment of the present disclosure all has no every frame (or every) image comprising face.
S104 carries out recognition of face processing to the initial pictures, obtains at least one candidate face image collection, each
Candidate face image in candidate face image collection belongs to a user.
The purpose of this step is that the facial image to each user for including in initial pictures identifies and classifies, to obtain
Each corresponding candidate face image collection of user.
It is found that will can directly be given if initial pictures carry out the facial image that a user is had in recognition of face processing
Multiple facial images of user are as a candidate face image collection.But if recognition of face is carried out to initial pictures
The facial image of at least two users is obtained after processing, then after executing the step, can get at least two candidate face figures
Image set closes, and in each candidate face image collection only includes the candidate face image of a user.
And for any one candidate face image collection, included in candidate face image number be extremely
It is one few.For example, passing through available 3 times of the step if initial pictures are the photo of a face comprising 3 users
Face image set is selected, only includes a facial image in each candidate face image collection.If initial pictures are multiple image
Data, then it is possible that therefore a case where user occurs in multiple image is wrapped in candidate face image collection
The number of the facial image of the user contained is multiple.
After aforementioned recognition of face processing, the identification for face in image is not only completed, is also completed at the same time
The classification of different user facial image.
S106 carries out age identifying processing respectively, obtains each use for one or more candidate face image collections
The age recognition result at family.
It is exactly that age identifying processing is carried out to each candidate face image collection when executing the step.It is found that if aforementioned obtain
To N number of candidate face image collection, then by N number of age recognition result can be obtained after the step, wherein N be greater than or
Integer equal to 1.
By scheme as shown in Figure 1, the embodiment of the present disclosure can be for the face figure of each user in initial pictures
As respectively realize the age identification, compared to the prior art in, the initial pictures of the facial image comprising N number of user are directly carried out
Age identifies and has to the scheme of 1 age recognition result, and technical solution provided by the embodiment of the present disclosure avoids difference
To the adverse effect of respective age recognition result between face, be conducive to the accuracy for improving age recognition result.
Hereinafter, being described further to each step in scheme as shown in Figure 1.
When implementing recognition of face step described in S104, the embodiment of the present disclosure provides at least as shown in Figure 2 two
Kind implementation:
In a kind of possible design, it can use face recognition technology and identify all face figures for including in initial pictures
Picture, as candidate face image, then, by carrying out face matching to each candidate face image, to obtain each candidate face
Set.
In specific implementation, it can use nerual network technique to realize recognition of face.It is, execute the step it
Before, building can be used to identify the neural network model of the facial image in initial pictures, and using sample data to the nerve net
Network model is trained, in this way, trained neural network model namely human face recognition model can be obtained.
In addition, the embodiment of the present disclosure is also not particularly limited the types of models of involved each neural network model.
It may include but be not limited to: convolutional neural networks model (Convolutional Neural Network, CNN) or recurrent neural
Network model (Recursive Neural Network, RNN).
As shown in Fig. 2, at this point, S104 may include steps of:
S1042-2 handles the initial pictures using the first human face recognition model, obtains at least one candidate face figure
Picture.
As previously mentioned, the input data of the first human face recognition model is image, output data is the face for including in image
Image.In addition, it is contemplated that initial pictures may be an at least frame image, and therefore, when executing the step, the first recognition of face mould
The input data of type can be a frame initial pictures, at this point, the first human face recognition model is used to carry out face knowledge to a frame image
Other places reason；Alternatively, the input data of the first human face recognition model is multiframe initial pictures, at this point, the first in another design
The processing of the recognition of face to multiple image can be achieved at the same time in face identification model.
S1042-4 carries out face matching to each candidate face image, obtains at least one described candidate face collection
It closes, the candidate face image in each candidate face image collection belongs to a user.
Since the output of aforementioned first human face recognition model is the facial image for including in initial pictures, and the same user
Face may occur in first frame initial pictures again, may also occur in third frame initial pictures, at this time, it may be necessary to preceding
It states the candidate face image identified and carries out matching treatment, to obtain the candidate face image collection of each user.
Specifically, feature extraction can be carried out to each candidate face image, the people of each candidate face image is obtained
Face feature；Then, the similarity of the face characteristic between each candidate face image is calculated, then, similarity is greater than default phase
Like facial image of the candidate face image as the same user of degree threshold value.In this way, realizing the face to candidate face image
Matching, obtains each candidate face set.
Alternatively, can use neural network algorithm also to realize that face matches.It is, design face Matching Model, people
The input data of face Matching Model is multiple facial images, and output data is at least one face image set, each face figure
Image set closes the facial image for including and belongs to the same user.In this way, directly being obtained aforementioned when executing the face matching step
Each candidate face image input the face Matching Model, at least one described candidate face set can be obtained.
In alternatively possible design, neural network algorithm can use, construct the second human face recognition model, and the second people
The input of face identification model is initial pictures, is exported as at least one candidate face image collection.
As shown in Fig. 2, at this point, S104 may include steps of:
S1044-2 handles the initial pictures using the second human face recognition model, obtains at least one candidate face image
Gather, the candidate face image in each candidate face image collection belongs to a user.
In this type of design, using second human face recognition model be directly realized by the identification to face in initial pictures and
Classification, can simplify processing step, but the requirement to neural network model is also higher to a certain extent.
By aforementioned either type, the embodiment of the present disclosure can be realized for each candidate face image collection identification and
It obtains.On this basis, when executing age identification step described in S106, trained age identification model is directly utilized, it is right
At least one aforementioned candidate face image collection carries out age identification.
At this point, S106 can be with specific manifestation are as follows: utilize trained age identification model, handle one or more institutes respectively
Candidate face image collection is stated, the age recognition result of each user is obtained.
In the embodiment of the present disclosure, based on different realization scenes, age identification model can be there are many different designs.Tool
Body, design method shown in Fig. 3 can be referred to.
On the one hand, for the input of age identification model, Fig. 3 can be referred to, at least may include such as under type:
All or part of candidate face image in one or more candidate face image collections；Alternatively,
One or more candidate face features, each candidate face feature extraction is from a candidate face image
Set.
Wherein, " one or more " be used to indicate the age identification model both can be only for a candidate face image set
The age identification for carrying out owning user is closed, multiple candidate face image collections can also be directed to, while handling it, to divide
The age recognition result of each candidate face image collection owning user is not identified.Hereinafter, for convenience of description, with the feelings of "one"
For condition, it is specifically described.
In the first design as shown in Figure 3, one or more candidate face image is as the defeated of human face recognition model
Enter.
It, can directly will be complete included in the candidate face image collection at this point, be directed to any candidate face image collection
Portion's candidate face image directly inputs the age identification model；Alternatively, can also be filtered out in the candidate face image collection
Part candidate face image, and the candidate face image filtered out is inputted into the age identification model.Present aspect embodiment for
The screening step is not particularly limited, and can according to need realization.For example, can be with random screening, alternatively, can be according to each candidate
At least one of the clarity of face image, integrity degree are screened, etc..
For example, the number of the input picture of human face recognition model can be preset according to actual needs, such as 3；So, at this
Kind is realized in scene, for the set comprising 3 or more candidate face images, then needs to filter out 3 candidate faces wherein
Image, at this point it is possible to which it is pre- to filter out clarity highest 3 candidate face images input age in each candidate face image
Survey model.
In the first design as shown in Figure 3, due to without to the candidate face image in candidate face image collection
Make other processing, implementation is relatively simple, and due to directly using candidate face image as input, to thin in facial image
Section loss is smaller, is conducive to the accuracy for improving recognition result.
And in design, then need to be for further processing to each candidate face image in as shown in Figure 3 second, to obtain
To the face characteristic of candidate face image collection owning user, and using face characteristic as the input of age identification model.
In other words, in this implementation, before executing the identification step, it is also necessary to further obtain candidate face figure
The face characteristic of image set conjunction owning user.
It should be noted that for any candidate face image collection, candidate face image collection owning user
Face characteristic can be with are as follows: and one of candidate face image collection represents the set face characteristic of set, or, or it is candidate
The face characteristic that every candidate face image respectively extracts in face image set.Wherein, foregoing assemblage face characteristic can
The face characteristic that wherein a candidate face image zooming-out comes out is thought, alternatively, can pick up by oneself for every candidate face image
The normalization of the face characteristic taken out indicates.And can be the same to the mode of each candidate face image into property feature extraction, no
It repeats again.
In this way, in second of design as shown in Figure 3, it can be by way of shifting to an earlier date feature extraction to each candidate face
Image collection is pre-processed, in this way, advantageously reducing the data processing amount of age identification model, is conducive to improve processing effect
Rate.
On the other hand, for the age recognition result of the output of age identification model, Fig. 3 can be referred to, at least can wrap
Include following at least one mode:
The age of user；
Age level locating for user；
It whether include target age section user；
Whether target user is target age section user.
Specifically, be directed to a candidate face image collection, i.e. a user, age identification scene in, pass through year
The processing of age identification model, the data that can directly export are as follows: the age of the candidate face image collection owning user, age level
With whether be at least one of target age section user information.
Specifically, be directed to multiple candidate face image collections, i.e., multiple users, age identification scene in, then can be with
There is different dispositions:
In one possible implementation, after by the processing of age identification model, the data that can directly export are as follows: each
Whether whether age of user, age level in each user are that target age section is used comprising target age section user, target user
At least one of family information.
In alternatively possible implementation, after the processing by age identification model, the data that can directly export are as follows:
Whether age level locating for age of target user, target user in each user includes target age section user, target user
It whether is at least one of target age section user information.
Wherein, age level and target age section can self-definition designs as needed.For example, can be by age bracket
Setting are as follows: low age, teenager, the young and the middle aged and old age.And the age of age level selects then self-definition design as needed, example
Such as, low age section can be 0~6 years old, and teenager's section can be 7~16 years old, etc..And target age section can also be according to different realities
Live scape selects at least one age bracket, is target age section.For example, the permission for low age user being previously mentioned controls
It, can be using low age section as target age section in scene；In another example the individualized content for old user is recommended in scene,
It can be using old section as target age section.
Wherein, target user involved by the embodiment of the present disclosure refers to the owning user of aforementioned initial pictures.Specific
Realization scene in, target user may be regarded as corresponding with delivering or uploading or provide the user account of aforementioned initial pictures.
Therefore, in the embodiment of the present disclosure, referring to FIG. 4, this method can also include the following steps:
S107 determines target user at least one described candidate face image collection owning user.
At this point, aforementioned being adapted to property of S108 adjustment are as follows:
S1082 carries out age identifying processing to the candidate face image collection of the target user, obtains the target and use
The age recognition result at family.
Alternatively, can also execute S108 during another realization and then determine target user, then, from each candidate
In the age recognition result of face image set, the age recognition result of the candidate face image collection of target user is got.
Specifically, can be at least accomplished in that when performance objective user determines
In a kind of implementation, frequency of occurrence of each candidate face image in initial pictures can use, to determine mesh
Mark facial image.At this point, S107 can be embodied as: obtaining the candidate for including in each candidate face image collection
Then the number of face image obtains the highest candidate face image collection of the number, using as the target face figure
Image set closes.
This design allows for be there is more facial image and is generally user in the aforementioned primary data of user's upload
The facial image of oneself, therefore, the number occurred in initial pictures are more, which is more likely to be target user's
Face.
It should be noted that during practical determining target user in this mode, it can be according to preceding frequency of occurrence
Sequence from high to low is ranked up the candidate face image of each user, alternatively, can also mode by comparing get
The higher target facial image of frequency of occurrence.
Specifically, can be using the highest candidate face image of frequency of occurrence as target facial image.Alternatively, may be used also
Using by frequency of occurrence sort forward (frequency of occurrence is higher) M (M is the integer greater than 1) user candidate face image as
Target facial image.As previously mentioned, the continuous age knows after execution at this point, being related to the target facial image of multiple and different users
When other places are managed, it should be noted that the target facial image of each user is respectively processed, to improve recognition accuracy.
In another implementation, the prior information of the target user can also be obtained, then, obtains each candidate
Matching degree between face image set and the prior information, thus, obtain the highest candidate of the matching degree
Face image set, using as the target face image set.
Wherein, the prior information involved by the embodiment of the present disclosure can include but is not limited to: basic data with go through
History behavioral data.Basic data be with the basic data of primary data owning user, can include but is not limited to: user's head portrait；
In addition, it can include but be not limited to: account title, account exclusive identification code, contact details, the network information (such as network address
Deng).Wherein, user's head portrait can include but is not limited to: user is used to verify the true head portrait or identity head portrait of identity
When specifically obtaining the matching degree of each candidate face image collection and the prior information, can be obtained by aforementioned
It takes image feature vector and the mode for obtaining similarity between vector repeats no more to get matching degree.
By aforementioned schemes, technical solution provided by the embodiment of the present disclosure be can be realized for target user and image
In any one face owning user age identification.
On this basis, the embodiment of the present disclosure further provides further applying for aforementioned age recognition result.At this point, can
To refer to Fig. 5, this method can also include the following steps:
S110 judges whether to meet preset low age protective condition according to the age recognition result of each user；If so, holding
Row S112；If it is not, terminating.
Specifically, low age protective condition can be according to actual scene self-definition design.
In a kind of possible design, according to aforementioned age recognition result, if age bracket locating for target user is low age section,
Determination meets preset low age protective condition.
In alternatively possible design, according to aforementioned age recognition result, if each portrait that the initial pictures are included
In owning user, there is user's portrait in low age section, it is determined that meet preset low age protective condition.
S112 carries out low age protection processing to the initial pictures owning user account.
At this point, the age recognition result of each user meets the low age protective condition, then to target user's account into
Row low age protection processing.Specifically, following at least one low age protection processing, Lai Shixian S112 can be taken.
A kind of low age is protected in the mode of processing, and the operating right of the initial pictures owning user account can be limited.
Wherein, the operating right can include but is not limited to: delivering permission, browse right, deletes permission, storage permission
With comment at least one of permission.For example, the browse right of low age user can be limited, avoids it from browsing to and be not suitable for its year
The content of age section viewing.In another example the comment permission of low age user can be limited, it is internal to avoid the random comment of low age user
Hold the comment deviation of publisher.
In the mode of another low age protection processing, the monitoring pair of the available initial pictures owning user account
As, and send monitoring to the guardianship and remind.
Wherein, the contact method of guardianship can obtain in the basic data of target user.In actual reality
In live scape, the contact method of guardianship can be with are as follows: communicating number, and/or, the account contact method in application program, this
Open embodiment is not particularly limited this.
It, can be according to the age recognition result of each user, described in determination in the mode of another low age protection processing
The recommendation of initial pictures owning user account, and recommend the recommendation to the initial pictures owning user account.
It is, carrying out personalization according to the interested content of low age user if target user is low age user for it
The recommendation of content.For example, animation video more popular in low age user is recommended target user's account.
By aforementioned schemes, safeguard measure can be taken to low age user, low age user is avoided to touch with its age not
The content of adaptation, the growth of power-assisted low age user.
It is understood that step or operation are only example, the embodiment of the present application some or all of in above-described embodiment
The deformation of other operations or various operations can also be performed.In addition, each step can be presented not according to above-described embodiment
With sequence execute, and it is possible to do not really want to execute all operationss in above-described embodiment.
Word used herein is only used for description embodiment and is not used in limitation claim.Such as embodiment with
And used in the description of claim, unless context clearly illustrates, otherwise "one" (a) of singular, "one"
(an) and " described " (the) is intended to include equally plural form.Similarly, term "and/or" as used in this specification
Refer to comprising one or more associated any and all possible combinations listed.In addition, when being used for the application
When middle, term " includes " (comprise) and its modification " comprising " (comprises) and/or refer to including (comprising) etc. old
The presence of feature, entirety, step, operation, element and/or the component stated, but be not excluded for one or more other features,
Entirety, step, operation, element, component and/or these grouping presence or addition.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application should be based on the protection scope of the described claims.
Embodiment two
The age recognition methods based on image provided by one, the embodiment of the present disclosure further provide based on the above embodiment
Realize the Installation practice of each step and method in above method embodiment.
The embodiment of the present disclosure provides a kind of age identification device based on image, referring to FIG. 6, should the year based on image
Age identification device 600, comprising:
Module 61 is obtained, includes face in the initial pictures for obtaining initial pictures；
First processing module 62 obtains at least one candidate for carrying out recognition of face processing to the initial pictures
Face image set, the candidate face image in each candidate face image collection belong to a user；
Second processing module 63, for carrying out age knowledge respectively for one or more candidate face image collections
Other places reason, obtains the age recognition result of each user.
In a kind of possible design, first processing module 62 is specifically used for:
The initial pictures are handled using the first human face recognition model, obtain at least one candidate face image；
Face matching is carried out to each candidate face image, obtains at least one described candidate face set, Mei Gehou
The candidate face image in face image set is selected to belong to a user.
In alternatively possible design, first processing module 62 is specifically used for:
The initial pictures are handled using the second human face recognition model, obtain at least one candidate face image collection, often
Candidate face image in a candidate face image collection belongs to a user.
In alternatively possible design, Second processing module 63 is specifically used for:
Using trained age identification model, the one or more candidate face image collections of processing, are obtained respectively
The age recognition result of each user.
Wherein, the input data of the age identification model includes:
All or part of candidate face image in one or more candidate face image collections；Alternatively,
One or more candidate face features, each candidate face feature extraction is from a candidate face image
Set.
Wherein, the age recognition result includes following at least one:
The age of user；
Age level locating for user；
It whether include target age section user；
Whether target user is target age section user.
In addition, in alternatively possible design, the age identification device 600 based on image can be with further include:
Determining module (Fig. 6 is not shown), for determining at least one described candidate face image collection owning user
Target user.
In alternatively possible design, determining module is specifically used for:
Obtain the number for the candidate face image for including in each candidate face image collection；
The highest candidate face image collection of the number is obtained, using as the target face image set.
In alternatively possible design, determining module is specifically used for:
Obtain the prior information of the target user, wherein the prior information includes: basic data and historical behavior number
According to；
Obtain the matching degree between each candidate face image collection and the prior information；
The highest candidate face image collection of the matching degree is obtained, using as the target face image set
It closes.
In the embodiment of the present disclosure, module 61 is obtained, is specifically used for:
The behavioral data of target user is obtained, the behavioral data includes: to deliver in data, storing data and acquisition data
At least one.
At least frame image data in the behavioral data is obtained, using as the initial pictures.
In addition, in alternatively possible design, the age identification device 600 based on image can be with further include:
Judgment module (Fig. 6 is not shown) judges whether to meet preset low for the age recognition result according to each user
Age protective condition；
Third processing module (Fig. 6 is not shown), if the age recognition result for each user meets the low age and protects
Guard strip part carries out low age protection processing to the initial pictures owning user account.
Wherein, third processing module is specifically used for following at least one processing mode:
Limit the operating right of the initial pictures owning user account；The operating right includes: to deliver permission, browsing
Permission deletes permission, storage permission and comments at least one of permission；
The guardianship of the initial pictures owning user account is obtained, and sends monitoring to the guardianship and reminds；
According to the age recognition result of each user, the recommendation of the initial pictures owning user account is determined,
And recommend the recommendation to the initial pictures owning user account.
The age identification device 600 based on image of embodiment illustrated in fig. 6 can be used for executing the skill of above method embodiment
Art scheme, implementing principle and technical effect can be with further reference to the associated descriptions in embodiment of the method, and optionally, this is based on
The age identification device 600 of image can be terminal.
It should be understood that the division of the modules of the age identification device 600 based on image shown in figure 6 above is only one kind
The division of logic function can be completely or partially integrated on a physical entity in actual implementation, can also be physically separate.
And these modules can be realized all by way of processing element calls with software；It can also be all real in the form of hardware
It is existing；It can realize that part of module passes through formal implementation of hardware by way of processing element calls with part of module with software.
For example, Second processing module 63 can be the processing element individually set up, also can integrate in the age identification dress based on image
It sets in 600, such as is realized in some chip of terminal, in addition it is also possible to be stored in the year based on image in the form of program
In the memory of age identification device 600, is called and executed by some processing element of the age identification device 600 based on image
The function of the above modules.The realization of other modules is similar therewith.Furthermore these modules completely or partially can integrate one
It rises, can also independently realize.Processing element described here can be a kind of integrated circuit, the processing capacity with signal.?
During realization, each step of the above method or the above modules can pass through the integration logic of the hardware in processor elements
The instruction of circuit or software form is completed.
For example, the above module can be arranged to implement one or more integrated circuits of above method, such as:
One or more specific integrated circuits (Application Specific Integrated Circuit, ASIC), or, one
Or multi-microprocessor (digital singnal processor, DSP), or, one or more field programmable gate array
(Field Programmable Gate Array, FPGA) etc..For another example, when some above module dispatches journey by processing element
When the form of sequence is realized, which can be general processor, such as central processing unit (Central Processing
Unit, CPU) or it is other can be with the processor of caller.For another example, these modules can integrate together, with system on chip
The form of (system-on-a-chip, SOC) is realized.
Also, the embodiment of the present disclosure provides a kind of age identification device based on image, referring to FIG. 7, should be based on figure
The age identification device 600 of picture, comprising:
Memory 610；
Processor 620；And
Computer program；
Wherein, computer program is stored in memory 610, and is configured as being executed by processor 620 to realize as above
State method described in embodiment.
Wherein, the number of processor 620 can be one or more, processing in the age identification device 600 based on image
Device 620 is referred to as processing unit, and certain control function may be implemented.The processor 620 can be general processor
Or application specific processor etc..In a kind of optionally design, processor 620 can also have instruction, and described instruction can be by institute
The operation of processor 620 is stated, so that the age identification device 600 based on image executes side described in above method embodiment
Method.
In another possible design, the age identification device 600 based on image may include circuit, and the circuit can
To realize the function of sending or receiving or communicate in preceding method embodiment.
Optionally, the number of memory 610 can be one or more in the age identification device 600 based on image
It is a, there are instruction or intermediate data on memory 610, described instruction can be run on the processor 620, so that described
Age identification device 600 based on image executes method described in above method embodiment.Optionally, the memory 610
In can also be stored with other related datas.Optionally it also can store instruction and/or data in processor 620.The processing
Device 620 and memory 610 can be separately provided, and also can integrate together.
In addition, as shown in fig. 7, being additionally provided with transceiver 630 in the age identification device 600 based on image, wherein
The transceiver 630 is properly termed as Transmit-Receive Unit, transceiver, transmission circuit or transceiver etc., for test equipment or its
His terminal device carries out data transmission or communicates, and details are not described herein.
As shown in fig. 7, memory 610, processor 620 are connected and communicated with transceiver 630 by bus.
If be somebody's turn to do the age identification device 600 based on image for realizing the method corresponded in Fig. 2, processor 620 is used
In completing corresponding determining or control operation, optionally, corresponding instruction can also be stored in memory 610.Each portion
The specific processing mode of part can refer to the associated description of previous embodiment.
In addition, in another possible design, referring to FIG. 8, should be based on may be used also in the age identification device 600 of image
To be further arranged: image collecting device 640；
Wherein, image collecting device 640, for acquiring the initial pictures.
Wherein, image collecting device 640 includes the device that can arbitrarily collect multi-media image, such as camera.
In addition, the embodiment of the present disclosure provides a kind of readable storage medium storing program for executing, it is stored thereon with computer program, the computer
Program is executed by processor to realize the method as described in embodiment one.
And the embodiment of the present disclosure provides a kind of terminal, referring to FIG. 9, the terminal 900 includes: the year based on image
Age identification device 600 and terminal body 910.Wherein, the age identification device 600 based on image is for executing as embodiment one is appointed
Age recognition methods described in one implementation based on image.
Wherein, image collecting device (such as camera) and display device are generally also configured in terminal body 910 (as shown
Screen) etc..At this point, the image collecting device in the age identification device 600 based on image as shown in Figure 8 can be with multiplex terminal
Some equipment.
The component that the embodiment of the present disclosure is included for terminal body 910 is not particularly limited.In a kind of actual realization field
Jing Zhong, may include following one or more components: processing component, memory, power supply module, multimedia component, audio component,
Input/output (I/O) interface, sensor module and communication component.
And terminal 900 involved by the embodiment of the present disclosure can be wireless terminal and be also possible to catv terminal.It is wireless whole
End can be directed to target user and provide voice and/or the equipment of other business datum connectivity, with wireless connecting function
Handheld device or other processing equipments for being connected to radio modem.Wireless terminal can be through wireless access network
(Radio Access Network, abbreviation RAN) is communicated with one or more equipments of the core network, and wireless terminal can be shifting
Dynamic terminal, such as mobile phone (or being " honeycomb " phone) and the computer with mobile terminal, for example, it may be portable,
Pocket, hand-held, built-in computer or vehicle-mounted mobile device, they exchange language and/or number with wireless access network
According to.For another example wireless terminal can also be personal communication service (Personal Communication Service, abbreviation
PCS) phone, wireless phone, Session initiation Protocol (Session Initiation Protocol, abbreviation SIP) phone, wireless
Local loop (Wireless Local Loop, abbreviation WLL) stands, personal digital assistant (Personal Digital
Assistant, abbreviation PDA) etc. equipment.Wireless terminal is referred to as system, subscriber unit (Subscriber Unit), orders
Family station (Subscriber Station), movement station (Mobile Station), mobile station (Mobile), distant station (Remote
Station), remote terminal (Remote Terminal), access terminal (Access Terminal), target terminal user
(User Terminal), destination user agent (User Agent), target UE (User Device or User
Equipment), it is not limited thereto.Optionally, above-mentioned terminal device can also be the equipment such as smartwatch, tablet computer.
Method shown in embodiment one is able to carry out as each module in this present embodiment, what the present embodiment was not described in detail
Part can refer to the related description to embodiment one.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the disclosure
Its embodiment.The disclosure is intended to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Person's adaptive change follows the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by following
Claims are pointed out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure is only limited by appended claims
System.
Claims (16)
1. a kind of age recognition methods based on image characterized by comprising
Initial pictures are obtained, include face in the initial pictures；
Recognition of face processing is carried out to the initial pictures, obtains at least one candidate face image collection, each candidate face
Candidate face image in image collection belongs to a user；
For one or more candidate face image collections, age identifying processing is carried out respectively, obtains the age of each user
Recognition result.
2. the method according to claim 1, wherein it is described to the initial pictures carry out recognition of face processing,
Obtain at least one candidate face image collection, comprising:
The initial pictures are handled using the first human face recognition model, obtain at least one candidate face image；
Face matching is carried out to each candidate face image, obtains at least one described candidate face set, each candidate
Candidate face image in face image set belongs to a user.
3. the method according to claim 1, wherein it is described to the initial pictures carry out recognition of face processing,
Obtain at least one candidate face image collection, comprising:
The initial pictures are handled using the second human face recognition model, obtain at least one candidate face image collection, Mei Gehou
The candidate face image in face image set is selected to belong to a user.
4. the method according to claim 1, wherein described for one or more candidate face image sets
It closes, carries out age identifying processing respectively, comprising:
Using trained age identification model, the one or more candidate face image collections of processing, obtain each use respectively
The age recognition result at family.
5. according to the method described in claim 4, it is characterized in that, the input data of the age identification model includes:
All or part of candidate face image in one or more candidate face image collections；Alternatively,
One or more candidate face features, each candidate face feature extraction is from a candidate face image set
It closes.
6. according to the method described in claim 4, it is characterized in that, the age recognition result includes following at least one:
The age of user；
Age level locating for user；
It whether include target age section user；
Whether target user is target age section user.
7. method according to claim 1-6, which is characterized in that the method also includes:
In at least one described candidate face image collection owning user, target user is determined.
8. method according to claim 7, which is characterized in that described belonging at least one described candidate face image collection
In user, target user is determined, comprising:
Obtain the number for the candidate face image for including in each candidate face image collection；
The highest candidate face image collection of the number is obtained, using as the target face image set.
9. method according to claim 7, which is characterized in that described belonging at least one described candidate face image collection
In user, target user is determined, comprising:
Obtain the prior information of the target user, wherein the prior information includes: basic data and historical behavior data；
Obtain the matching degree between each candidate face image collection and the prior information；
The highest candidate face image collection of the matching degree is obtained, using as the target face image set.
10. method according to claim 1-6, which is characterized in that the acquisition initial pictures, comprising:
The behavioral data of target user is obtained, the behavioral data includes: to deliver in data, storing data and acquisition data extremely
Few one kind；
At least frame image data in the behavioral data is obtained, using as the initial pictures.
11. method according to claim 1-6, which is characterized in that the method also includes:
According to the age recognition result of each user, judge whether to meet preset low age protective condition；
If the age recognition result of each user meets the low age protective condition, to the initial pictures owning user account
Carry out low age protection processing.
12. according to the method for claim 11, which is characterized in that described to be carried out to the initial pictures owning user account
Low age protection processing, including following at least one:
Limit the operating right of the initial pictures owning user account；The operating right includes: to deliver permission, browsing power
Limit deletes permission, storage permission and comments at least one of permission；
The guardianship of the initial pictures owning user account is obtained, and sends monitoring to the guardianship and reminds；
According to the age recognition result of each user, the recommendation of the initial pictures owning user account is determined, and to
The initial pictures owning user account recommends the recommendation.
13. a kind of age identification device based on image characterized by comprising
Module is obtained, includes face in the initial pictures for obtaining initial pictures；
First processing module obtains at least one candidate face image for carrying out recognition of face processing to the initial pictures
Gather, the candidate face image in each candidate face image collection belongs to a user；
Second processing module, for carrying out age identifying processing respectively for one or more candidate face image collections,
Obtain the age recognition result of each user.
14. a kind of age identification device based on image characterized by comprising
Memory；
Processor；And
Computer program；
Wherein, the computer program stores in the memory, and is configured as being executed by the processor to realize such as
The described in any item methods of claim 1-12.
15. a kind of computer readable storage medium, which is characterized in that it is stored thereon with computer program,
The computer program is executed by processor to realize such as the described in any item methods of claim 1-12.
16. a kind of terminal characterized by comprising
Age identification device based on image, for realizing such as described in any item methods of claim 1-12；
Terminal body.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910615158.2A CN110334663A (en) | 2019-07-09 | 2019-07-09 | Age recognition methods and device, storage medium and terminal based on image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910615158.2A CN110334663A (en) | 2019-07-09 | 2019-07-09 | Age recognition methods and device, storage medium and terminal based on image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110334663A true CN110334663A (en) | 2019-10-15 |
Family
ID=68144839
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910615158.2A Pending CN110334663A (en) | 2019-07-09 | 2019-07-09 | Age recognition methods and device, storage medium and terminal based on image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110334663A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112016623A (en) * | 2020-08-28 | 2020-12-01 | 佳都新太科技股份有限公司 | Face clustering method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150089370A (en) * | 2014-01-27 | 2015-08-05 | 주식회사 에스원 | Age Cognition Method that is powerful to change of Face Pose and System thereof |
| CN109034078A (en) * | 2018-08-01 | 2018-12-18 | 腾讯科技（深圳）有限公司 | Training method, age recognition methods and the relevant device of age identification model |
| CN109190449A (en) * | 2018-07-09 | 2019-01-11 | 北京达佳互联信息技术有限公司 | Age recognition methods, device, electronic equipment and storage medium |
| CN109885461A (en) * | 2019-01-07 | 2019-06-14 | 武汉游戏群科技有限公司 | A kind of system of Anti-addiction, method and smart machine |
-
2019
- 2019-07-09 CN CN201910615158.2A patent/CN110334663A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150089370A (en) * | 2014-01-27 | 2015-08-05 | 주식회사 에스원 | Age Cognition Method that is powerful to change of Face Pose and System thereof |
| CN109190449A (en) * | 2018-07-09 | 2019-01-11 | 北京达佳互联信息技术有限公司 | Age recognition methods, device, electronic equipment and storage medium |
| CN109034078A (en) * | 2018-08-01 | 2018-12-18 | 腾讯科技（深圳）有限公司 | Training method, age recognition methods and the relevant device of age identification model |
| CN109885461A (en) * | 2019-01-07 | 2019-06-14 | 武汉游戏群科技有限公司 | A kind of system of Anti-addiction, method and smart machine |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112016623A (en) * | 2020-08-28 | 2020-12-01 | 佳都新太科技股份有限公司 | Face clustering method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110321863A (en) | Age recognition methods and device, storage medium | |
| US11477268B2 (en) | Sharing digital media assets for presentation within an online social network | |
| US20220062707A1 (en) | Privacy Preserving Personalized Workout Recommendations | |
| CN106326391B (en) | Multimedia resource recommendation method and device | |
| CN106604051A (en) | Live channel recommending method and device | |
| CN109660823A (en) | Video distribution method, apparatus, electronic equipment and storage medium | |
| CN103970830B (en) | Information recommendation method and device | |
| CN108551462A (en) | A kind of information-pushing method, device and equipment | |
| CN103412951A (en) | Individual-photo-based human network correlation analysis and management system and method | |
| JP2006344215A (en) | Method for assembling collection of digital picture | |
| US20140006418A1 (en) | Method and apparatus for ranking apps in the wide-open internet | |
| CN109101658A (en) | Information search method, device and equipment/terminal/server | |
| CN107465886A (en) | Intermediary method, mediating device and system and record media | |
| CN110659581A (en) | Image processing method, device, equipment and storage medium | |
| CN111191133B (en) | Service search processing method, device and equipment | |
| CN110298405A (en) | Classification recognition methods and device, storage medium and terminal | |
| CN107368550A (en) | Information acquisition method, device, medium, electronic equipment, server and system | |
| CN109727208A (en) | Filter recommended method, device, electronic equipment and storage medium | |
| WO2012109330A1 (en) | Content storage management in cameras | |
| WO2013079124A1 (en) | Portable electronic equipment and method of recording media using a portable electronic equipment | |
| CN105893562A (en) | Conversation message processing method and device and terminal | |
| CN110335626A (en) | Age recognition methods and device, storage medium based on audio | |
| CN108665347A (en) | Products Show method, apparatus, equipment and storage medium based on user demand | |
| CN110852224B (en) | Expression recognition method and related device | |
| CN110990625A (en) | Movie recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |