CN105072327B - A kind of method and apparatus of the portrait processing of anti-eye closing - Google Patents
A kind of method and apparatus of the portrait processing of anti-eye closing Download PDFInfo
- Publication number
- CN105072327B CN105072327B CN201510416911.7A CN201510416911A CN105072327B CN 105072327 B CN105072327 B CN 105072327B CN 201510416911 A CN201510416911 A CN 201510416911A CN 105072327 B CN105072327 B CN 105072327B
- Authority
- CN
- China
- Prior art keywords
- image
- eye
- eyes
- portrait
- 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.)
- Expired - Fee Related
Links
- 238000012545 processing Methods 0.000 title claims abstract description 79
- 238000000034 method Methods 0.000 title claims abstract description 35
- 210000001508 eye Anatomy 0.000 claims abstract description 196
- 238000001514 detection method Methods 0.000 claims abstract description 35
- 238000012546 transfer Methods 0.000 claims abstract description 11
- 210000001747 pupil Anatomy 0.000 claims description 53
- 230000006835 compression Effects 0.000 claims description 42
- 238000007906 compression Methods 0.000 claims description 42
- 230000001815 facial effect Effects 0.000 claims description 13
- 210000005252 bulbus oculi Anatomy 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 230000012010 growth Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims 2
- 238000010586 diagram Methods 0.000 description 6
- 238000012790 confirmation Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004087 cornea Anatomy 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 210000000887 face Anatomy 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 210000000554 iris Anatomy 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000002087 whitening effect Effects 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The present invention relates to the method and apparatus that a kind of portrait of anti-eye closing is handled, and including shooting unit, detection unit, transfer unit and processing unit；The shooting unit, for obtaining image to be confirmed；The detection unit, for obtaining first image in the image to be confirmed and detecting the state of the wherein eyes of owner's picture；It is described to transfer unit, if being in closed-eye state for the eyes of at least one portrait, first image is preserved, and the subsequent image after transferring；The processing unit, the eye for being in closed-eye state to first image according to the subsequent image do Local treatment, so that the eyes of owner's picture are in eyes-open state in first image.As it can be seen that the method and apparatus of the portrait processing of the anti-eye closing, can effectively avoid the eyes for taking portrait from being in the photo of closed-eye state, greatly improve the quality of photo in itself, optimize user experience.
Description
Technical field
The method and dress handled the present invention relates to technical field of image processing more particularly to a kind of portrait of anti-eye closing
It puts.
Background technology
With mobile terminal, such as the continuous development of the shooting technology of mobile phone, more and more users play self-timer using mobile phone
Or group claps, various U.S.'s face camera functions also occur therewith.U.S. face camera function by captured image to be confirmed into pedestrian
Face identifies, whitening, convergence, the mill post-processing operations such as skin are then carried out to face, realizes that the image that confirmation is allowed to take seems face
Skin is more smooth, the visual effect of Bai Liang.U.S. face camera function almost becomes the essential function that female user is taken pictures now.So
And when more people together group clap, particularly travel outdoors souvenir when, it is easy to just take some human eye and be in closed-eye state
Photo, the quality of this photo not only big heavy discount can also more or less influence the mood of playing of people.
The content of the invention
It is an object of the invention to propose a kind of method and apparatus of the portrait of anti-eye closing processing, can effectively avoid
The eyes for taking portrait are in the photo of closed-eye state, greatly improve the quality of photo in itself, optimize user experience.
For this purpose, the present invention uses following technical scheme：
In a first aspect, a kind of method of the portrait processing of anti-eye closing is provided, including：
It obtains and confirms shooting instruction, image to be confirmed, the image to be confirmed are obtained by the interface of finding a view for shooting portrait
Several including continuously acquiring do not do the image that digital image compression is handled；
First image in the image to be confirmed is obtained, is detected using Face datection algorithm in first image
The state of the eyes of owner's picture；
If the eyes of at least one portrait are in closed-eye state, first image is preserved, and transfers described
Subsequent image after one image；
Whether the eyes for detecting the portrait for being in closed-eye state corresponding to first image in the subsequent image open
It opens；If opening, first image is corresponded in closed-eye state using corresponding this subsequent image of eyes-open state
Eye does Local treatment, so that the eyes of owner's picture are in eyes-open state in first image.
Second aspect provides a kind of device of the portrait processing of anti-eye closing, including：
Shooting unit confirms shooting instruction for obtaining, and image to be confirmed, institute are obtained by the interface of finding a view for shooting portrait
State image to be confirmed include continuously acquire several do not do digital image compression processing image；
For obtaining first image in the image to be confirmed, institute is detected using Face datection algorithm for detection unit
State the state of the eyes of owner's picture in first image；
Unit is transferred, if being in closed-eye state for the eyes of at least one portrait, preserves first image,
And transfer the subsequent image after first image；
Processing unit, for detecting the portrait for corresponding to first image in the subsequent image and being in closed-eye state
Eyes whether open；If opening, using corresponding this subsequent image of eyes-open state to first image corresponding position
Local treatment is done in the eye of closed-eye state, so that the eyes of owner's picture are in eyes-open state in first image.
The beneficial effects of the present invention are：A kind of method and apparatus of the portrait processing of anti-eye closing, it is single including shooting
Member, detection unit transfer unit and processing unit；The shooting unit confirms shooting instruction, by shooting portrait for obtaining
Interface of finding a view obtain image to be confirmed, the image to be confirmed include continuously acquiring several do not do at digital image compression
The image of reason；The detection unit for obtaining first image in the image to be confirmed, is examined using Face datection algorithm
Survey the state of the eyes of owner's picture in first image；It is described to transfer unit, if the eye at least one portrait
Eyeball is in closed-eye state, then preserves first image, and transfers the subsequent image after first image；The place
Manage unit, for detect in the subsequent image correspond to first image be in closed-eye state portrait eyes whether
It opens；If opening, first image is corresponded in closed-eye state using corresponding this subsequent image of eyes-open state
Eye do Local treatment so that the eyes of owner's picture are in eyes-open state in first image.As it can be seen that the anti-eye closing
Portrait processing method and apparatus, the eyes for taking portrait can effectively be avoided to be in the photo of closed-eye state, significantly
The quality of photo in itself is improved, optimizes user experience.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, institute in being described below to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without creative efforts, can also implement according to the present invention
The content of example and these attached drawings obtain other attached drawings.
Fig. 1 is the method flow diagram of method one embodiment of the portrait processing of anti-eye closing provided by the invention.
Fig. 2 is the method flow diagram of second embodiment of method of the portrait processing of anti-eye closing provided by the invention.
Fig. 3 is the block diagram of device one embodiment of the portrait processing of anti-eye closing provided by the invention.
Fig. 4 is the block diagram of second embodiment of device of the portrait processing of anti-eye closing provided by the invention.
Specific embodiment
For make present invention solves the technical problem that, the technical solution that uses and the technique effect that reaches it is clearer, below
The technical solution of the embodiment of the present invention will be described in further detail with reference to attached drawing, it is clear that described embodiment is only
It is part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those skilled in the art exist
All other embodiments obtained under the premise of creative work are not made, belong to the scope of protection of the invention.
It please refers to Fig.1, is the side of method one embodiment of the portrait processing of anti-eye closing provided by the invention
Method flow chart.The method of the portrait processing of the anti-eye closing of the embodiment of the present invention can be applied to various possess camera function
Intelligent terminal, such as mobile phone, tablet, smartwatch.
The method of the portrait processing of the anti-eye closing, including：
Step S101, obtain and confirm shooting instruction, image to be confirmed is obtained by the interface of finding a view for shooting portrait, it is described to treat
Confirm image include continuously acquire several do not do digital image compression processing image.
The interface of finding a view of the shooting portrait is the form or display screen for the intelligent terminal for possessing camera function.
Image can be such that picture signal is transmitted in high quality after digital image compression is handled, and convenient for image
Retrieval, analysis, processing and storage.
Step S102, first image in the image to be confirmed is obtained, utilizes Face datection algorithm detection described the
The state of the eyes of owner's picture in one image.
Confirm shooting instruction when obtaining, that is, represent that user is satisfied in the main to the image to be confirmed at interface of finding a view at present, pass through
Confirmation shooting key is pressed, processing output is carried out to image to be confirmed, forms final portrait picture.
The process of the state of the eyes of owner's picture in first image is detected using Face datection algorithm, it is main to divide
For the detection of people face part, the ocular detection based on face sorting survey and three parts of eyes detection.
Face sorting surveys the detection of the characteristic point on Main Basiss face, and carrying out characteristic point to facial image detects it
Before, it is also necessary to carry out feature point extraction and classification.During detecting face, first by facial image and face template to be determined
Template matches are carried out, if it does, being so projected into face subspace, face is determined whether by feature sub-face technology.
Feature sub-face technology finds the basic element of facial image distribution, i.e. facial image sample set covariance square from the angle of statistics
The feature vector of battle array, facial image is approx characterized with this.These feature vectors are known as eigenface (Eigenface).In fact,
Eigenface reflects the structural relation for lying in information and face inside face sample set.By eyes, cheek, lower jaw sample
The feature vector of this collection covariance matrix is known as eigen eyes, feature jaw and feature lip, is referred to as sub-face of feature.Sub-face of feature is corresponding
Image space in be turned into subspace, be known as sub-face space.Projector distance of the test image window in sub-face space is calculated, if
Video in window meets threshold value comparison condition, then judges it for face.The corresponding region behaviour face part of the face.
It is that the above-mentioned face sorting of ocular detection Main Basiss is surveyed as a result, figure according to the definite face part
Picture utilizes the initial position of the three front yards ratio estimate nose of five.Define two initial ranging rectangles, difference two to left and right
Residing approximate location growth.The characteristics of being significantly lower than face gray scale according to human eye gray scale, eye is found using search rectangular
Edge, and the center of pupil in eyeball is positioned, the ocular of final definite face part.
Eyes detection Main Basiss eye feature value, such as the inspection of pupil middle height, interpupillary distance or pupil gray value
It surveys as a result, being in eyes-open state to eyes or closed-eye state judges.
If step S103, the eyes of at least one portrait are in closed-eye state, first image is preserved, and is adjusted
Take the subsequent image after first image.
If it was found that the eyes of at least one portrait are in closed-eye state, illustrate that picture quality to be confirmed at this time is relatively low,
The requirement of user is not achieved, it is necessary to carry out the portrait processing of anti-eye closing to image to be confirmed.
Step S104, detect and correspond to the eye that first image is in the portrait of closed-eye state in the subsequent image
Whether eyeball is opened；If opening, first image is corresponded to be in using corresponding this subsequent image of eyes-open state and is closed
The eye of eye shape state does Local treatment, so that the eyes of owner's picture are in eyes-open state in first image.
Preferably, eye Local treatment is described in the eye being in using eye in the subsequent image of eyes-open state is replaced
The eye that portrait is in closed-eye state is corresponded in first image, on the premise of face characteristic coincide substantially, eyes-open state
Eye be directly covered in the eye of closed-eye state.
The method of the portrait processing provided in an embodiment of the present invention for moving anti-eye closing, can effectively avoid taking portrait
Eyes are in the photo of closed-eye state, greatly improve the quality of photo in itself, optimize user experience.
It please refers to Fig.2, is the side of second embodiment of method of the portrait processing of anti-eye closing provided by the invention
Method flow chart.On the basis of the one embodiment for the method that the embodiment of the present invention is handled in the portrait of anti-eye closing, to eye
It does Local treatment and is illustrated the step of judging the state of portrait eye in portion.
The method of the portrait processing of the anti-eye closing, including：
Step S201, camera function is started.
Start camera function, that is, terminal is allowed to enter state of taking pictures, the interface of finding a view for shooting portrait starts to find a view.
Step S202, obtain and confirm shooting instruction, image to be confirmed is obtained by the interface of finding a view for shooting portrait, it is described to treat
Confirm image include continuously acquire several do not do digital image compression processing image.
Step S203, first image in the image to be confirmed is obtained, utilizes Face datection algorithm detection described the
The state of the eyes of owner's picture in one image.
Preferably, the state that the eyes of owner's picture in first image are detected using Face datection algorithm,
Including：
The eye feature value of owner's picture in first image, the eye feature are detected using Face datection algorithm
Pupil middle height, interpupillary distance or the pupil gray value that value includes；
When any portrait pupil middle height be less than or equal to preset height, then judge that the eyes of the portrait are in eye closing shape
State；When any portrait pupil middle height be more than the preset height, then judge that the eyes of the portrait are in eyes-open state,
Or；
When the interpupillary distance for not detecting any portrait, then judge that the eyes of the portrait are in closed-eye state；When detection is taken office
The interpupillary distance of one portrait, then judge the portrait eyes be in eyes-open state or；
When the pupil gray value of any portrait is less than or equal to default pupil gray value, then judges that the eyes of the portrait are in and close
Eye shape state；When any portrait pupil gray value be more than the default pupil gray value, then judge the portrait eyes be in open
Eye shape state.
Eyeball refers to center section before eyeball, this position is made of cornea, iris and pupil.It is high in the middle part of pupil
Spend the centerline height for circular pupil, the color depth that interpupillary distance is the distance between two pupils, pupil gray value is pupil.
When preset height is in normal eyes-open state for eyes pre-set, corresponding to the face size in image
Pupil middle height, it is that eyes pre-set, corresponding to the face size in image are in normal to preset pupil gray value
Pupil gray value during eyes-open state.Model that can be according to different terminals in actual application and configuration, to preset height
It is configured with default pupil gray value.
If step S204a, the eyes of at least one portrait are in closed-eye state, first image is preserved, and
Transfer the subsequent image after first image.
If step S204b, the eyes of nobody's picture are in closed-eye state, digital picture is carried out to first image
Compression handle, through the digital image compression treated finally portrait picture be shown in screen.
It, can be directly to image to be captured into brief biography of a deceased person digital picture pressure if the eyes of nobody's picture are in closed-eye state
Contracting is handled, to export.
It should be noted that step S204a and step S204b, there is no sequencing relation, alternatively implements.
Step S205, in preset period of time, detect in the subsequent image and be in corresponding to first image
Whether the eyes of the portrait of closed-eye state are opened.
Preferably, the preset period of time is 2s.In 2s, the eye of portrait is not opened still, is continuously in and closes
If eye shape state, then may determine that this be in order to build more lively shooting effect deliberately for it.It can directly treat at this time
The image of shooting is into the processing of brief biography of a deceased person digital image compression, to export.
If step S206a, opening, first image is corresponded to using corresponding this subsequent image of eyes-open state
Eye in closed-eye state does Local treatment, so that the eyes of owner's picture are in eyes-open state in first image；
The image that the eyes for making the owner's picture obtained after the Local treatment are in eyes-open state carries out digital image compression processing,
Through the digital image compression, treated that final portrait picture is shown in screen.
If step S206b, not opening, digital image compression processing is carried out to first image, described in warp
Treated that final portrait picture is shown in screen for digital image compression.
Step S206a and step S206b allow terminal anti-eye closing portrait processing process is more intelligent and people
Property improves the satisfaction of user.It should be noted that step S206a and step S206b is there is no sequencing relation,
Alternatively implements.
Preferably, the digital image compression processing is to be handled using the digital image compression of JPEG encryption algorithms.
JPEG encryption algorithms are the canonical algorithm of Static Picture Compression.JPEG encryption algorithms must meet the following requirement：
1st, JPEG encryption algorithms are independently of the resolution ratio of image.
2nd, there is the encoding rate less than 1bit/ pixels, and image can be established in five seconds, in real time with meeting
It asks.
3rd, original image can be recovered without distortions in the case where compression ratio is about 2.
4th, the encoding and decoding of support order and progressive encoding and decoding.
5th, to the adaptive ability of various iconic elements and data precision.
6th, coding/decoding apparatus is simply easily realized.
The method of the portrait processing of anti-eye closing provided in an embodiment of the present invention, it is real first by Face datection algorithm
When detection shooting portrait interface of finding a view in all faces position and corresponding eye state.Be pressed when confirmation shoots key,
It sends after confirming shooting instruction, before image to be confirmed does JPEG codings, takes first image, and judge first image
First image if there is human eye is in closed-eye state, is then first preserved, not done by the state of middle portrait eyes
JPEG is encoded.Every image gives more sustained attention the eye state of eye closing people below.Until the eyes of eye closing people are in eyes-open state,
According to the eyes of the people and the feature of face, to the eye closing people of first image, corresponding eye Local treatment is carried out, makes its eye
Eyeball is in eyes-open state.If eye closing people is continuously in closed-eye state more than 2s, without eye Local treatment, acquiescence is special
Expect the effect for taking eye closing rather than because the photo of eye closing is taken caused by blink.Finally by the by eye processing
One image carries out JPEG codings, exports final portrait picture.
The method of the portrait processing of anti-eye closing provided in an embodiment of the present invention, can be obtained by Face datection algorithm
To the state of human eye, eye Local treatment then is carried out to the portrait of eye closing, the eyes-open state of portrait is synthesized to first figure
As upper, eye closing portrait is made to be in eyes-open state, to take higher-quality portrait photo, effectively avoids taking eye closing portrait
Photo.
The embodiment of the device handled below for the portrait of anti-eye closing provided in an embodiment of the present invention.The people of anti-eye closing
As the embodiment of the embodiment and the method for the portrait processing of above-mentioned anti-eye closing of the device of shooting processing belongs to same structure
Think, the detail content of not detailed description in the embodiment of the device of the portrait of anti-eye closing processing, may be referred to above-mentioned anti-close
The embodiment of the method for the portrait processing of eye.
It please refers to Fig.3, is the knot of device one embodiment of the portrait processing of anti-eye closing provided by the invention
Structure block diagram.
The device of the portrait processing of the anti-eye closing, including：
Shooting unit confirms shooting instruction for obtaining, and image to be confirmed, institute are obtained by the interface of finding a view for shooting portrait
State image to be confirmed include continuously acquire several do not do digital image compression processing image；
For obtaining first image in the image to be confirmed, institute is detected using Face datection algorithm for detection unit
State the state of the eyes of owner's picture in first image；
Unit is transferred, if being in closed-eye state for the eyes of at least one portrait, preserves first image,
And transfer the subsequent image after first image；
Processing unit, for detecting the portrait for corresponding to first image in the subsequent image and being in closed-eye state
Eyes whether open；If opening, using corresponding this subsequent image of eyes-open state to first image corresponding position
Local treatment is done in the eye of closed-eye state, so that the eyes of owner's picture are in eyes-open state in first image.
The device of the portrait processing of anti-eye closing provided in an embodiment of the present invention effectively can avoid taking the eye of portrait
Eyeball is in the photo of closed-eye state, greatly improves the quality of photo in itself, optimizes user experience.
It please refers to Fig.4, is the structure of second embodiment of device of the portrait processing of anti-eye closing provided by the invention
Block diagram.On the basis of the one embodiment for the device that the embodiment of the present invention is handled in the portrait of anti-eye closing, add
Display unit and start unit.
The device of the portrait processing of the anti-eye closing, including：
Shooting unit confirms shooting instruction for obtaining, and image to be confirmed, institute are obtained by the interface of finding a view for shooting portrait
State image to be confirmed include continuously acquire several do not do digital image compression processing image；
For obtaining first image in the image to be confirmed, institute is detected using Face datection algorithm for detection unit
State the state of the eyes of owner's picture in first image；
Unit is transferred, if being in closed-eye state for the eyes of at least one portrait, preserves first image,
And transfer the subsequent image after first image；
Processing unit, for detecting the portrait for corresponding to first image in the subsequent image and being in closed-eye state
Eyes whether open；If opening, using corresponding this subsequent image of eyes-open state to first image corresponding position
Local treatment is done in the eye of closed-eye state, so that the eyes of owner's picture are in eyes-open state in first image.
Wherein, the processing unit, is specifically used for：
In preset period of time, detect in the subsequent image and be in closed-eye state corresponding to first image
Whether the eyes of portrait are opened；
If opening, first image is corresponded in eye closing shape using corresponding this subsequent image of eyes-open state
The eye of state does Local treatment, so that the eyes of owner's picture are in eyes-open state in first image；To doing the office
The image that the eyes of the owner's picture obtained after portion's processing are in eyes-open state carries out digital image compression processing, through the number
Final portrait picture after word image Compression is shown in screen；
If not opening, digital image compression processing is carried out to first image, through the digital picture pressure
Treated that final portrait picture is shown in screen for contracting.
Wherein, the digital image compression processing is to be handled using the digital image compression of JPEG encryption algorithms.
Wherein, the device of the portrait processing of the anti-eye closing further includes display unit, the display unit, if for
The eyes of nobody's picture are in closed-eye state, then digital image compression processing are carried out to first image, through the number
Final portrait picture after word image Compression is shown in screen.
Wherein, the state that the eyes of owner's picture in first image are detected using Face datection algorithm, bag
It includes：
The eye feature value of owner's picture in first image, the eye feature are detected using Face datection algorithm
Pupil middle height, interpupillary distance or the pupil gray value that value includes；
When any portrait pupil middle height be less than or equal to preset height, then judge that the eyes of the portrait are in eye closing shape
State；When any portrait pupil middle height be more than the preset height, then judge that the eyes of the portrait are in eyes-open state,
Or；
When the interpupillary distance for not detecting any portrait, then judge that the eyes of the portrait are in closed-eye state；When detection is taken office
The interpupillary distance of one portrait, then judge the portrait eyes be in eyes-open state or；
When the pupil gray value of any portrait is less than or equal to default pupil gray value, then judges that the eyes of the portrait are in and close
Eye shape state；When any portrait pupil gray value be more than the default pupil gray value, then judge the portrait eyes be in open
Eye shape state.
Wherein, the device of the portrait processing of the anti-eye closing further includes start unit, the start unit, for opening
Dynamic camera function.
The device of the portrait processing of anti-eye closing provided in an embodiment of the present invention, remarkable face detection algorithm can be got
Then the state of human eye carries out eye Local treatment to the portrait of eye closing, the eyes-open state of portrait is synthesized to first image
On, eye closing portrait is made to be in eyes-open state, to take higher-quality portrait photo, effectively avoids taking the photograph of eye closing portrait
Piece.
A kind of method and apparatus of the portrait processing of anti-eye closing can effectively avoid the eyes for taking portrait from being in and close
The photo of eye shape state greatly improves the quality of photo in itself, optimizes user experience.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
It completes, relevant hardware can also be instructed to complete by program, which can be stored in a computer-readable storage medium
In matter, storage medium can include memory, disk or CD etc..
More than content is only presently preferred embodiments of the present invention, according to the invention for those of ordinary skill in the art
Thought, there will be changes, this specification content should not be construed as to the present invention in specific embodiments and applications
Limitation.
Claims (10)
1. a kind of method of the portrait processing of anti-eye closing, which is characterized in that including：
It obtains and confirms shooting instruction, image to be confirmed is obtained by the interface of finding a view for shooting portrait, the image to be confirmed includes
Several continuously acquired do not do the image that digital image compression is handled；
First image in the image to be confirmed is obtained, is detected in first image and owned using Face datection algorithm
The state of the eyes of portrait, wherein the detection includes the ocular that face sorting is surveyed, surveyed based on the face sorting
Detection and eyes detection；
The face sorting survey includes：Carry out the feature point extraction on face and classification；It will be to be determined according to the characteristic point
Facial image and face template carry out template matches, if it does, then it is empty that facial image to be determined is projected to face
Between；Projector distance of the test image window in the face subspace is calculated, if video in window meets threshold value comparison condition,
Facial image to be determined is judged for face, is turned into wherein the face subspace is sub-face of feature in corresponding image space
Subspace, the spy that the feature vector of the sample set covariance matrix of the sub-face of feature including eyes, cheek and lower jaw is formed
Levy eye, feature jaw and feature lip, the feature vector of the sample set covariance matrix is from the angle of statistics by feature sub-face technology
Spend the basic element for the facial image distribution to be determined found；
The ocular detection includes：According to the image of the definite face part, the three front yards ratio-dependent of five is utilized
The initial position of nose；Two initial ranging rectangles are defined, respectively the location of two growths to left and right, according to human eye ash
The characteristics of degree is less than face gray scale, the edge of eye is found using described search rectangle, and positions the center of pupil in eyeball；Root
The ocular of the face part is determined according to the center of the initial position of the nose, the edge of eye and pupil；
The eyes detection includes：According to eye feature value, eyes-open state is in eyes or closed-eye state is sentenced
It is disconnected, wherein the eye feature value includes pupil middle height, interpupillary distance or pupil gray value；
If the eyes of at least one portrait are in closed-eye state, first image is preserved, and transfers described first
Subsequent image after image；
In preset period of time 2s, detect and correspond to the people that first image is in closed-eye state in the subsequent image
Whether the eyes of picture are opened；If opening, first image is corresponded to using corresponding this subsequent image of eyes-open state
Eye in closed-eye state does Local treatment, so that the eyes of owner's picture are in eyes-open state in first image；
The image that the eyes for making the owner's picture obtained after the Local treatment are in eyes-open state carries out digital image compression processing,
Through the digital image compression, treated that final portrait picture is shown in screen；Wherein, the Local treatment is using eye
The eye that portion is in the subsequent image of eyes-open state, which replaces, to be corresponded to portrait in first image and is in closed-eye state
The eye of eyes-open state on the premise of face characteristic coincide, is directly covered in the eye of closed-eye state by eye；
If not opening, digital image compression processing is carried out to first image, at through the digital image compression
Final portrait picture after reason is shown in screen.
2. the method for the portrait processing of anti-eye closing according to claim 1, which is characterized in that the digital picture pressure
Contracting processing is to be handled using the digital image compression of JPEG encryption algorithms.
3. the method for the portrait processing of anti-eye closing according to claim 1, which is characterized in that described to be examined using face
In method of determining and calculating detection first image after the state of the eyes of owner's picture, further include：
If the eyes of nobody's picture are in closed-eye state, digital image compression processing is carried out to first image, warp
Treated that final portrait picture is shown in screen for the digital image compression.
4. the method for the portrait processing of the anti-eye closing according to claim 1 or 3, which is characterized in that described to utilize people
Face detection algorithm detects the state of the eyes of owner's picture in first image, including：
The eye feature value of owner's picture in first image, the eye feature value bag are detected using Face datection algorithm
Pupil middle height, interpupillary distance or the pupil gray value included；
When any portrait pupil middle height be less than or equal to preset height, then judge that the eyes of the portrait are in closed-eye state；
When any portrait pupil middle height be more than the preset height, then judge the portrait eyes be in eyes-open state or；
When the interpupillary distance for not detecting any portrait, then judge that the eyes of the portrait are in closed-eye state；When detecting any people
The interpupillary distance of picture, then judge the portrait eyes be in eyes-open state or；
When the pupil gray value of any portrait is less than or equal to default pupil gray value, then judge that the eyes of the portrait are in eye closing shape
State；When any portrait pupil gray value be more than the default pupil gray value, then judge that the eyes of the portrait are in eye opening shape
State.
5. the method for the portrait processing of anti-eye closing according to claim 1, which is characterized in that described obtain confirms bat
It takes the photograph before instruction, further includes：Start camera function.
6. a kind of device of the portrait processing of anti-eye closing, which is characterized in that including：
Shooting unit confirms shooting instruction for obtaining, and image to be confirmed is obtained by the interface of finding a view for shooting portrait, described to treat
Confirm image include continuously acquire several do not do digital image compression processing image；
Detection unit for obtaining first image in the image to be confirmed, utilizes Face datection algorithm detection described the
The state of the eyes of owner's picture in one image is surveyed, wherein the detection includes face sorting based on the face part
The ocular detection of detection and eyes detection；
The face sorting survey includes：Carry out the feature point extraction on face and classification；It will be to be determined according to the characteristic point
Facial image and face template carry out template matches, if it does, then it is empty that facial image to be determined is projected to face
Between；Projector distance of the test image window in the face subspace is calculated, if video in window meets threshold value comparison condition,
Facial image to be determined is judged for face, is turned into wherein the face subspace is sub-face of feature in corresponding image space
Subspace, the spy that the feature vector of the sample set covariance matrix of the sub-face of feature including eyes, cheek and lower jaw is formed
Levy eye, feature jaw and feature lip, the feature vector of the sample set covariance matrix is from the angle of statistics by feature sub-face technology
Spend the basic element for the facial image distribution to be determined found；
The ocular detection includes：According to the image of the definite face part, the three front yards ratio-dependent of five is utilized
The initial position of nose；Two initial ranging rectangles are defined, respectively the location of two growths to left and right, according to human eye ash
The characteristics of degree is less than face gray scale, the edge of eye is found using described search rectangle, and positions the center of pupil in eyeball；Root
The ocular of the face part is determined according to the center of the initial position of the nose, the edge of eye and pupil；
The eyes detection includes：According to eye feature value, eyes-open state is in eyes or closed-eye state is sentenced
It is disconnected, wherein the eye feature value includes pupil middle height, interpupillary distance or pupil gray value；
Unit is transferred, if being in closed-eye state for the eyes of at least one portrait, first image is preserved, and adjusts
Take the subsequent image after first image；
Processing unit corresponds in preset period of time 2s, detecting in the subsequent image at first image
Whether opened in the eyes of the portrait of closed-eye state；If opening, using corresponding this subsequent image of eyes-open state to described
First image corresponds to the eye in closed-eye state and does Local treatment so that in first image owner's picture eyes
In eyes-open state；The image that the eyes for making the owner's picture obtained after the Local treatment are in eyes-open state carries out number
Image Compression, treated that final portrait picture is shown in screen through the digital image compression；Wherein, the part
It handles to replace in first image for the eye being in the subsequent image of eyes-open state using eye and correspond at portrait
In the eye of closed-eye state, on the premise of face characteristic coincide, the eye of eyes-open state is directly covered in closed-eye state
Eye；
If not opening, digital image compression processing is carried out to first image, at through the digital image compression
Final portrait picture after reason is shown in screen.
7. the device of the portrait processing of anti-eye closing according to claim 6, which is characterized in that the digital picture pressure
Contracting processing is to be handled using the digital image compression of JPEG encryption algorithms.
8. the device of the portrait processing of anti-eye closing according to claim 6, which is characterized in that the people of the anti-eye closing
Display unit, the display unit are further included as shooting the device handled, if being in closed-eye state for the eyes of nobody's picture,
Digital image compression processing then is carried out to first image, through the digital image compression treated final portrait figure
Piece is shown in screen.
9. the device of the portrait processing of the anti-eye closing according to claim 6 or 8, which is characterized in that described to utilize people
Face detection algorithm detects the state of the eyes of owner's picture in first image, including：
The eye feature value of owner's picture in first image, the eye feature value bag are detected using Face datection algorithm
Pupil middle height, interpupillary distance or the pupil gray value included；
When any portrait pupil middle height be less than or equal to preset height, then judge that the eyes of the portrait are in closed-eye state；
When any portrait pupil middle height be more than the preset height, then judge the portrait eyes be in eyes-open state or；
When the interpupillary distance for not detecting any portrait, then judge that the eyes of the portrait are in closed-eye state；When detecting any people
The interpupillary distance of picture, then judge the portrait eyes be in eyes-open state or；
When the pupil gray value of any portrait is less than or equal to default pupil gray value, then judge that the eyes of the portrait are in eye closing shape
State；When any portrait pupil gray value be more than the default pupil gray value, then judge that the eyes of the portrait are in eye opening shape
State.
10. the device of the portrait processing of anti-eye closing according to claim 6, which is characterized in that the anti-eye closing
The device of portrait processing further includes start unit, the start unit, for starting camera function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510416911.7A CN105072327B (en) | 2015-07-15 | 2015-07-15 | A kind of method and apparatus of the portrait processing of anti-eye closing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510416911.7A CN105072327B (en) | 2015-07-15 | 2015-07-15 | A kind of method and apparatus of the portrait processing of anti-eye closing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105072327A CN105072327A (en) | 2015-11-18 |
| CN105072327B true CN105072327B (en) | 2018-05-25 |
Family
ID=54501596
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510416911.7A Expired - Fee Related CN105072327B (en) | 2015-07-15 | 2015-07-15 | A kind of method and apparatus of the portrait processing of anti-eye closing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105072327B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105592249B (en) * | 2015-12-23 | 2018-12-21 | 努比亚技术有限公司 | The terminal and method of eyelid reconstruction are carried out to personage in photo based on eyeprint identification |
| CN105827969A (en) * | 2016-03-29 | 2016-08-03 | 乐视控股（北京）有限公司 | Intelligent camera method and device, and mobile device |
| CN106250426A (en) * | 2016-07-25 | 2016-12-21 | 深圳天珑无线科技有限公司 | A kind of photo processing method and terminal |
| US10148884B2 (en) * | 2016-07-29 | 2018-12-04 | Microsoft Technology Licensing, Llc | Facilitating capturing a digital image |
| CN107292261B (en) * | 2017-06-16 | 2021-07-13 | 深圳天珑无线科技有限公司 | Photographing method and mobile terminal thereof |
| CN107622483A (en) * | 2017-09-15 | 2018-01-23 | 深圳市金立通信设备有限公司 | A kind of image combining method and terminal |
| CN107743195A (en) * | 2017-10-09 | 2018-02-27 | 惠州Tcl移动通信有限公司 | A kind of photographic method, mobile terminal and storage medium |
| CN107833197B (en) * | 2017-10-31 | 2021-03-02 | Oppo广东移动通信有限公司 | Image processing method and device, computer readable storage medium and electronic equipment |
| CN108022206A (en) * | 2017-11-30 | 2018-05-11 | 广东欧珀移动通信有限公司 | Image processing method, device, electronic equipment and computer-readable recording medium |
| CN107911608A (en) * | 2017-11-30 | 2018-04-13 | 西安科锐盛创新科技有限公司 | The method of anti-shooting of closing one's eyes |
| CN108234878A (en) * | 2018-01-31 | 2018-06-29 | 广东欧珀移动通信有限公司 | Image processing method, device and terminal |
| CN108229443A (en) * | 2018-02-08 | 2018-06-29 | 三星电子（中国）研发中心 | A kind of photographic method and device of anti-eye closing |
| CN108462831B (en) * | 2018-03-18 | 2020-09-08 | Oppo广东移动通信有限公司 | Image processing method, image processing device, storage medium and electronic equipment |
| CN108259766B (en) * | 2018-03-29 | 2020-05-19 | 宁波大学 | Photographing and shooting processing method for mobile intelligent terminal |
| CN108259769B (en) * | 2018-03-30 | 2020-08-14 | Oppo广东移动通信有限公司 | Image processing method, image processing device, storage medium and electronic equipment |
| CN110717349A (en) * | 2018-07-11 | 2020-01-21 | 深圳纬目信息技术有限公司 | Table mat and state monitoring method |
| CN114170069A (en) * | 2021-11-25 | 2022-03-11 | 杭州电子科技大学上虞科学与工程研究院有限公司 | Automatic eye closing processing method based on continuous multiple pictures |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101617339A (en) * | 2007-02-15 | 2009-12-30 | 索尼株式会社 | Image processing apparatus and image processing method |
| WO2010029922A1 (en) * | 2008-09-12 | 2010-03-18 | アイシン精機株式会社 | Open/closed-eye judgment device, device for estimating degree of eye openness, and program |
| CN103369248A (en) * | 2013-07-20 | 2013-10-23 | 厦门美图移动科技有限公司 | Method for photographing allowing closed eyes to be opened |
| CN103716539A (en) * | 2013-12-16 | 2014-04-09 | 乐视致新电子科技（天津）有限公司 | Photographing method and apparatus |
| CN104052908A (en) * | 2013-03-15 | 2014-09-17 | 卡西欧计算机株式会社 | Display device and display method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8102417B2 (en) * | 2006-10-25 | 2012-01-24 | Delphi Technologies, Inc. | Eye closure recognition system and method |
| CN100463000C (en) * | 2007-09-27 | 2009-02-18 | 上海交通大学 | Human eye state detection method based on cascade classification and hough circle transform |
| JP5169139B2 (en) * | 2007-10-25 | 2013-03-27 | 株式会社ニコン | Camera and image recording program |
| CN102193621A (en) * | 2010-03-17 | 2011-09-21 | 三星电子（中国）研发中心 | Vision-based interactive electronic equipment control system and control method thereof |
| JP6304999B2 (en) * | 2013-10-09 | 2018-04-04 | アイシン精機株式会社 | Face detection apparatus, method and program |
-
2015
- 2015-07-15 CN CN201510416911.7A patent/CN105072327B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101617339A (en) * | 2007-02-15 | 2009-12-30 | 索尼株式会社 | Image processing apparatus and image processing method |
| WO2010029922A1 (en) * | 2008-09-12 | 2010-03-18 | アイシン精機株式会社 | Open/closed-eye judgment device, device for estimating degree of eye openness, and program |
| CN102150189A (en) * | 2008-09-12 | 2011-08-10 | 爱信精机株式会社 | Open-eye or closed-eye determination apparatus, degree of eye openness estimation apparatus and program |
| CN104052908A (en) * | 2013-03-15 | 2014-09-17 | 卡西欧计算机株式会社 | Display device and display method |
| CN103369248A (en) * | 2013-07-20 | 2013-10-23 | 厦门美图移动科技有限公司 | Method for photographing allowing closed eyes to be opened |
| CN103716539A (en) * | 2013-12-16 | 2014-04-09 | 乐视致新电子科技（天津）有限公司 | Photographing method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105072327A (en) | 2015-11-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105072327B (en) | A kind of method and apparatus of the portrait processing of anti-eye closing | |
| TWI751161B (en) | Terminal equipment, smart phone, authentication method and system based on face recognition | |
| CN105184246B (en) | Living body detection method and living body detection system | |
| CN105975935B (en) | A kind of face image processing process and device | |
| CN106127167B (en) | Recognition methods, device and the mobile terminal of target object in a kind of augmented reality | |
| CN108076290B (en) | Image processing method and mobile terminal | |
| US20170161553A1 (en) | Method and electronic device for capturing photo | |
| CN110837750B (en) | Face quality evaluation method and device | |
| CN109190522B (en) | Living body detection method based on infrared camera | |
| WO2017211139A1 (en) | Method and apparatus for implementing video communication | |
| CN107330370B (en) | Forehead wrinkle action detection method and device and living body identification method and system | |
| CN102946516A (en) | Mobile terminal and method for detecting blink action and realizing autodyne by mobile terminal | |
| CN107231529A (en) | Image processing method, mobile terminal and storage medium | |
| WO2015158087A1 (en) | Method and apparatus for detecting health status of human eyes and mobile terminal | |
| CN115599219B (en) | Eye protection control method, system and equipment for display screen and storage medium | |
| CN106881716A (en) | Human body follower method and system based on 3D cameras robot | |
| CN109977846B (en) | Living body detection method and system based on near-infrared monocular photography | |
| CN106412420B (en) | It is a kind of to interact implementation method of taking pictures | |
| CN106157262B (en) | Augmented reality processing method and device and mobile terminal | |
| CN112329727A (en) | Living body detection method and device | |
| CN114187166A (en) | Image processing method, intelligent terminal and storage medium | |
| JP2010239499A (en) | Communication terminal unit, communication control unit, method of controlling communication of communication terminal unit, and communication control program | |
| CN110309693A (en) | Multi-level state detecting system and method | |
| CN106599779A (en) | Human ear recognition method | |
| WO2020119518A1 (en) | Control method and device based on spatial awareness of artificial retina |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Changan town in Guangdong province Dongguan 523860 usha Beach Road No. 18 Patentee after: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS Corp.,Ltd. Address before: Changan town in Guangdong province Dongguan 523860 usha Beach Road No. 18 Patentee before: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS Corp.,Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180525 |