CN110390229A - A kind of face picture screening technique, device, electronic equipment and storage medium - Google Patents
A kind of face picture screening technique, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110390229A CN110390229A CN201810360907.7A CN201810360907A CN110390229A CN 110390229 A CN110390229 A CN 110390229A CN 201810360907 A CN201810360907 A CN 201810360907A CN 110390229 A CN110390229 A CN 110390229A
- Authority
- CN
- China
- Prior art keywords
- picture
- detected
- faceform
- threshold value
- human face
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
- G06V20/653—Three-dimensional objects by matching three-dimensional models, e.g. conformal mapping of Riemann surfaces
-
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides a kind of face picture screening technique, device, electronic equipment and storage mediums, which is characterized in that the described method includes: obtaining the picture to be detected comprising human face target；Human face target in picture to be detected is compared with the three-dimensional references faceform in default three-dimensional references database, determines the characteristic value of human face target；Picture to be detected is divided into different classes of according to comparison result by the numerical values recited between comparative feature value and preset threshold, and wherein classification includes high quality picture and normal quality picture.By the way that characteristic value compared with the carry out numerical values recited between preset threshold, can more accurately be measured out the requirement whether picture to be detected meets subsequent operation, so that more accurate to the screening of picture to be detected.
Description
Technical field
The present invention relates to technical field of image processing, set more particularly to a kind of face picture screening technique, device, electronics
Standby and storage medium.
Background technique
With the continuous development of internet and intelligent hardware devices technology, face recognition technology has been applied to more and more
Field, such as security protection, finance, amusement etc..More and more conveniences and enjoyment are provided for people's lives.Pass through face
Identification technology can identify the face in true face or picture, so that it is determined that corresponding identity information etc. out,
And then subsequent response can be executed, such as information registration, payment operation or process.Especially in safety-security area, recognition of face skill
Art can quickly identify the identity of people, to be conducive to the development of security protection work.
In the identity information determined by face recognition technology, video camera, camera, camera etc. can be passed through first
Capture apparatus obtains the picture containing face, and in practical application, can generally obtain multiple pictures containing face, so
Multiple face picture can be screened or be classified by certain rule afterwards, to select the face figure to suit the requirements
Then piece carries out subsequent operation or process to the face picture selected again.
However, in the prior art, when being screened to face picture, it is general only according to the clarity of face picture come
It is screened, so the face picture that can meet subsequent operation requirement can not accurately be selected in many cases, thus
The operation or process that can not accurately continue after the completion.For example, when the face in face picture bows or comes back, although picture is clear
Clear degree is very high, but since the angle that face bows or comes back is excessive, causes that face is subregional to be blocked, passes through the face
Picture can accurately not identify the corresponding identity information of face.
Summary of the invention
Be designed to provide a kind of face picture screening technique, device, electronic equipment and the storage of the embodiment of the present invention are situated between
Matter can more accurately screen face picture.Specific technical solution is as follows:
The embodiment of the invention provides a kind of face picture screening techniques, comprising:
Obtain the picture to be detected comprising human face target；
By the human face target in the picture to be detected and the three-dimensional references in the default three-dimensional references database
Faceform is compared, and determines the characteristic value of the human face target；
Compare the numerical values recited between the characteristic value and preset threshold, according to comparison result, by the picture to be detected
It is divided into different classes of, the classification includes high quality picture and normal quality picture.
Optionally, in the human face target by the picture to be detected and the default three-dimensional references database
Three-dimensional references faceform be compared, determine the characteristic value of the human face target, comprising:
By the human face target in the picture to be detected, respectively with it is multiple in the default three-dimensional references database
Three-dimensional references faceform is compared respectively, wherein the multiple three-dimensional references faceform is three with different characteristic value
The quasi- faceform of Wiki；
Determine the three-dimensional references faceform to match with the human face target；
Characteristic value by characteristic value possessed by the three-dimensional references faceform, as the human face target.
Optionally, the characteristic value includes binocular interpupillary distance, pitch angle and left and right side face angle；
The preset threshold includes binocular interpupillary distance threshold value, pitch angle threshold value and left and right side face angle threshold.
Optionally, the numerical values recited between characteristic value and preset threshold will be described according to comparison result
Picture to be detected is divided into different classes of, comprising:
When the binocular interpupillary distance be greater than the binocular interpupillary distance threshold value, and the pitch angle be less than the pitch angle threshold
Value, and the left and right side face angle be less than the left and right side face angle threshold when, the picture to be detected be high quality picture；
Otherwise, the picture to be detected is normal quality picture.
Optionally, the binocular interpupillary distance threshold value includes the first binocular interpupillary distance threshold value and the second binocular interpupillary distance threshold value, described to bow
Facing upward angle threshold includes the first pitch angle threshold value and the second pitch angle threshold value, and the left and right side face angle threshold includes first
Left and right side face angle threshold and second or so side face angle thresholds, the classification further include mean quality picture；
The numerical values recited between characteristic value and preset threshold will be described to be detected according to comparison result
Picture is divided into different classes of, comprising:
When the binocular interpupillary distance be greater than the first binocular interpupillary distance threshold value, and the pitch angle be less than first pitching
Angle threshold, and the left and right side face angle be less than the first or so side face angle threshold when, the picture to be detected be height
Quality picture；
When the binocular interpupillary distance be less than the first binocular interpupillary distance threshold value and be greater than the second binocular interpupillary distance threshold value,
And the pitch angle is greater than the first pitch angle threshold value and is less than the second pitch angle threshold value,
And the left and right side face angle is greater than the first or so side face angle threshold and is less than second or so the side face
When angle threshold, the picture to be detected is mean quality picture；
Otherwise, the picture to be detected is normal quality picture.
Optionally, in the human face target by the picture to be detected and the default three-dimensional references database
Three-dimensional references faceform be compared, determine the characteristic value of the human face target, comprising:
The human face target is identified in the picture to be detected；
The human face target is compared with the three-dimensional references faceform in the default three-dimensional references database, really
Make the characteristic value of the human face target.
Optionally, the method also includes:
Establish the corresponding faceform of the human face target in the picture to be detected；
When the picture to be detected is high quality picture, the faceform is inputted into presetting database, and it is described pre-
If each default faceform stored in database is compared；
When there is the default faceform to match with the faceform in the presetting database, alarm signal is exported
Breath；
Alternatively, when the default faceform to match with the faceform is not present in the presetting database, it is defeated
Warning message out.
The embodiment of the invention also provides a kind of face picture screening plants, comprising:
Module is obtained, for obtaining the picture to be detected comprising human face target；
Comparison module, for by the picture to be detected the human face target and the default three-dimensional references database
In three-dimensional references faceform be compared, determine the characteristic value of the human face target；
Screening module, for the numerical values recited between the characteristic value and preset threshold, according to comparison result, by institute
It states picture to be detected and is divided into different classes of, the classification includes high quality picture and normal quality picture.
Optionally, the comparison module, is specifically used for:
By the human face target in the picture to be detected, respectively with it is multiple in the default three-dimensional references database
Three-dimensional references faceform is compared respectively, wherein the multiple three-dimensional references faceform is three with different characteristic value
The quasi- faceform of Wiki；Determine the three-dimensional references faceform to match with the human face target；By the three-dimensional references people
Characteristic value possessed by face model, the characteristic value as the human face target.
Optionally, the characteristic value in described device includes binocular interpupillary distance, pitch angle and left and right side face angle；
The preset threshold includes binocular interpupillary distance threshold value, pitch angle threshold value and left and right side face angle threshold.
Optionally, the screening module, is specifically used for:
When the binocular interpupillary distance be greater than the binocular interpupillary distance threshold value, and the pitch angle be less than the pitch angle threshold
Value, and the left and right side face angle be less than the left and right side face angle threshold when, the picture to be detected be high quality picture；It is no
Then, the picture to be detected is normal quality picture.
Optionally, the binocular interpupillary distance threshold value in the comparison module includes the first binocular interpupillary distance threshold value and the second binocular
Interpupillary distance threshold value, the pitch angle threshold value include the first pitch angle threshold value and the second pitch angle threshold value, the left and right side face
Angle threshold includes the first or so side face angle threshold and second or so side face angle thresholds, the class in the screening module
It does not further include mean quality picture；
The screening module, is specifically used for:
When the binocular interpupillary distance be greater than the first binocular interpupillary distance threshold value, and the pitch angle be less than first pitching
Angle threshold, and the left and right side face angle be less than the first or so side face angle threshold when, the picture to be detected be height
Quality picture；
When the binocular interpupillary distance is less than the first binocular interpupillary distance threshold value and is greater than the second binocular interpupillary distance threshold value, and institute
Pitch angle is stated greater than the first pitch angle threshold value and is less than the second pitch angle threshold value, and left and right side face angle
When degree is greater than the first or so side face angle threshold and is less than the second or so side face angle threshold, the picture to be detected
For mean quality picture；
Otherwise, the picture to be detected is normal quality picture.
Optionally, the comparison module, is specifically used for:
The human face target is identified in the picture to be detected；
The human face target is compared with the three-dimensional references faceform in the default three-dimensional references database, really
Make the characteristic value of the human face target.
Optionally, described device further include:
Modeling module, the corresponding faceform of the human face target for establishing in the picture to be detected；
Analysis module, for when the picture to be detected is high quality picture, the faceform to be inputted present count
According to library, it is compared with each default faceform stored in the presetting database；
Alarm module, for working as the default faceform for existing in the presetting database and matching with the faceform
When, export warning message；Alternatively, when there is no the default face moulds to match with the faceform in the presetting database
When type, warning message is exported.
The embodiment of the invention also provides a kind of electronic equipment, including processor, communication interface, memory and communication are total
Line, wherein processor, communication interface, memory complete mutual communication by communication bus；
Memory, for storing computer program；
Processor when for executing the program stored on memory, realizes any of the above-described face picture screening
Method.
At the another aspect that the present invention is implemented, a kind of computer readable storage medium is additionally provided, it is described computer-readable
Instruction is stored in storage medium, when run on a computer, so that computer executes any of the above-described face figure
Piece screening technique.
At the another aspect that the present invention is implemented, the embodiment of the invention also provides a kind of, and the computer program comprising instruction is produced
Product, when run on a computer, so that computer executes any of the above-described face picture screening technique.
A kind of face picture screening technique, device, electronic equipment and storage medium provided in an embodiment of the present invention, pass through by
Three-dimensional references faceform in human face target in acquired picture to be detected, with default three-dimensional references database compares
It is right, the characteristic value of human face target is obtained, then again by this feature value compared with the carry out numerical values recited between preset threshold, from
And determining picture to be detected is high quality picture or normal quality picture.Due in embodiments of the present invention, to be detected
It when picture is screened, can be screened by various characteristic values, rather than only only considered one factor of clarity,
So as to more accurately measure out the requirement whether picture to be detected meets subsequent operation, so as to the sieve of picture to be detected
It is more accurate to select.Also, since preset threshold can be configured in advance, it is possible to more flexible adjustment picture to be detected
Screening criteria, to adapt to the application under different actual conditions or scene.Certainly, it implements any of the products of the present invention or method must
It is not necessarily required to reach all the above advantage simultaneously.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow chart of face picture screening technique provided in an embodiment of the present invention；
Fig. 2 is another flow chart of face picture screening technique provided in an embodiment of the present invention；
Fig. 3 is the structure chart of face picture screening plant provided in an embodiment of the present invention；
Fig. 4 is the structure chart of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Referring to Fig. 1, Fig. 1 is a kind of flow chart of face picture screening technique provided in an embodiment of the present invention, comprising:
Step 101, the picture to be detected comprising human face target is obtained.
The embodiment of the present invention can be used for all kinds of electronic equipments with data-handling capacity, can by the electronic equipment
Image data is handled, for example, electronic equipment can be mobile phone, capture machine, camera, server, monitoring camera and
The monitoring system etc. that monitoring camera is connected.
Picture to be detected can be image or picture containing one or more human face targets, and human face target is to refer to instead
Mirror the image of the face feature of people.For example, picture to be detected can be mounted to the capture machine of particular place entrance at any time
The picture of candid photograph contains one or more human face targets in the picture；Alternatively, picture to be detected can also be monitoring camera institute
The monitored picture of shooting, wherein the human face target in monitoring scene can be contained.
Electronic equipment acquisition picture to be detected can there are many modes, for example, picture or electronics to be detected is manually entered
Equipment can also directly acquire the picture to be detected, for example, containing the picture of human face target captured by camera or capture machine,
It can be directly as the picture to be detected.
Step 102, by the human face target in picture to be detected and the three-dimensional references face in default three-dimensional references database
Model is compared, and determines the characteristic value of human face target.
Characteristic value is the parameter value for being able to reflect out the feature of human face target, and characteristic value can there are many types, for example, special
Value indicative can for the position angle of face, the size of human face target, the positional relationship of characteristic point in face, in human face target
Ratio, clarity between face etc..
After obtaining picture to be detected, which can be input in default three-dimensional references database.In
In default three-dimensional references database, it is stored with a large amount of three-dimensional references faceform.Three-dimensional references faceform is to pass through acquisition
Image, various differences when the direct picture of benchmark face, the side image of different angle, different new line angles bow angle when
Image and various different shooting distances captured by image, then by orthogonal pictorial manner, the three-dimensional references established out
Faceform.So each benchmark face model can have the spies such as different sizes, position angle or position and attitude
Value indicative.
So specifically, in embodiments of the present invention, step 102, by the human face target and default three in picture to be detected
Three-dimensional references faceform in dimension benchmark database is compared, and determines the characteristic value of human face target, may include:
Step 102a, by the human face target in picture to be detected, respectively with multiple three in default three-dimensional references database
The quasi- faceform of Wiki is compared respectively, and plurality of three-dimensional references faceform is the three-dimensional references with different characteristic value
Faceform.
Step 102b determines the three-dimensional references faceform to match with the human face target.
Human face target is compared with three-dimensional references faceform one by one, so as to determine and the human face target
Size, position angle or position and attitude etc., closest three-dimensional references faceform, and by three-dimensional references faceform
As the three-dimensional references faceform to match with human face target.
Also, in practical applications, in order to faster carrying out of human face target Yu three-dimensional references faceform
Match, the human face target can be directed to, establish the corresponding faceform of the human face target, then pass through the faceform and multiple three
The quasi- faceform of Wiki is compared, to more efficiently determine the three-dimensional references face mould to match with the human face target
Type.
Step 102c, the feature by characteristic value possessed by the three-dimensional references faceform, as the human face target
Value.
After the three-dimensional references faceform for determining to match with the human face target, so that it may the three-dimensional that this matches
Characteristic value possessed by benchmark face model, as the face clarification of objective value, so that it is determined that going out the face clarification of objective
Value.
Step 103, the numerical values recited between comparative feature value and preset threshold, according to comparison result, by picture to be detected
It is divided into different classes of, classification includes high quality picture and normal quality picture.
After picture to be detected is inputted default three-dimensional references database, three-dimensional references database is preset by this, can be obtained
The characteristic value of human face target into the picture to be detected.After having obtained characteristic value, so that it may using this feature value and preset
Threshold value is compared.
Preset threshold is the threshold value pre-set, by comparing the size of characteristic value and preset threshold, it will be able to weigh
It measures characteristic value and whether meets whether required standard or characteristic value are in suitable range.When meeting standard or be in
When suitable range, then it represents that this feature is worth corresponding human face target and meets the condition or requirement for executing follow-up process or operation,
Picture to be detected where may thereby determine that out the human face target is the picture of high quality, conversely, then the picture to be detected is
The picture of high quality is the picture of normal quality.
Specifically, corresponding different types of characteristic value, can there is different preset thresholds.When the type of characteristic value is different
When, correspondingly, preset threshold is also different, and specific manner of comparison is also different.
For example, preset threshold can be an angle value, when more than the angle when characteristic value is the angle that face comes back
When, then it is unsatisfactory for requiring, can not accurately identify the corresponding identity information of the human face target, so when characteristic value is pre- greater than this
If when threshold value, then it represents that characteristic value has been more than the angle, then the picture to be detected is normal quality picture.
Alternatively, preset threshold can be an area value, when less than the area when characteristic value is the area of human face target
When, then it is unsatisfactory for requiring, so when the preset threshold, then the picture to be detected is normal quality picture under characteristic value.
So needing to be determined according to different feature Value Types various when characteristic value is compared with preset threshold
In the case where different comparison results, picture to be detected is high quality picture or normal quality picture.
In embodiments of the present invention, by by the human face target in acquired picture to be detected, with default three-dimensional references
Three-dimensional references faceform in database is compared, and obtains the characteristic value of human face target, then again by this feature value and pre-
If the comparison of the carry out numerical values recited between threshold value, so that it is determined that picture to be detected is high quality picture or normal quality figure out
Piece.Due to that in embodiments of the present invention, when being screened to picture to be detected, can be sieved by various characteristic values
Choosing, rather than only only considered one factor of clarity, so as to more accurately measure out whether picture to be detected accords with
The requirement for closing subsequent operation, so that more accurate to the screening of picture to be detected.Also, since preset threshold can carry out in advance
Configuration, it is possible to the screening criteria of more flexible adjustment picture to be detected, to adapt under different actual conditions or scene
Using.
After filtering out the picture to be detected is high quality picture or normal quality picture, so that it may according to picture to be detected
Different classification, to carry out corresponding subsequent affected or process.So in order to more easily execute subsequent step, referring to figure
2, in face picture screening technique provided in an embodiment of the present invention, further includes:
Step 104, the corresponding faceform of human face target in picture to be detected is established.
Faceform is using the face characteristic in human face target, and e.g., relative position and ratio between human face five-sense-organ are closed
System etc., for digital model constructed by the human face target.It, can be to the face realization of goal by the digital model
Various operations or processing, for example, carrying out person-portion validation of information etc. by digital model.
The step for corresponding faceform is established to the human face target in picture to be detected, can obtain it is to be detected
Any stage after picture carries out, for example, can just be directed to the picture to be detected immediately after obtaining picture to be detected
In human face target establish corresponding faceform, so that each picture to be detected can obtain a corresponding face mould
Type；Alternatively, can also filter out whether picture to be detected is to be carried out after high quality picture again, so as to just for screening
Out, classification is that the picture to be detected of high quality picture establishes corresponding faceform, reduces data processing amount and improves operation
Efficiency.Certainly, if when human face target is compared with the three-dimensional references faceform in default three-dimensional references data,
The corresponding faceform of the human face target is established, then can also directly apply the faceform, does not need to repeat to establish the people
The corresponding faceform of face target.
If the classification of picture to be detected be high quality picture, then it represents that the human face target in the picture to be detected meet into
Row subsequent operation or the requirement of process.It is able to carry out subsequent operation or process.It, can be with specifically, in embodiments of the present invention
The identification of personnel identity is carried out using the picture to be detected of the high quality, and according to all kinds of identity identified, is alarmed.
So in face picture screening technique provided in an embodiment of the present invention, further includes:
Step 105, when picture to be detected is high quality picture, faceform is inputted into presetting database, with present count
It is compared according to each default faceform stored in library.
Presetting database is to preserve the database of a large amount of personal information, largely presets faceform wherein containing,
Each default corresponding identity information of faceform.Default faceform is the acquisition by early period to personnel's face information
And the model established can acquire the face information of school internal staff for example, being directed to this application scenarios of school in advance, it is raw
It at respective faceform, and is stored in presetting database, which can be used as the people of school internal staff
Face model database.
After having obtained the human face target in picture to be detected and corresponding to faceform, so that it may by the face mode input
It into presetting database, is then compared with each default faceform in presetting database, so as to determine that this is default
With the presence or absence of the default faceform to match with the faceform in database.
Specifically, can be compared by the way of deep learning algorithm, by deep learning algorithm, available people
The similarity between each default faceform in face model and presetting database, as the faceform and a default face
When the similarity of model is greater than a threshold value pre-set, it may be considered that the faceform and the default faceform
Match.Certainly, in embodiments of the present invention, can also carry out between faceform in other way in the prior art
Comparison, can according to need and selected, and be applied to the embodiment of the present invention in.
Step 106, when there is the default faceform to match with faceform in presetting database, alarm signal is exported
Breath；Alternatively, when, there is no when the default faceform to match with faceform, exporting warning message in presetting database.
When presetting database is black list database, preset when existing in presetting database with what faceform matched
Faceform, then it represents that the corresponding human face target of the faceform matches with the identity information in the black list database, institute
Can be warned for the human face target alert in acquired picture to be detected for example, being issued in monitoring system
Report, or alarm mail or short message are sent to the relevant personnel.
Alternatively, when presetting database is white list database, for example, in example above-mentioned, presetting database is
When faceform's database of school internal staff.If there is no the default people to match with faceform in presetting database
Face model, then it represents that the human face target in acquired picture to be detected is non-school internal staff, strange so as to trigger
People's alarm.Inform in monitoring range non-campus internal staff occur in time to the relevant personnel.
In embodiments of the present invention, due to being screened first to picture to be detected, it is higher to select picture quality
Picture to be detected, is compared and identification in the preset database.To improve the accuracy rate of comparison and identification,
Avoid in the prior art, due to being all compared and identifying for all pictures to be detected, caused by operational efficiency it is low
Problem.Also, also solve and only select to need the picture that is compared by clarity, caused by compare and identification not
Accurate problem reduces the rate of false alarm of alarm.
Simultaneously as in the embodiment of the present invention picture to be detected can be adjusted by configuring different preset thresholds
Scale or screening criteria are screened, so as to be directed to different application scenarios, flexibly screens picture to be detected.For example, it is desired to drop
When low rate of failing to report, preset threshold can be set it is more loose, so as to so that more pictures to be detected by screening, into
The step of subsequent comparison of row or identification, and then reduce rate of failing to report；When needing reduces rate of false alarm, then preset threshold can be set
It is fixed more stringent so that the quality of the picture to be detected filtered out further increases, thus more accurately into
The subsequent comparison of row or identification, and then reduce rate of false alarm.
In practical applications, in order to more accurately judging the classification of picture to be detected, institute by characteristic value
State the characteristic value that uses can be it is multiple, i.e., be compared by multiple characteristic values with multiple corresponding preset thresholds, so as to
It is enough that picture to be detected more accurately measure or evaluate, it is more accurately screened and determines picture to be detected for height
Quality picture or normal quality picture.
So in embodiments of the present invention, characteristic value may include binocular interpupillary distance, pitch angle and left and right side face angle.
Correspondingly, preset threshold includes: binocular interpupillary distance threshold value, pitch angle threshold value and left and right side face angle threshold.
Wherein binocular interpupillary distance be human face target in the distance between two, the binocular interpupillary distance according to electronic equipment be taken
The distance between human face target it is far and near and change, when the distance between electronic equipment and human face target are remoter, then the binocular pupil
Away from smaller.So in order to keep the human face target in picture to be detected clearer, binocular interpupillary distance it is more big in a certain range more
It is good.For example, the range of binocular interpupillary distance generally can be with 18 to 571 pixel units, the more big then table of binocular interpupillary distance in the range
Show that human face target is clearer.
Pitch angle refers to the angle that the human face target that is taken comes back or bows, in order to when human face target it is more clear
Chu, more easy to identify, coverage extent is smaller, and the pitch angle is the smaller the better.When the pitch angle is 0, indicate that human face target does not have
It bows or comes back.The range of pitch angle generally can be 90 degree to 90 degree of ﹣, wherein positive and negative respectively indicate new line or bow.
Left and right side face angle refers to the angle turned over when the human face target being taken carries out rotary head to the left or to the right, together
Sample, the left and right side face angle are the smaller the better.When the left and right side face angle is smaller, then it represents that the closer front of human face target,
So that the feature of the appearance of human face target shows more, become apparent from, coverage extent is smaller.The range of left and right side face angle
It generally can be 90 degree to 90 degree of ﹣, wherein positive and negative respectively indicate turns left or turn right.
In embodiments of the present invention, by above three characteristic value, people in picture to be detected can more accurately be reflected
The position angle and circumstance of occlusion of face target.To pass through three characteristic values and corresponding preset threshold, energy
Enough pictures for more accurately filtering out satisfactory high quality.
Specifically, implementing when characteristic value may include binocular interpupillary distance, pitch angle and left and right side face angle in the present invention
Example provide face picture screening technique in, step 103, the numerical values recited between comparative feature value and preset threshold, according to than
Compared with as a result, picture to be detected is divided into different classes of, may include:
Step 103a, when binocular interpupillary distance be greater than binocular interpupillary distance threshold value, and pitch angle be less than pitch angle threshold value, and left and right
When side face angle is less than left and right side face angle threshold, picture to be detected is high quality picture.
Step 103b, otherwise, picture to be detected are normal quality picture.
Higher-quality picture in order to obtain, in an implementation of the embodiment of the present invention, when above three feature
Value, it is even when meeting corresponding preset threshold, it may be considered that the picture to be detected where the human face target is the picture of high quality.
For example, human face target in picture to be detected by with the three-dimensional references face mould in default three-dimensional references database
Type is compared, and determining that the characteristic value of human face target is respectively as follows: binocular interpupillary distance is 50, pitch angle 20, and left and right angle is
30。
Meanwhile preset threshold is respectively as follows: that binocular interpupillary distance threshold value is 18, pitch angle threshold value is 45 and left and right side face angle threshold
Value 45.
By comparing, the binocular interpupillary distance of human face target is greater than binocular interpupillary distance threshold value in the picture to be detected, and pitch angle is small
In pitch angle threshold value, when left and right side face angle is less than left and right side face angle threshold, so, which is high-quality
Piece.
In another implementation of the embodiment of the present invention, when any two characteristic value in above three characteristic value is full
When the corresponding preset threshold of foot, which can also be divided into the classification of high quality picture.So as to avoid screening
Percent of pass it is too low, more adapt to it is of less demanding to picture quality in the case where, the screening to face picture.
In conjunction with each embodiment above, in practical applications, a variety of different applied fields are adapted in order to more flexible
Scape can also screen picture to be detected when screening to picture to be detected as the classification of multiple class, to meet not
Same business needs.
So in embodiments of the present invention, when being screened to picture to be detected, in addition to high quality picture and common matter
It can also include this classification of mean quality picture other than the two classifications of spirogram piece.Also, the binocular interpupillary distance in preset threshold
Threshold value may include the first binocular interpupillary distance threshold value and the second binocular interpupillary distance threshold value, and pitch angle threshold value may include the first pitch angle
Threshold value and the second pitch angle threshold value are spent, left and right side face angle threshold may include that the first or so side face angle threshold and second are left
Right side face angle threshold.
Wherein, the first binocular interpupillary distance threshold value is greater than the second binocular interpupillary distance threshold value；First pitch angle threshold value is bowed less than second
Face upward angle threshold；First or so side face angle threshold is less than the second or so side face angle threshold.
Correspondingly, in face picture screening technique provided in an embodiment of the present invention, step 103, comparative feature value and pre-
If the numerical values recited between threshold value, according to comparison result, picture to be detected is divided into different classes of, may include:
Step 103c, when binocular interpupillary distance is greater than the first binocular interpupillary distance threshold value, and pitch angle is less than the first pitch angle threshold
Value, and when left and right side face angle is less than the first or so side face angle threshold, picture to be detected is high quality picture.
Step 103d, when binocular interpupillary distance less than the first binocular interpupillary distance threshold value and is greater than the second binocular interpupillary distance threshold value, and pitching
Angle is greater than the first pitch angle threshold value and less than the second pitch angle threshold value, and left and right side face angle is greater than the first or so side face
Angle threshold and when less than the second or so side face angle threshold, picture to be detected is mean quality picture.
Step 103e, otherwise, picture to be detected are normal quality picture.
Since the first binocular interpupillary distance threshold value is greater than the second binocular interpupillary distance threshold value；First pitch angle threshold value is less than the second pitching
Angle threshold；First or so side face angle threshold is less than the second or so side face angle threshold.
So corresponding with the first binocular interpupillary distance threshold value, the first pitch angle when the corresponding each characteristic value of human face target
When measurement standard constructed by threshold value and first or so side face angle thresholds, then it represents that the motor of face target range shooting is set
It is standby to be closer, and the deflections such as position angle of face are smaller, belong to the picture of high quality.
When the corresponding each characteristic value of human face target, do not meet the first binocular interpupillary distance threshold value, the first pitch angle threshold value and
Measurement standard constructed by first or so side face angle threshold, but meet the second binocular interpupillary distance threshold value, the second pitch angle
When measurement standard constructed by threshold value and second or so side face angle thresholds, then it represents that the human face target is although of low quality, but
Can be applied in is not in very high scene to picture quality requirement, so, which can be mean quality
Picture.
If the corresponding each characteristic value of human face target does not meet the second binocular interpupillary distance threshold value, the second pitch angle threshold value
With measurement standard constructed by the second or so side face angle threshold, then it represents that the picture of the picture to be detected where the human face target
Quality is lower, is the picture of normal quality, can not carry out subsequent operation or process to it, can be to avoid inaccurate due to identifying
The problem of false alarm caused by really.
Certainly, in practical applications, the preset threshold of multiple gears can also be set as needed, is treated to realize
Detect the more careful screening of picture so that more accurate to the screening to picture to be detected, with meet different application scene or
The needs of different business.
In conjunction with above-mentioned each embodiment, in practical applications, by the human face target and default three in picture to be detected
When three-dimensional references faceform in dimension benchmark database is compared, in order to further increase the efficiency of comparison, so that comparing
Process it is more accurate.It, step 102, will be in picture to be detected in the face picture screening technique provided in the embodiment of the present invention
Human face target be compared with the three-dimensional references faceform in default three-dimensional references database, determine the spy of human face target
Value indicative can also include:
The first step identifies human face target in picture to be detected.
It, can be first by each before being compared with the three-dimensional references faceform in default three-dimensional references database
The method of class Face datection, for example, neural network algorithm, Principal Component Analysis Algorithm, are based on singular value at independent composition analysis algorithm
Characterization method etc. determines human face target from picture to be detected.Specifically, determining that human face target can be determines people
Image-region where face target, and the human face target is identified in such a way that frame selects.
Alternatively, can also be after determining the region where human face target, by scratching the modes such as figure from picture to be detected
In extract the human face target.
Second step human face target is compared with the three-dimensional references faceform in default three-dimensional references database, really
Make the characteristic value of human face target.
After determining the human face target, so that it may the human face target is inputted in default three-dimensional references database, and it is pre-
If the three-dimensional references faceform in three-dimensional references database is compared, to more fast and easily determine human face target
Characteristic value.
Referring to Fig. 3, Fig. 3 is the structure chart for the face picture screening plant that inventive embodiments provide, comprising:
Module 301 is obtained, for obtaining the picture to be detected comprising human face target；
Comparison module 302, for by the picture to be detected the human face target and the default three-dimensional references number
It is compared according to the three-dimensional references faceform in library, determines the characteristic value of the human face target；
Screening module 303, for the numerical values recited between the characteristic value and preset threshold, according to comparison result,
The picture to be detected is divided into different classes of, the classification includes high quality picture and normal quality picture.
In embodiments of the present invention, by by the human face target in acquired picture to be detected, with default three-dimensional references
Three-dimensional references faceform in database is compared, and obtains the characteristic value of human face target, then again by this feature value and pre-
If the comparison of the carry out numerical values recited between threshold value, so that it is determined that picture to be detected is high quality picture or normal quality figure out
Piece.Due to that in embodiments of the present invention, when being screened to picture to be detected, can be sieved by various characteristic values
Choosing, rather than only only considered one factor of clarity, so as to more accurately measure out whether picture to be detected accords with
The requirement for closing subsequent operation, so that more accurate to the screening of picture to be detected.Also, since preset threshold can carry out in advance
Configuration, it is possible to the screening criteria of more flexible adjustment picture to be detected, to adapt under different actual conditions or scene
Using.
Optionally, in face picture screening plant provided in an embodiment of the present invention, the comparison module 302 is specific to use
In:
By the human face target in the picture to be detected, respectively with it is multiple in the default three-dimensional references database
Three-dimensional references faceform is compared respectively, wherein the multiple three-dimensional references faceform is three with different characteristic value
The quasi- faceform of Wiki；Determine the three-dimensional references faceform to match with the human face target；By the three-dimensional references people
Characteristic value possessed by face model, the characteristic value as the human face target.
Optionally, the characteristic value in face picture screening plant provided in an embodiment of the present invention, in described device
Including binocular interpupillary distance, pitch angle and left and right side face angle；
The preset threshold includes binocular interpupillary distance threshold value, pitch angle threshold value and left and right side face angle threshold.
Optionally, in face picture screening plant provided in an embodiment of the present invention, the screening module 303 is specific to use
In:
When the binocular interpupillary distance be greater than the binocular interpupillary distance threshold value, and the pitch angle be less than the pitch angle threshold
Value, and the left and right side face angle be less than the left and right side face angle threshold when, the picture to be detected be high quality picture；It is no
Then, the picture to be detected is normal quality picture.
Optionally, described in the comparison module 302 in face picture screening plant provided in an embodiment of the present invention
Binocular interpupillary distance threshold value includes the first binocular interpupillary distance threshold value and the second binocular interpupillary distance threshold value, and the pitch angle threshold value is bowed including first
Face upward angle threshold and the second pitch angle threshold value, the left and right side face angle threshold includes the first or so side face angle threshold and the
Two or so side face angle thresholds, the classification in the screening module 303 further includes mean quality picture；
The screening module 303, is specifically used for:
When the binocular interpupillary distance be greater than the first binocular interpupillary distance threshold value, and the pitch angle be less than first pitching
Angle threshold, and the left and right side face angle be less than the first or so side face angle threshold when, the picture to be detected be height
Quality picture；
When the binocular interpupillary distance is less than the first binocular interpupillary distance threshold value and is greater than the second binocular interpupillary distance threshold value, and institute
Pitch angle is stated greater than the first pitch angle threshold value and is less than the second pitch angle threshold value, and left and right side face angle
When degree is greater than the first or so side face angle threshold and is less than the second or so side face angle threshold, the picture to be detected
For mean quality picture；
Otherwise, the picture to be detected is normal quality picture.
Optionally, in face picture screening plant provided in an embodiment of the present invention, the comparison module 302 is specific to use
In:
The human face target is identified in the picture to be detected；
The human face target is compared with the three-dimensional references faceform in the default three-dimensional references database, really
Make the characteristic value of the human face target.
Optionally, in face picture screening plant provided in an embodiment of the present invention, described device further include:
Modeling module, the corresponding faceform of the human face target for establishing in the picture to be detected；
Analysis module, for when the picture to be detected is high quality picture, the faceform to be inputted present count
According to library, it is compared with each default faceform stored in the presetting database；
Alarm module, for working as the default faceform for existing in the presetting database and matching with the faceform
When, export warning message；Alternatively, when there is no the default face moulds to match with the faceform in the presetting database
When type, warning message is exported.
The embodiment of the invention also provides a kind of electronic equipment, as shown in figure 4, include processor 401, communication interface 402,
Memory 403 and communication bus 404, wherein processor 401, communication interface 402, memory 403 are complete by communication bus 404
At mutual communication,
Memory 403, for storing computer program；
Processor 401 when for executing the program stored on memory 403, realizes following steps:
Obtain the picture to be detected comprising human face target；
By the human face target in the picture to be detected and the three-dimensional references in the default three-dimensional references database
Faceform is compared, and determines the characteristic value of the human face target；
Compare the numerical values recited between the characteristic value and preset threshold, according to comparison result, by the picture to be detected
It is divided into different classes of, the classification includes high quality picture and normal quality picture.
The communication bus that above-mentioned electronic equipment is mentioned can be PCI (Peripheral Component
Interconnect, Peripheral Component Interconnect standard) bus or EISA (Extended Industry Standard
Architecture, expanding the industrial standard structure) bus etc..The communication bus can be divided into address bus, data/address bus, control
Bus etc..Only to be indicated with a thick line in figure convenient for indicating, it is not intended that an only bus or a type of total
Line.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include RAM (Random Access Memory, random access memory), also may include NVM
(Non-Volatile Memory, nonvolatile memory), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including CPU (Central Processing Unit, central processing
Device), NP (Network Processor, network processing unit) etc.；Can also be DSP (Digital Signal Processor,
Digital signal processor), ASIC (Application Specific Integrated Circuit, specific integrated circuit),
FPGA (Field-Programmable Gate Array, field programmable gate array) or other programmable logic device are divided
Vertical door or transistor logic, discrete hardware components.
In another embodiment provided by the invention, a kind of computer readable storage medium is additionally provided, which can
It reads to be stored with instruction in storage medium, when run on a computer, so that computer executes any institute in above-described embodiment
The face picture screening technique stated.
In another embodiment provided by the invention, a kind of computer program product comprising instruction is additionally provided, when it
When running on computers, so that computer executes any face picture screening technique in above-described embodiment.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program
Product includes one or more computer instructions.When loading on computers and executing the computer program instructions, all or
It partly generates according to process or function described in the embodiment of the present invention.The computer can be general purpose computer, dedicated meter
Calculation machine, computer network or other programmable devices.The computer instruction can store in computer readable storage medium
In, or from a computer readable storage medium to the transmission of another computer readable storage medium, for example, the computer
Instruction can pass through wired (such as coaxial cable, optical fiber, number from a web-site, computer, server or data center
User's line (DSL)) or wireless (such as infrared, wireless, microwave etc.) mode to another web-site, computer, server or
Data center is transmitted.The computer readable storage medium can be any usable medium that computer can access or
It is comprising data storage devices such as one or more usable mediums integrated server, data centers.The usable medium can be with
It is magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk
Solid State Disk (SSD)) etc..
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is all made of relevant mode and describes, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device etc.
For embodiment, since it is substantially similar to the method embodiment, so being described relatively simple, related place is implemented referring to method
The part explanation of example.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (16)
1. a kind of face picture screening technique characterized by comprising
Obtain the picture to be detected comprising human face target；
By the human face target in the picture to be detected and the three-dimensional references face in the default three-dimensional references database
Model is compared, and determines the characteristic value of the human face target；
Compare the numerical values recited between the characteristic value and preset threshold, according to comparison result, the picture to be detected is divided into
Different classes of, the classification includes high quality picture and normal quality picture.
2. the method according to claim 1, wherein the human face target by the picture to be detected
It is compared with the three-dimensional references faceform in the default three-dimensional references database, determines the feature of the human face target
Value, comprising:
By the human face target in the picture to be detected, respectively with multiple three-dimensionals in the default three-dimensional references database
Benchmark face model is compared respectively, wherein the multiple three-dimensional references faceform is three Wikis with different characteristic value
Quasi- faceform；
Determine the three-dimensional references faceform to match with the human face target；
Characteristic value by characteristic value possessed by the three-dimensional references faceform, as the human face target.
3. the method according to claim 1, wherein the characteristic value includes binocular interpupillary distance, pitch angle and a left side
Right side face angle；
The preset threshold includes binocular interpupillary distance threshold value, pitch angle threshold value and left and right side face angle threshold.
4. according to the method described in claim 3, it is characterized in that, the number between characteristic value and preset threshold
The picture to be detected is divided into different classes of by value size according to comparison result, comprising:
When the binocular interpupillary distance be greater than the binocular interpupillary distance threshold value, and the pitch angle be less than the pitch angle threshold value, and
When the left and right side face angle is less than the left and right side face angle threshold, the picture to be detected is high quality picture；
Otherwise, the picture to be detected is normal quality picture.
5. according to the method described in claim 3, it is characterized in that, the binocular interpupillary distance threshold value includes the first binocular interpupillary distance threshold value
With the second binocular interpupillary distance threshold value, the pitch angle threshold value includes the first pitch angle threshold value and the second pitch angle threshold value, institute
Stating left and right side face angle threshold includes that the first or so side face angle threshold and second or so side face angle thresholds, the classification are also wrapped
Include mean quality picture；
The numerical values recited between characteristic value and preset threshold, according to comparison result, by the picture to be detected
It is divided into different classes of, comprising:
When the binocular interpupillary distance be greater than the first binocular interpupillary distance threshold value, and the pitch angle be less than first pitch angle
Threshold value, and the left and right side face angle be less than the first or so side face angle threshold when, the picture to be detected be high quality
Picture；
When the binocular interpupillary distance be less than the first binocular interpupillary distance threshold value and be greater than the second binocular interpupillary distance threshold value,
And the pitch angle is greater than the first pitch angle threshold value and is less than the second pitch angle threshold value,
And the left and right side face angle is greater than the first or so side face angle threshold and is less than the second or so side face angle
When threshold value, the picture to be detected is mean quality picture；
Otherwise, the picture to be detected is normal quality picture.
6. the method according to claim 1, wherein the human face target by the picture to be detected
It is compared with the three-dimensional references faceform in the default three-dimensional references database, determines the feature of the human face target
Value, comprising:
The human face target is identified in the picture to be detected；
The human face target is compared with the three-dimensional references faceform in the default three-dimensional references database, is determined
The characteristic value of the human face target.
7. method according to any one of claim 1 to 6, which is characterized in that the method also includes:
Establish the corresponding faceform of the human face target in the picture to be detected；
When the picture to be detected is high quality picture, the faceform is inputted into presetting database, with the present count
It is compared according to each default faceform stored in library；
When there is the default faceform to match with the faceform in the presetting database, warning message is exported；
Alternatively, when, there is no when the default faceform to match with the faceform, output is reported in the presetting database
Alert information.
8. a kind of face picture screening plant characterized by comprising
Module is obtained, for obtaining the picture to be detected comprising human face target；
Comparison module, for by the picture to be detected the human face target and the default three-dimensional references database in
Three-dimensional references faceform is compared, and determines the characteristic value of the human face target；
Screening module, for the numerical values recited between the characteristic value and preset threshold, according to comparison result, will it is described to
Detection picture is divided into different classes of, and the classification includes high quality picture and normal quality picture.
9. device according to claim 8, which is characterized in that the comparison module is specifically used for:
By the human face target in the picture to be detected, respectively with multiple three-dimensionals in the default three-dimensional references database
Benchmark face model is compared respectively, wherein the multiple three-dimensional references faceform is three Wikis with different characteristic value
Quasi- faceform；Determine the three-dimensional references faceform to match with the human face target；By the three-dimensional references face mould
Characteristic value possessed by type, the characteristic value as the human face target.
10. device according to claim 8, which is characterized in that the characteristic value in described device include binocular interpupillary distance,
Pitch angle and left and right side face angle；
The preset threshold includes binocular interpupillary distance threshold value, pitch angle threshold value and left and right side face angle threshold.
11. device according to claim 10, which is characterized in that the screening module is specifically used for:
When the binocular interpupillary distance be greater than the binocular interpupillary distance threshold value, and the pitch angle be less than the pitch angle threshold value, and
When the left and right side face angle is less than the left and right side face angle threshold, the picture to be detected is high quality picture；Otherwise, institute
Stating picture to be detected is normal quality picture.
12. device according to claim 10, which is characterized in that the binocular interpupillary distance threshold value packet in the comparison module
Include the first binocular interpupillary distance threshold value and the second binocular interpupillary distance threshold value, the pitch angle threshold value includes the first pitch angle threshold value and the
Two pitch angle threshold values, the left and right side face angle threshold include the first or so side face angle threshold and second or so side face angles
Threshold value, the classification in the screening module further includes mean quality picture；
The screening module, is specifically used for:
When the binocular interpupillary distance be greater than the first binocular interpupillary distance threshold value, and the pitch angle be less than first pitch angle
Threshold value, and the left and right side face angle be less than the first or so side face angle threshold when, the picture to be detected be high quality
Picture；
When the binocular interpupillary distance is less than the first binocular interpupillary distance threshold value and is greater than the second binocular interpupillary distance threshold value, and described bow
Elevation angle degree is greater than the first pitch angle threshold value and is less than the second pitch angle threshold value, and the left and right side face angle is big
In the first or so side face angle threshold and when being less than the second or so side face angle threshold, during the picture to be detected is
Etc. quality picture；
Otherwise, the picture to be detected is normal quality picture.
13. device according to claim 8, which is characterized in that the comparison module is specifically used for:
The human face target is identified in the picture to be detected；
The human face target is compared with the three-dimensional references faceform in the default three-dimensional references database, is determined
The characteristic value of the human face target.
14. the device according to any one of claim 8 to 13, which is characterized in that described device further include:
Modeling module, the corresponding faceform of the human face target for establishing in the picture to be detected；
Analysis module, for when the picture to be detected is high quality picture, the faceform to be inputted presetting database,
It is compared with each default faceform stored in the presetting database；
Alarm module, for when in the presetting database exist match with the faceform default faceform when,
Export warning message；Alternatively, when there is no the default faceforms to match with the faceform in the presetting database
When, export warning message.
15. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein processing
Device, communication interface, memory complete mutual communication by communication bus；
Memory, for storing computer program；
Processor when for executing the program stored on memory, realizes method and step as claimed in claim 1 to 7.
16. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium
Program, the computer program realize method and step as claimed in claim 1 to 7 when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810360907.7A CN110390229B (en) | 2018-04-20 | 2018-04-20 | Face picture screening method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810360907.7A CN110390229B (en) | 2018-04-20 | 2018-04-20 | Face picture screening method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110390229A true CN110390229A (en) | 2019-10-29 |
| CN110390229B CN110390229B (en) | 2022-03-04 |
Family
ID=68283691
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810360907.7A Active CN110390229B (en) | 2018-04-20 | 2018-04-20 | Face picture screening method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110390229B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111046806A (en) * | 2019-12-12 | 2020-04-21 | 天地伟业技术有限公司 | Heterogeneous image face recognition target library generation method |
| CN111159445A (en) * | 2019-12-30 | 2020-05-15 | 深圳云天励飞技术有限公司 | Picture filtering method and device, electronic equipment and storage medium |
| CN111178162A (en) * | 2019-12-12 | 2020-05-19 | 北京迈格威科技有限公司 | Image recognition method and device, computer equipment and storage medium |
| CN111640207A (en) * | 2020-06-04 | 2020-09-08 | 福建票付通信息科技有限公司 | Non-inductive face collection ticket checking method |
| CN112220450A (en) * | 2020-08-21 | 2021-01-15 | 上海交通大学医学院附属第九人民医院 | Orbit disease screening method, system and terminal based on three-dimensional model |
| CN112580434A (en) * | 2020-11-25 | 2021-03-30 | 奥比中光科技集团股份有限公司 | Face false detection optimization method and system based on depth camera and face detection equipment |
| CN113095110A (en) * | 2019-12-23 | 2021-07-09 | 浙江宇视科技有限公司 | Method, device, medium and electronic equipment for dynamically warehousing face data |
| EP4124029A4 (en) * | 2020-03-17 | 2023-04-05 | NEC Corporation | Gate system, gate device, image processing method therefor, program, and arrangement method for gate device |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799877A (en) * | 2012-09-11 | 2012-11-28 | 上海中原电子技术工程有限公司 | Method and system for screening face images |
| CN103745208A (en) * | 2014-01-27 | 2014-04-23 | 中国科学院深圳先进技术研究院 | Face recognition method and device |
| US20140152956A1 (en) * | 2012-07-03 | 2014-06-05 | Reverse Engineering, Lda | System for the measurement of the interpupillary distance using a device equipped with a display and a camera |
| CN103927515A (en) * | 2014-04-08 | 2014-07-16 | 章建国 | Three-dimension dynamic facial comparison method |
| CN104463117A (en) * | 2014-12-02 | 2015-03-25 | 苏州科达科技股份有限公司 | Sample collection method and system used for face recognition and based on video |
| CN104657817A (en) * | 2015-01-28 | 2015-05-27 | 四川君逸易视科技有限公司 | Face snapshotting, comparing, identifying, retrieving, and inquiring method for bank counter |
| CN105243357A (en) * | 2015-09-15 | 2016-01-13 | 深圳市环阳通信息技术有限公司 | Identity document-based face recognition method and face recognition device |
| CN105631439A (en) * | 2016-02-18 | 2016-06-01 | 北京旷视科技有限公司 | Human face image collection method and device |
| CN106557723A (en) * | 2015-09-25 | 2017-04-05 | 北京市商汤科技开发有限公司 | A kind of system for face identity authentication with interactive In vivo detection and its method |
| CN107122054A (en) * | 2017-04-27 | 2017-09-01 | 青岛海信医疗设备股份有限公司 | A kind of detection method and device of face deflection angle and luffing angle |
| CN107277265A (en) * | 2017-07-13 | 2017-10-20 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
-
2018
- 2018-04-20 CN CN201810360907.7A patent/CN110390229B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140152956A1 (en) * | 2012-07-03 | 2014-06-05 | Reverse Engineering, Lda | System for the measurement of the interpupillary distance using a device equipped with a display and a camera |
| CN102799877A (en) * | 2012-09-11 | 2012-11-28 | 上海中原电子技术工程有限公司 | Method and system for screening face images |
| CN103745208A (en) * | 2014-01-27 | 2014-04-23 | 中国科学院深圳先进技术研究院 | Face recognition method and device |
| CN103927515A (en) * | 2014-04-08 | 2014-07-16 | 章建国 | Three-dimension dynamic facial comparison method |
| CN104463117A (en) * | 2014-12-02 | 2015-03-25 | 苏州科达科技股份有限公司 | Sample collection method and system used for face recognition and based on video |
| CN104657817A (en) * | 2015-01-28 | 2015-05-27 | 四川君逸易视科技有限公司 | Face snapshotting, comparing, identifying, retrieving, and inquiring method for bank counter |
| CN105243357A (en) * | 2015-09-15 | 2016-01-13 | 深圳市环阳通信息技术有限公司 | Identity document-based face recognition method and face recognition device |
| CN106557723A (en) * | 2015-09-25 | 2017-04-05 | 北京市商汤科技开发有限公司 | A kind of system for face identity authentication with interactive In vivo detection and its method |
| CN105631439A (en) * | 2016-02-18 | 2016-06-01 | 北京旷视科技有限公司 | Human face image collection method and device |
| CN107122054A (en) * | 2017-04-27 | 2017-09-01 | 青岛海信医疗设备股份有限公司 | A kind of detection method and device of face deflection angle and luffing angle |
| CN107277265A (en) * | 2017-07-13 | 2017-10-20 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
Non-Patent Citations (2)
| Title |
|---|
| MASAYUKI ASANO等: "Eye detection method robust to facial pose changes for eye input device", 《2011 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS》 * |
| 张国亮 等: "基于视觉表情分析的交互式表情机器人系统研究", 《小型微型计算机系统》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111046806A (en) * | 2019-12-12 | 2020-04-21 | 天地伟业技术有限公司 | Heterogeneous image face recognition target library generation method |
| CN111178162A (en) * | 2019-12-12 | 2020-05-19 | 北京迈格威科技有限公司 | Image recognition method and device, computer equipment and storage medium |
| CN111178162B (en) * | 2019-12-12 | 2023-11-07 | 北京迈格威科技有限公司 | Image recognition method, device, computer equipment and storage medium |
| CN113095110A (en) * | 2019-12-23 | 2021-07-09 | 浙江宇视科技有限公司 | Method, device, medium and electronic equipment for dynamically warehousing face data |
| CN113095110B (en) * | 2019-12-23 | 2024-03-08 | 浙江宇视科技有限公司 | Method, device, medium and electronic equipment for dynamically warehousing face data |
| CN111159445A (en) * | 2019-12-30 | 2020-05-15 | 深圳云天励飞技术有限公司 | Picture filtering method and device, electronic equipment and storage medium |
| EP4124029A4 (en) * | 2020-03-17 | 2023-04-05 | NEC Corporation | Gate system, gate device, image processing method therefor, program, and arrangement method for gate device |
| CN111640207A (en) * | 2020-06-04 | 2020-09-08 | 福建票付通信息科技有限公司 | Non-inductive face collection ticket checking method |
| CN112220450A (en) * | 2020-08-21 | 2021-01-15 | 上海交通大学医学院附属第九人民医院 | Orbit disease screening method, system and terminal based on three-dimensional model |
| CN112220450B (en) * | 2020-08-21 | 2023-08-15 | 上海交通大学医学院附属第九人民医院 | Orbital disease screening method, system and terminal based on three-dimensional model |
| CN112580434A (en) * | 2020-11-25 | 2021-03-30 | 奥比中光科技集团股份有限公司 | Face false detection optimization method and system based on depth camera and face detection equipment |
| CN112580434B (en) * | 2020-11-25 | 2024-03-15 | 奥比中光科技集团股份有限公司 | Face false detection optimization method and system based on depth camera and face detection equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110390229B (en) | 2022-03-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110390229A (en) | A kind of face picture screening technique, device, electronic equipment and storage medium | |
| CN110866891B (en) | Image recognition device, image recognition method, and storage medium | |
| CN108090458A (en) | Tumble detection method for human body and device | |
| CN106650662B (en) | Target object shielding detection method and device | |
| CN105631439B (en) | Face image processing process and device | |
| CN108351967A (en) | A kind of plurality of human faces detection method, device, server, system and storage medium | |
| EP2919153A1 (en) | Event detection apparatus and event detection method | |
| CN109165589A (en) | Vehicle based on deep learning recognition methods and device again | |
| CN107240078A (en) | Lens articulation Method for Checking, device and electronic equipment | |
| JP4714749B2 (en) | Real-time image detection using polarization data | |
| CN109086734B (en) | Method and device for positioning pupil image in human eye image | |
| CN109766779A (en) | It hovers personal identification method and Related product | |
| CN107111704A (en) | System and method based on biological recognition system fraud detection in iris | |
| CN112215037B (en) | Object tracking method and device, electronic equipment and computer readable storage medium | |
| CN110879995A (en) | Target object detection method and device, storage medium and electronic device | |
| CN106355367A (en) | Warehouse monitoring management device | |
| CN109167998A (en) | Detect method and device, the electronic equipment, storage medium of camera status | |
| CN109724993A (en) | Detection method, device and the storage medium of the degree of image recognition apparatus | |
| CN108875519A (en) | Method for checking object, device and system and storage medium | |
| CN109711346A (en) | A kind of birds abnormality recognition method and device | |
| CN112633255A (en) | Target detection method, device and equipment | |
| CN111751279A (en) | Optical image capturing parameter adjusting method and sensing device | |
| CN110287361A (en) | A kind of personage's picture screening technique and device | |
| CN109727417A (en) | Video processing unit is controlled to promote the method and controller of detection newcomer | |
| CN109583396A (en) | A kind of region prevention method, system and terminal based on CNN two stages human testing |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |