CN108491780A - Image landscaping treatment method, apparatus, storage medium and terminal device - Google Patents
Image landscaping treatment method, apparatus, storage medium and terminal device Download PDFInfo
- Publication number
- CN108491780A CN108491780A CN201810217900.XA CN201810217900A CN108491780A CN 108491780 A CN108491780 A CN 108491780A CN 201810217900 A CN201810217900 A CN 201810217900A CN 108491780 A CN108491780 A CN 108491780A
- Authority
- CN
- China
- Prior art keywords
- face
- identification
- beautification
- image
- parameter information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013527 convolutional neural network Methods 0.000 claims description 44
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 230000001815 facial effect Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 210000002569 neuron Anatomy 0.000 claims description 8
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 27
- 238000010586 diagram Methods 0.000 description 21
- 238000012545 processing Methods 0.000 description 18
- 210000000887 face Anatomy 0.000 description 14
- 230000002093 peripheral effect Effects 0.000 description 9
- 241000208340 Araliaceae Species 0.000 description 8
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 8
- 235000003140 Panax quinquefolius Nutrition 0.000 description 8
- 235000008434 ginseng Nutrition 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000013480 data collection Methods 0.000 description 2
- 210000004709 eyebrow Anatomy 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 208000003351 Melanosis Diseases 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
- 230000002087 whitening effect Effects 0.000 description 1
- 210000000216 zygoma Anatomy 0.000 description 1
Classifications
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/04—Context-preserving transformations, e.g. by using an importance map
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
A kind of image landscaping treatment method, apparatus, storage medium and the terminal device provided in the embodiment of the present application, this method determine the identification face in pending image by Face datection；Wherein, identification face is at least two, and identification face includes one and the corresponding main identification face of current focusing area, other identification faces are secondary identification face；The main corresponding first beautification parameter information of identification face and the corresponding second beautification parameter information of secondary identification face are determined respectively, wherein the corresponding beautification degree of the first beautification parameter information beautification degree corresponding higher than the second beautification parameter information；Corresponding landscaping treatment is carried out to the image-region of each identification face according to the corresponding beautification parameter information of each identification face.By using above-mentioned technical proposal, the landscaping effect of picture can be optimized, avoid secondary identification face excessively fuzzy, while the personage of focusing area can be protruded again, make mainly to identify that face is more prominent, improve the effect of image beautification.
Description
Technical field
The invention relates to technical field of image processing more particularly to a kind of image landscaping treatment method, apparatus, deposit
Storage media and terminal device.
Background technology
With the development of terminal device, it is equipped with camera on most terminal device, is carried out by terminal device
Taking pictures becomes a very universal function.Due to shooting environmental or the hardware problem of terminal device, lead to the picture of shooting not
Photographer can be allowed to be satisfied with, so picture progress landscaping treatment that generally can be according to image algorithm to shooting, but the prior art
Beautification mode can lead to the problem that regional area is excessively fuzzy in picture, so the landscaping treatment technology needs of picture are further excellent
Change.
Invention content
A kind of image landscaping treatment method, apparatus, storage medium and terminal device provided by the embodiments of the present application, Ke Yiyou
Change the image landscaping treatment scheme based on face.
In a first aspect, the embodiment of the present application provides a kind of image landscaping treatment method, including：
The identification face in pending image is determined by Face datection；Wherein, the identification face is at least two, institute
It includes one and the corresponding main identification face of current focusing area to state identification face, other identification faces are secondary identification
Face；
Determine that the corresponding first beautification parameter information of the main identification face and the secondary identification face correspond to respectively
The second beautification parameter information, wherein the first corresponding beautification degree of beautification parameter information is higher than the second beautification ginseng
The corresponding beautification degree of number information；
Phase is carried out to the image-region of each identification face according to each identification face corresponding beautification parameter information
The landscaping treatment answered.
Second aspect, the embodiment of the present application provide a kind of image landscaping treatment device, including：
Face recognition module, for determining the identification face in pending image by Face datection；Wherein, the identification
Face is at least two, and the identification face includes one and the corresponding main identification face of current focusing area, other
Identification face is secondary identification face；
Parameter determination module, for determining the corresponding first beautification parameter information and described of the main identification face respectively
Secondary identification face corresponding second beautifies parameter information, wherein described first beautifies the corresponding beautification degree height of parameter information
In the corresponding beautification degree of the second beautification parameter information；
Landscaping treatment module, for identifying the corresponding beautification parameter information of face to each identification face according to each
Image-region carry out corresponding landscaping treatment.
The third aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey
Sequence realizes the image landscaping treatment method as described in the embodiment of the present application when the program is executed by processor.
Fourth aspect, the embodiment of the present application provide a kind of terminal device, including memory, processor and are stored in storage
It can realize on device and when the computer program of processor operation, the processor execute the computer program as the application is real
Apply the image landscaping treatment method described in example.
A kind of image landscaping treatment scheme provided in the embodiment of the present application, is determined by Face datection in pending image
Identification face；Wherein, the identification face is at least two, and the identification face includes one and current focusing area phase
Corresponding main identification face, other identification faces are secondary identification face；Determine that the main identification face is corresponding respectively
First beautification parameter information and the corresponding second beautification parameter information of the secondary identification face, wherein the first beautification ginseng
The corresponding beautification degree of number information beautifies the corresponding beautification degree of parameter information higher than described second；According to each identification face point
Not corresponding beautification parameter information carries out corresponding landscaping treatment to the image-region of each identification face.By using above-mentioned skill
Art scheme, can identify the face in pending image, and adjust beautification parameter information according to the distance of face, so to away from
The face different from focusing area carries out rational landscaping treatment, can optimize the landscaping effect of picture, avoid secondary identification people
Face is excessively fuzzy, while can protrude the personage of focusing area again, makes mainly to identify that face is more prominent, improves the effect of image beautification
Fruit.
Description of the drawings
Fig. 1 is a kind of flow diagram of image landscaping treatment method provided by the embodiments of the present application；
Fig. 2 is a kind of schematic diagram of a scenario of image landscaping treatment method provided by the embodiments of the present application；
Fig. 3 is the schematic diagram of a scenario of pending image provided by the embodiments of the present application；
Fig. 4 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application；
Fig. 5 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application；
Fig. 6 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application；
Fig. 7 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application；
Fig. 8 is the schematic diagram of a scenario of the pending image of another kind provided by the embodiments of the present application；
Fig. 9 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application；
Figure 10 is a kind of structure diagram of image landscaping treatment device provided by the embodiments of the present application；
Figure 11 is a kind of structural schematic diagram of terminal device provided by the embodiments of the present application；
Figure 12 is the structural schematic diagram of another terminal device provided by the embodiments of the present application.
Specific implementation mode
Further illustrate the technical solution of the application below with reference to the accompanying drawings and specific embodiments.It is appreciated that
It is that specific embodiment described herein is used only for explaining the application, rather than the restriction to the application.It further needs exist for illustrating
, illustrate only for ease of description, in attached drawing and the relevant part of the application rather than entire infrastructure.
It should be mentioned that some exemplary embodiments are described as before exemplary embodiment is discussed in greater detail
The processing described as flow chart or method.Although each step is described as the processing of sequence, many of which by flow chart
Step can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of each step can be rearranged.When its operation
The processing can be terminated when completion, it is also possible to the additional step being not included in attached drawing.The processing can be with
Corresponding to method, function, regulation, subroutine, subprogram etc..
Terminal device may include smart mobile phone, camera, tablet computer and other equipment with operating system.User can be with
It is taken pictures by the camera on terminal device.The image that can also be shot to terminal device of terminal device carries out at beautification
Reason, and generally when shooting includes the image of multiple personages, the picture after landscaping treatment will appear the portrait far from focusing area
The problem excessively obscured, the technical solution of the embodiment of the present application can pass through portrait in the image of shooting and focusing area
Relationship dynamically adjusts landscaping treatment.
Fig. 1 is a kind of flow diagram of image landscaping treatment method provided by the embodiments of the present application, and this method can be by
Image landscaping treatment device executes, and wherein the device can generally be integrated in terminal device by software and or hardware realization
In, other can also be integrated in and be equipped in the equipment of operating system.As shown in Figure 1, this method includes：
S110, identification face in pending image is determined by Face datection；Wherein, the identification face is at least two
A, the identification face includes one and the corresponding main identification face of current focusing area, other identification faces are secondary
Identify face.
Wherein, the pending image includes at least two recognizable face.It is alternatively possible to pass through preset people
Face detection algorithm determines the identification face in pending image.
The pending image can be carried out acquired in captured in real-time by terminal device, optionally, described to wait locating
When reason image can be that user is shot by the camera lens of terminal device, the figure of real-time display on the screen of terminal device
Picture, with customer mobile terminal equipment, the image of real-time display is also changing on screen.
The pending image can also be that user passes through figure acquired after the camera lens progress shooting operation of terminal device
Picture.The pending image can also be that the image being stored in terminal device, pending image are not limited to pass through terminal
Captured by the camera lens of equipment, it can also be through the character image acquired in other channels.
The current focusing area can be when being shot by the camera lens of terminal device, the position area that camera lens is focused
The image-region in domain, the current focusing area described at this time is most clearly；User can also adjust current focusing area by being arranged
Domain.If the pending image is stored in the image in terminal device, the focusing area can be that user voluntarily sets
The region set.The current focusing area is corresponding to identify face to be main, other identification faces are secondary identification face, will be waited for
Identification face in processing image distinguishes, and may further determine that the beautification parameter information of each identification face.
S111, the corresponding first beautification parameter information of the main identification face and the secondary identification face are determined respectively
Corresponding second beautification parameter information, wherein described first beautifies the corresponding beautification degree of parameter information higher than described second U.S.
Change the corresponding beautification degree of parameter information.
The beautification parameter information includes attribute value corresponding with image beautification operation.Different image beautification operation packets
The parameter value for adjusting different image property values or being handled image according to different operational formulas is included, so that image reaches
Corresponding landscaping effect.
Illustratively, described image beautification operation includes mill skin, whitening, nti-freckle, toning and thin face etc., executes different figures
As the different attribute value of the corresponding adjustment image of beautification operation needs, for example, image property value includes the brightness value of image, contrast
With saturation degree etc..Calculation process can also be carried out to obtain by image-processing operations formula to image by executing image beautification operation
Image after beautification, illustratively, the operational formula include convolution algorithm and filtering operation etc..
It can be preset beautification parameter information that the main identification face corresponding first, which beautifies parameter information, preset
Beautify parameter information and current focusing area corresponds to.Because the secondary identification face is not in focusing area, secondary knowledge
The image definition of others' face is less than the main identification face, so beautifying the corresponding beautification of parameter information by described first
Degree beautifies the corresponding beautification degree of parameter information higher than described second, can make the beautification of the not identification face in focusing area
Effect reduces.Illustratively, the skin area of face, which carries out mill skin processing, to be identified to each of images to be recognized, if to secondary
It identifies that face and main identification face are handled using the mill skin of same degree, then secondary identification face can be caused excessively fuzzy
Problem.
So beautifying parameter letter higher than described second by the way that the corresponding beautification degree of the first beautification parameter information is arranged
Corresponding beautification degree is ceased, it can be to avoid secondary identification face because similarly beautifying parameter information using with main identification face
And there is fuzzy problem.
The beautification degree and the effect intensity of corresponding image beautification operation are directly proportional.The beautification degree is higher, right
The effect for the image beautification operation answered is stronger, and it is higher that the beautification of corresponding and image operates corresponding attribute value.Illustratively, if
Image beautification operation is to carry out mill skin processing to the skin area in image, and mill skin degree is higher, and mill skin handles corresponding fortune
The value for calculating the Fuzzy Influence parameter in formula is higher, and mill bark effect is also stronger.
S112, according to the corresponding beautification parameter information of each identification face to the image-region of each identification face into
The corresponding landscaping treatment of row.
Wherein, the image-region of the identification face is identification face region residing in images to be recognized, phase
The basis answered each identifies that the corresponding beautification parameter information of face carries out landscaping treatment to its image-region.
Optionally, can also according to identification face beautification parameter information to it is described identification face image-region except
Background area carries out landscaping treatment, and the background area between adjacent two identification facial images identifies face according to two
The difference for beautifying parameter information carries out gradual change landscaping treatment.Background area can also beautify parameter information according to benchmark and be beautified
Processing, or gradual change landscaping treatment is carried out according to the difference of beautification parameter information to the benchmark beautification parameter information of identification face.
A kind of image landscaping treatment scheme provided in the embodiment of the present application, can be to the pending image of acquisition into pedestrian
Face identifies, identifies the face in pending image, and adjust beautification parameter letter according to the distance apart from focusing area of face
Breath, and then rational landscaping treatment is carried out to the identification face inside and outside focusing area, the landscaping effect of picture can be optimized, avoided
Secondary identification face is excessively fuzzy, while can protrude the personage of focusing area again, makes mainly to identify that face is more prominent, improves figure
As the effect of beautification.
Fig. 2 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application, in above-mentioned implementation
On the basis of the technical solution that example is provided, optionally, as shown in Fig. 2, this method includes：
S120, identification face in pending image is determined by Face datection；Wherein, the identification face is at least two
A, the identification face includes one and the corresponding main identification face of current focusing area, other identification faces are secondary
Identify face.
S121, the corresponding first beautification parameter information of the main identification face is determined.
The specific implementation mode of aforesaid operations can refer to associated description above, and details are not described herein.
S122, the secondary identification face quantity be greater than or equal to two when, for each secondary identification face, really
Distance difference of the settled preceding secondary identification face with the main identification face in shooting direction.
The distance difference is the difference of the shooting distance of secondary identification face and the shooting distance of main identification face, institute
State the distance value to taking lens in real space where shooting distance identifies face.As shown in figure 3, passing through terminal device 00
Camera lens 01 shooting personage A and personage B, the shooting distance of personage A is L1, the shooting distance L2 of personage B.Based on personage A
Identify face, it is possible to determine that the distance difference of personage B and personage A in shooting direction is L2-L1.
Optionally, if the pending image is the image acquired in the camera lens captured in real-time by terminal device,
The distance value of the identification face and taking lens can be determined by the lens sensors of terminal device.If described pending
Image is the image being stored in terminal device, then the area of the image-regions of faces can be identified according to the difference in image,
Determine distance difference of the secondary face with the main identification face in shooting direction.
Wherein, different secondary identification faces corresponds to different distance differences, and different distance differences is then corresponding different
The second beautification parameter information, it is each to beautify parameter information and each identify that face is corresponding.The number of the secondary identification face
When amount is greater than or equal to two, it is determined that each secondary identification face is at a distance from the main identification face is in shooting direction
Difference, and then corresponding second beautification parameter information can be determined according to the distance difference of each secondary identification face.
S123, each secondary identification face is determined according to the distance difference and the first beautification parameter information respectively
Corresponding second beautification parameter information；Wherein, the distance difference is smaller, and described second beautifies the corresponding U.S. of parameter information
Change degree is higher.
Wherein, corresponding U.S. of the second beautification parameter information apart from the main identification closer secondary identification face of face
Change degree is higher.
The beautification degree is higher, and the error amount of corresponding beautification parameter information and standard parameter information is smaller.That is, beautification
The sum of included different classes of error amount of beautification parameter is smaller in parameter information and standard parameter information.The standard ginseng
Number information includes standard parameter corresponding with different classes of landscaping treatment, is beautified to image according to standard parameter information
The picture obtained after processing reaches best landscaping effect closer to the aesthetic effect of default.
And general standard parameter information is to be directed to standard-sized image, and the face proportion in image is standard ratio
Example, so being carried out at beautification using standard parameter information for the different secondary identification face of the distance of the main identification face of distance
The effect of reason is also different.So for the second beautification parameter letter of the main identification closer secondary identification face of face of distance
Breath is closer to established standards parameter information.
Illustratively, mill skin processing is carried out to the skin area of the identification face in images to be recognized, apart from described main
The closer secondary identification face of face is identified, than the mill skin degree of the secondary identification face apart from the main identification face farther out
Want high.Beautification parameter information is adjusted according to the distance of the main identification face of secondary identification face distance, and then to different distance
Face carry out rational landscaping treatment, the landscaping effect of picture can be optimized.For another example, images to be recognized is carried out at toning
Reason, beautification degree is higher, and the application region of toning operation in the picture is also more extensive.To the face of the face in images to be recognized
When size is adjusted, beautification degree is higher, and corresponding face size adjustment ratio is bigger.
Illustratively, Fig. 4 is the schematic diagram of a scenario of pending image provided by the embodiments of the present application, illustratively, such as Fig. 4
Shown, images to be recognized includes the image of 4 personages, wherein mainly identification personage is personage 13, wherein personage 10 and main
Identify that the distance difference of personage 13 is farthest, the distance difference of personage 11 and personage 12 reduce successively, so to wherein four people
When the image-region of object carries out landscaping treatment, personage 13 is beautified by the first beautification parameter information, beautifies degree highest；
Correspondingly, the second corresponding beautification degree of beautification parameter information of personage 10 is minimum, the beautification degree of personage 11 and personage 12 according to
It is secondary to get higher.
By determining the beautification degree in beautification parameter information corresponding with the identification face according to the distance difference,
Distance difference is smaller, and corresponding degree of beautifying is higher, and the beautification degree of the face remoter apart from focusing area can be made lower, kept away
Exempt from apart from the facial image of focusing area farther out excessively to be beautified and the fuzzy problem that arrives.
S124, according to the corresponding beautification parameter information of each identification face to the image-region of each identification face into
The corresponding landscaping treatment of row.
Specific implementation mode can refer to associated description above, and details are not described herein.
Fig. 5 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application, in above-mentioned implementation
On the basis of the technical solution that is provided of example, to according to the corresponding beautification parameter information of each identification face to each identification face
Image-region carry out the operation of corresponding landscaping treatment and be optimized, optionally, as shown in figure 5, this method includes：
S130, identification face in pending image is determined by Face datection；Wherein, the identification face is at least two
A, the identification face includes one and the corresponding main identification face of current focusing area, other identification faces are secondary
Identify face.
S131, the corresponding first beautification parameter information of the main identification face and the secondary identification face are determined respectively
Corresponding second beautification parameter information, wherein described first beautifies the corresponding beautification degree of parameter information higher than described second U.S.
Change the corresponding beautification degree of parameter information.
The specific implementation mode of aforesaid operations can refer to associated description above, and details are not described herein.
The face key position information of S132, each identification face of extraction.
S133, the corresponding beautification parameter information of face and face key position information are identified according to each, to each
Identify that the image-region of face carries out corresponding landscaping treatment.
Wherein, the face key position information includes the division area information for embodying face characteristic position, example
Property, it may include brow region, eye areas, nasal area, face region, cheekbone area, chin area or other characteristic areas
Domain etc..The region that the face key position information includes can be systemic presupposition, and the embodiment of the present application is not limited thereto.
After determining identification face, the face key position information that extraction identification face includes is then it was determined that described
Identify which face characteristic position each region is respectively in the corresponding image-region of face, and then can be according to face characteristic portion
Position and beautification parameter information more meticulously carry out landscaping treatment.
The face key position information of face is identified by extraction, and according to the corresponding beautification of each identification face
Parameter information and face key position information carry out corresponding landscaping treatment to the image-region of each identification face, can be with needle
To face key position information to identifying that the image-region of face carries out detailed-oriented landscaping treatment, image can be advanced optimized
The effect of landscaping treatment.
It should be noted that the execution sequence of the operation S133 is not limited to shown in attached drawing, as long as operation S133 exists
After operating S130, and executed before operation S134.
Optionally, as shown in fig. 6, the basis each identifies that the corresponding beautification parameter information of face and face are crucial
Location information, carrying out corresponding landscaping treatment to the image-region of each identification face can be implemented by following manner：
S141, the corresponding position beautification of key position is determined according to the corresponding beautification parameter information of each identification face
Parameter；Wherein, each face key position information includes at least one key position.
S142, beautify parameter according to the position to the progress landscaping treatment of corresponding key position.
Key position corresponds to the division region of different face characteristics respectively in the wherein described face key position information, crucial
Position may include face characteristic portion and other characteristic portions.For the difference of key position, the requirement of corresponding beautification also has
Institute is different, illustratively, if carrying out mill skin processing to face, needs to skin area progress Fuzzy Processing, and brow region
If or eye areas can cause eyebrow and eye areas to be also blurred similarly by carry out Fuzzy Processing.So being directed to
Eyebrow and the fog-level of eye areas need the fog-level less than skin area, can be to avoid brow region and eye areas
It is excessively fuzzy.
Beautify parameter according to the corresponding position of each different key position, which is carried out at corresponding beautification
Reason can carry out suitable landscaping treatment to different key positions, advanced optimize the effect of image beautification.
Fig. 7 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application, above-mentioned arbitrary
On the basis of the technical solution that embodiment is provided, closed to being determined according to the corresponding beautification parameter information of each identification face
The operation of the corresponding position beautification parameter in key position is optimized, optionally, as shown in fig. 7, this method includes：
S150, identification face in pending image is determined by Face datection；Wherein, the identification face is at least two
A, the identification face includes one and the corresponding main identification face of current focusing area, other identification faces are secondary
Identify face.
S151, the corresponding first beautification parameter information of the main identification face and the secondary identification face are determined respectively
Corresponding second beautification parameter information, wherein described first beautifies the corresponding beautification degree of parameter information higher than described second U.S.
Change the corresponding beautification degree of parameter information.
The face key position information of S152, each identification face of extraction.
The specific implementation mode of aforesaid operations can refer to associated description above, and details are not described herein.
S153, determine that the reference position of each identification face is sat according to the face key position information of each identification face
Mark.
Wherein, the focusing area is the region for all key positions for including the identification face, the reference position
Coordinate is reference position when different key positions are carried out with landscaping treatment.Illustratively, as shown in figure 8, focusing area is Fig. 8
Region in the box, the reference position coordinate is the center C points of focusing area, using C points as reference position.
S154, the corresponding position beautification of key position is determined according to the corresponding beautification parameter information of each identification face
Parameter, and parameter is beautified according to the corresponding position of key position described in the reference position Coordinate Adjusting.
It is alternatively possible to adjust position beautification ginseng according to the distance of the position coordinates of key position to reference position coordinate
Number.Illustratively, when carrying out shape of face adjustment landscaping treatment to identification portrait, using C points as reference position, the nearlyr correspondence of distance C points
Adjustment degree it is smaller, the more remote corresponding adjustment degree of distance C points is bigger, so shape of face adjustment effect can be made more natural, avoided out
The problem of being now adjusted to portrait distortion.
By the reference position coordinate of determining focusing area, and according to the corresponding beautification parameter of each identification face
Information determines the corresponding position beautification parameter of key position, and according to the corresponding portion of the reference position Coordinate Adjusting key position
Position beautification parameter, can advanced optimize the effect of image beautification.
S155, beautify parameter according to the position to the progress landscaping treatment of corresponding key position.
Specific implementation mode can refer to associated description above, and details are not described herein.
Fig. 9 is the flow diagram of another image landscaping treatment method provided by the embodiments of the present application, above-mentioned arbitrary
On the basis of the technical solution that embodiment is provided, to determining the operation of the identification face in pending image by Face datection
It is optimized, optionally, as shown in figure 9, this method includes：
S160, pending image is obtained.
It is alternatively possible to which the camera lens by terminal device obtains pending image, it can also be that acquisition prestores pending
Image.The pending image is character image, and the pending image includes at least two recognizable face.
S161, face key position identification is carried out to the pending image by default convolutional neural networks, obtains institute
The face key position information in pending image is stated, and the pending figure is determined according to the face key position information
Identification face as in.
Convolutional neural networks (CNN, Constitutional Neural Networks) are one kind in multilayer neural network
On the basis of a kind of deep learning method for being designed for image classification and identification, the face that can be directed to facial image is crucial
Position is trained convolutional neural networks to generate default convolutional neural networks.
In the prior art terminal device carry out face Atomatic focusing method be by the head of people, face, eyes, nose, mouth etc. by
It is written in the firmware or software of terminal device in advance according to information such as shape, tone, arrangement modes, this mode needs to understand in advance
The details of different objects, and if when some faces are not recorded in database, which can not be identified, be led
Focusing failure is caused, it is bad to further result in landscaping effect.
Default convolutional neural networks can carry out face key position identification to the pending image of acquisition, described in acquisition
Face key position information in pending image determines the knowledge in the pending image further according to face key position information
Others' face.Each identification face has corresponding face key position information, if the pending image includes multiple knowledges
Others' face, then multiple face key position information can be obtained by presetting convolutional neural networks, and according to each face key position
Information determines the identification face in corresponding pending image.
It is optionally, described that face key position identification is carried out to the pending image by default convolutional neural networks,
The face key position information obtained in the pending image can be implemented by following manner：
S1610, multilayer convolution algorithm is carried out to pending image, every layer of convolution algorithm, which corresponds to, extracts the one of pending image
A eigenmatrix.
The eigenmatrix that S1611, basis are extracted successively gradually determines the high level of the face key position in pending image
Eigenmatrix.
Wherein, the multilayer convolution algorithm includes at least one convolutional layer, carries out convolution algorithm to pending image, often
Convolution algorithm can obtain an eigenmatrix, and high-rise spy can be obtained by carrying out multilayer convolution algorithm to pending image
Levy matrix；For pending image, high-level characteristic matrix is more abstract.
S1612, face key position information is determined according to the high-level characteristic matrix of the face key position.
The high-level characteristic matrix handled by multilayer convolution algorithm, which can more embody, to be needed to extract in pending image
Feature can determine face key position information according to high-level characteristic matrix.
Face key position identification is carried out to pending image by default convolutional neural networks, can automatically extract and wait locating
The face key position information in image is managed, and quickly determines the position of identification face according to the face key position information of extraction
It sets, the efficiency of Face datection can be improved.
Optionally, before carrying out face key position identification to the pending image by default convolutional neural networks,
Further include：
It is trained according to human face data set pair convolutional neural networks, obtains default convolutional neural networks, it is described default
Convolutional neural networks are used to determine the face key position information of the pending image.
The method that can be shared by part connection and weights, makes convolutional neural networks become the god with local receptor field
Through network, the neural network with local receptor field can extract visual signature corresponding with face key position information.
Part connection refers to each neuron on convolutional layer and establishes connection with the local neuron in preceding layer convolutional layer,
The local neuron corresponds to the characteristic area that convolutional layer is extracted；
Weights are shared refer to neuron in same convolutional layer with one group of identical weights with preceding layer is local connect, pass through
Setting weights are shared can to reduce network training parameter.One group of identical weights is a feature extractor, or is a volume
Product core, the convolution kernel numerical value of most initial can be default, and convolution kernel numerical value can be updated with the training of convolutional neural networks.
Can the facial image that human face data is concentrated be input to convolutional neural networks to be trained, and update every layer of convolution
The corresponding weights of layer obtain default convolutional neural networks.
Wherein, human face data concentration includes multiple and different facial images and the corresponding face key position mark of facial image
Remember the data acquisition system of information, the personage's angle and character features of different facial images are all different, different facial images tool
There is different face key position label informations.Illustratively, human face data collection can be CelebA human face data collection, convolution god
Can be VGG (Visual Geometry Group) through network, VGG is by Oxford University's computer vision group and Google
The depth convolutional neural networks that DeepMind companies research and develop together.
It is trained by human face data set pair convolutional neural networks, convolutional neural networks are multiple not to human face data concentration
Same facial image, and each corresponding face key position label information of facial image are learnt, and are generated and preset convolution
Neural network, default convolutional neural networks can determine the face key position information of the pending image of acquisition.
After being trained by human face data set pair convolutional neural networks, default convolutional neural networks can be improved and treat place
The identification accuracy of the identification face in image is managed, even if pending image includes having corresponding different face key position letters
The identification face of breath can also be predetermined convolutional neural networks and identify.
S162, the corresponding first beautification parameter information of the main identification face and the secondary identification face are determined respectively
Corresponding second beautification parameter information, wherein described first beautifies the corresponding beautification degree of parameter information higher than described second U.S.
Change the corresponding beautification degree of parameter information.
S163, according to the corresponding beautification parameter information of each identification face to the image-region of each identification face into
The corresponding landscaping treatment of row.
The specific implementation mode of aforesaid operations can refer to associated description above, and details are not described herein.
Figure 10 is a kind of structure diagram of image landscaping treatment device provided by the embodiments of the present application, which can execute
Image landscaping treatment method, as shown in Figure 10, which includes：
Face recognition module 210, for determining the identification face in pending image by Face datection；Wherein, described
Identifying that face is at least two, the identification face includes one and the corresponding main identification face of current focusing area,
Other identification faces are secondary identification face；
Parameter determination module 211, for determine respectively the main identification face it is corresponding first beautification parameter information and
The secondary identification face corresponding second beautifies parameter information, wherein described first beautifies the corresponding beautification journey of parameter information
Degree beautifies the corresponding beautification degree of parameter information higher than described second；
Landscaping treatment module 212, for identifying the corresponding beautification parameter information of face to each identification according to each
The image-region of face carries out corresponding landscaping treatment.
A kind of image landscaping treatment device provided in the embodiment of the present application, is determined by Face datection in pending image
Identification face；Wherein, the identification face is at least two, and the identification face includes one and current focusing area phase
Corresponding main identification face, other identification faces are secondary identification face；Determine that the main identification face is corresponding respectively
First beautification parameter information and the corresponding second beautification parameter information of the secondary identification face, wherein the first beautification ginseng
The corresponding beautification degree of number information beautifies the corresponding beautification degree of parameter information higher than described second；According to each identification face point
Not corresponding beautification parameter information carries out corresponding landscaping treatment to the image-region of each identification face.By using above-mentioned skill
Art scheme, can identify the face in pending image, and adjust beautification parameter information according to the distance of face, so to away from
The face different from focusing area carries out rational landscaping treatment, can optimize the landscaping effect of picture, avoid secondary identification people
Face is excessively fuzzy, while can protrude the personage of focusing area again, makes mainly to identify that face is more prominent, improves the effect of image beautification
Fruit.
Optionally, the parameter determination module is specifically used for：
When the quantity of the secondary identification face is greater than or equal to two, for each secondary identification face, determination is worked as
Distance difference of the preceding secondary identification face with the main identification face in shooting direction；
Each secondary identification face difference is determined respectively according to the distance difference and the first beautification parameter information
Corresponding second beautification parameter information；Wherein, the distance difference is smaller, and described second beautifies the corresponding beautification journey of parameter information
Degree is higher.
Optionally, the beautification degree is higher, and the error amount of corresponding beautification parameter information and standard parameter information is smaller.
Optionally, further include：
Location information extraction module carries after the identification face in determining pending image by Face datection
Take the face key position information of each identification face；
Correspondingly, landscaping treatment module is specifically used for：According to the corresponding beautification parameter information of each identification face and
Face key position information carries out corresponding landscaping treatment to the image-region of each identification face.
Optionally, landscaping treatment module specifically includes：
Positional parameter determining module, for determining key portion according to the corresponding beautification parameter information of each identification face
Parameter is beautified at the corresponding position in position；Wherein, each face key position information includes at least one key position；
Key beautification module carries out landscaping treatment for beautifying parameter according to the position to corresponding key position.
Optionally, crucial beautification module includes：
Reference coordinate determining module, for determining each identification people according to the face key position information of each identification face
The reference position coordinate of face；
Beautify parameter adjustment module, for determining key portion according to the corresponding beautification parameter information of each identification face
Parameter is beautified at the corresponding position in position, and according to the corresponding position beautification ginseng of key position described in the reference position Coordinate Adjusting
Number.
Optionally, face recognition module specifically includes：
Image collection module, for obtaining pending image；
Neural Network for Face Recognition module, for carrying out face to the pending image by default convolutional neural networks
Key position identifies, obtains the face key position information in the pending image, and according to the face key position
Information determines the identification face in the pending image.
Optionally, Neural Network for Face Recognition module is specifically used for：
Multilayer convolution algorithm is carried out to pending image, every layer of convolution algorithm corresponds to a feature for extracting pending image
Matrix；
According to the eigenmatrix extracted successively, the high-level characteristic square of the face key position in pending image is gradually determined
Battle array；
Face key position information is determined according to the high-level characteristic matrix of the face key position.
Optionally, further include：
Neural metwork training module, for carrying out face pass to the pending image by default convolutional neural networks
It before the identification of key position, is trained according to human face data set pair convolutional neural networks, obtains default convolutional neural networks, it is described
Default convolutional neural networks are used to determine the face key position information of the pending image.
Optionally, the neural metwork training module, is specifically used for：
Each neuron on the convolutional layer of convolutional neural networks is built with the local neuron in preceding layer convolutional layer
Vertical connection, and corresponding weights are set for every layer of convolutional layer, so that convolutional neural networks have extraction face key position information
The local receptor field of corresponding visual signature；
The facial image that human face data is concentrated is input to convolutional neural networks to be trained, and updates every layer of convolutional layer pair
The weights answered obtain default convolutional neural networks.
A kind of storage medium including computer executable instructions that the embodiment of the present application is provided, computer are executable
The image landscaping treatment operation being not limited to the described above is instructed, it is beautiful that the image that the application any embodiment is provided can also be performed
Change the relevant operation in processing method.
The embodiment of the present application also provides a kind of storage medium including computer executable instructions, and the computer is executable
When being executed by computer processor for executing image landscaping treatment method, this method includes for instruction：
The identification face in pending image is determined by Face datection；Wherein, the identification face is at least two, institute
It includes one and the corresponding main identification face of current focusing area to state identification face, other identification faces are secondary identification
Face；
Determine that the corresponding first beautification parameter information of the main identification face and the secondary identification face correspond to respectively
The second beautification parameter information, wherein the first corresponding beautification degree of beautification parameter information is higher than the second beautification ginseng
The corresponding beautification degree of number information；
Phase is carried out to the image-region of each identification face according to each identification face corresponding beautification parameter information
The landscaping treatment answered.
Storage medium --- any various types of memory devices or storage device.Term " storage medium " is intended to wrap
It includes：Install medium, such as CD-ROM, floppy disk or magnetic tape equipment；Computer system memory or random access memory, such as
DRAM, DDR RAM, SRAM, EDO RAM, blue Bath (Rambus) RAM etc.；Nonvolatile memory, such as flash memory, magnetic medium
(such as hard disk or optical storage)；The memory component etc. of register or other similar types.Storage medium can further include other
Memory of type or combinations thereof.In addition, storage medium can be located at program in the first computer system being wherein performed,
Or can be located in different second computer systems, second computer system is connected to the by network (such as internet)
One computer system.Second computer system can provide program instruction to the first computer for executing." storage is situated between term
Matter " may include may reside in different location two of (such as in different computer systems by network connection) or
More storage mediums.Storage medium can store the program instruction that can be executed by one or more processors and (such as implement
For computer program).
The embodiment of the present application provides a kind of terminal device, and figure provided by the embodiments of the present application can be integrated in the terminal device
As landscaping treatment device.
Figure 11 is a kind of structural schematic diagram of terminal device provided by the embodiments of the present application, and the embodiment of the present application provides one
Kind of terminal device 30, including memory 31, processor 32 and are stored in the computer that can be run on memory 31 and in processor
Program, the processor realize the image landscaping treatment method described in above-described embodiment when executing the computer program.This Shen
Please the terminal device that provides of embodiment, can identify the face in pending image, and beautification is adjusted according to the distance of face
Parameter information, and then the rational landscaping treatment of face progress that focusing area of adjusting the distance is different, can optimize the beautification effect of picture
Fruit avoids secondary identification face excessively fuzzy, while can protrude the personage of focusing area again, makes mainly to identify that face is more prominent
Go out, improves the effect of image beautification.
Figure 12 is a kind of structural schematic diagram of terminal device provided by the embodiments of the present application.As shown in figure 12, which sets
It is standby to may include：Shell (not shown), touch screen (not shown), touch key-press (not shown), memory
301, central processing unit (Central Processing Unit, CPU) 302 (also known as processor, hereinafter referred to as CPU), circuit board
(not shown) and power circuit (not shown).The circuit board is placed in the space interior that the shell surrounds；Institute
It states CPU302 and the memory 301 is arranged on the circuit board；The power circuit, for being each of the terminal device
A circuit or device power supply；The memory 301, for storing executable program code；The CPU302 passes through described in reading
The executable program code stored in memory 301 runs computer program corresponding with the executable program code, with
Realize following steps：
The identification face in pending image is determined by Face datection；Wherein, the identification face is at least two, institute
It includes one and the corresponding main identification face of current focusing area to state identification face, other identification faces are secondary identification
Face；
Determine that the corresponding first beautification parameter information of the main identification face and the secondary identification face correspond to respectively
The second beautification parameter information, wherein the first corresponding beautification degree of beautification parameter information is higher than the second beautification ginseng
The corresponding beautification degree of number information；
Phase is carried out to the image-region of each identification face according to each identification face corresponding beautification parameter information
The landscaping treatment answered.
The terminal device further includes：Peripheral Interface 303, RF (Radio Frequency, radio frequency) circuit 305, audio-frequency electric
Road 306, loud speaker 311, power management chip 308, input/output (I/O) subsystem 309, touch screen 312, other input/controls
Control equipment 310 and outside port 304, these components are communicated by one or more communication bus or signal wire 307.
It should be understood that graphic terminal 300 is only an example of terminal device, and terminal device 300
Can have than shown in the drawings more or less component, can combine two or more components, or can be with
It is configured with different components.Various parts shown in the drawings can be including one or more signal processings and/or special
It is realized in the combination of hardware, software or hardware and software including integrated circuit.
Just the terminal device provided in this embodiment for realizing image landscaping treatment is described in detail below, the end
End equipment is by taking mobile phone as an example.
Memory 301, the memory 301 can be by access such as CPU302, Peripheral Interfaces 303, and the memory 301 can
Can also include nonvolatile memory to include high-speed random access memory, such as one or more disk memory,
Flush memory device or other volatile solid-state parts.
The peripheral hardware that outputs and inputs of equipment can be connected to CPU302 and deposited by Peripheral Interface 303, the Peripheral Interface 303
Reservoir 301.
I/O subsystems 309, the I/O subsystems 309 can be by the input/output peripherals in equipment, such as touch screen 312
With other input/control devicess 310, it is connected to Peripheral Interface 303.I/O subsystems 309 may include 3091 He of display controller
One or more input controllers 3092 for controlling other input/control devicess 310.Wherein, one or more input controls
Device 3092 processed receives electric signal from other input/control devicess 310 or sends electric signal to other input/control devicess 310,
Other input/control devicess 310 may include physical button (pressing button, rocker buttons etc.), dial, slide switch, behaviour
Vertical pole clicks idler wheel.It is worth noting that input controller 3092 can with it is following any one connect：Keyboard, infrared port,
The indicating equipment of USB interface and such as mouse.
Touch screen 312, the touch screen 312 are the input interface and output interface between subscriber terminal equipment and user,
Visual output is shown to user, visual output may include figure, text, icon, video etc..
Display controller 3091 in I/O subsystems 309 receives electric signal from touch screen 312 or is sent out to touch screen 312
Electric signals.Touch screen 312 detects the contact on touch screen, and the contact detected is converted to and is shown by display controller 3091
The interaction of user interface object on touch screen 312, that is, realize human-computer interaction, the user interface being shown on touch screen 312
Object can be the icon of running game, be networked to the icon etc. of corresponding network.It is worth noting that equipment can also include light
Mouse, light mouse are the extensions for the touch sensitive surface for not showing the touch sensitive surface visually exported, or formed by touch screen.
RF circuits 305 are mainly used for establishing the communication of mobile phone and wireless network (i.e. network side), realize mobile phone and wireless network
The data receiver of network and transmission.Such as transmitting-receiving short message, Email etc..Specifically, RF circuits 305 receive and send RF letters
Number, RF signals are also referred to as electromagnetic signal, and RF circuits 305 convert electrical signals to electromagnetic signal or electromagnetic signal is converted to telecommunications
Number, and communicated with communication network and other equipment by the electromagnetic signal.RF circuits 305 may include for executing
The known circuit of these functions comprising but it is not limited to antenna system, RF transceivers, one or more amplifiers, tuner, one
A or multiple oscillators, digital signal processor, CODEC (COder-DECoder, coder) chipset, user identifier mould
Block (Subscriber Identity Module, SIM) etc..
Voicefrequency circuit 306 is mainly used for receiving audio data from Peripheral Interface 303, which is converted to telecommunications
Number, and the electric signal is sent to loud speaker 311.
Loud speaker 311, the voice signal for receiving mobile phone from wireless network by RF circuits 305, is reduced to sound
And play the sound to user.
Power management chip 308, the hardware for being connected by CPU302, I/O subsystem and Peripheral Interface are powered
And power management.
Terminal device provided by the embodiments of the present application can identify the face in pending image, and according to the remote of face
Recently adjustment beautification parameter information, and then the rational landscaping treatment of face progress that focusing area of adjusting the distance is different, can optimize
The landscaping effect of picture avoids secondary identification face excessively fuzzy, while can protrude the personage of focusing area again, makes mainly to know
Others' face is more prominent, improves the effect of image beautification.
It is arbitrary that image landscaping treatment device, storage medium and the terminal device provided in above-described embodiment can perform the application
The image landscaping treatment method that embodiment is provided has and executes the corresponding function module of this method and advantageous effect.Not upper
The technical detail of detailed description in embodiment is stated, reference can be made to the image landscaping treatment method that the application any embodiment is provided.
Note that above are only preferred embodiment and the institute's application technology principle of the application.It will be appreciated by those skilled in the art that
The application is not limited to specific embodiment described here, can carry out for a person skilled in the art it is various it is apparent variation,
The protection domain readjusted and substituted without departing from the application.Therefore, although being carried out to the application by above example
It is described in further detail, but the application is not limited only to above example, in the case where not departing from the application design, also
May include other more equivalent embodiments, and scope of the present application is determined by scope of the appended claims.
Claims (13)
1. a kind of image landscaping treatment method, which is characterized in that including：
The identification face in pending image is determined by Face datection；Wherein, the identification face is at least two, the knowledge
Others' face includes one and the corresponding main identification face of current focusing area, other identification faces are secondary identification people
Face；
Determine that the main identification face corresponding first beautifies parameter information and the secondary identification face corresponding the respectively
Two beautification parameter informations, wherein described first, which beautifies the corresponding beautification degree of parameter information, beautifies parameter letter higher than described second
Cease corresponding beautification degree；
The image-region of each identification face is carried out according to the corresponding beautification parameter information of each identification face corresponding
Landscaping treatment.
2. the method as described in claim 1, which is characterized in that be greater than or equal to two in the quantity of the secondary identification face
When, the corresponding second beautification parameter information of the determination secondary identification face includes：
For each secondary identification face, determine current secondary identification face with the main identification face in shooting direction
Distance difference；
Determine that each secondary identification face corresponds to respectively respectively according to the distance difference and the first beautification parameter information
Second beautification parameter information；Wherein, the distance difference is smaller, and described second, which beautifies the corresponding beautification degree of parameter information, gets over
It is high.
3. method as claimed in claim 2, which is characterized in that the beautification degree is higher, corresponding beautification parameter information and
The error amount of standard parameter information is smaller.
4. the method as described in claim 1, which is characterized in that the identification determined by Face datection in pending image
After face, further include：
The face key position information of each identification face of extraction；
Correspondingly, the image-region of each identification face is carried out according to the corresponding beautification parameter information of each identification face
Corresponding landscaping treatment includes：
According to the corresponding beautification parameter information of each identification face and face key position information, to each identification face
Image-region carries out corresponding landscaping treatment.
5. method as claimed in claim 4, which is characterized in that the basis each identifies the corresponding beautification parameter of face
Information and face key position information carry out corresponding landscaping treatment to the image-region of each identification face, including：
Determine that parameter is beautified at the corresponding position of key position according to the corresponding beautification parameter information of each identification face；Its
In, each face key position information includes at least one key position；
Beautify parameter according to the position and landscaping treatment is carried out to corresponding key position.
6. method as claimed in claim 5, which is characterized in that the basis each identifies the corresponding beautification parameter of face
Information determines that the corresponding position beautification parameter of key position includes：
The reference position coordinate of each identification face is determined according to the face key position information of each identification face；
Determine that parameter, and root are beautified in the corresponding position of key position according to the corresponding beautification parameter information of each identification face
Beautify parameter according to the corresponding position of key position described in the reference position Coordinate Adjusting.
7. such as claim 1 to 6 any one of them method, which is characterized in that described to determine pending figure by Face datection
Identification face as in, including：
Obtain pending image；
Face key position identification is carried out to the pending image by default convolutional neural networks, obtains the pending figure
Face key position information as in, and the identification in the pending image is determined according to the face key position information
Face.
8. the method for claim 7, which is characterized in that described by presetting convolutional neural networks to the pending figure
As progress face key position identification, the face key position information in the pending image is obtained, including：
Multilayer convolution algorithm is carried out to pending image, every layer of convolution algorithm corresponds to a feature square for extracting pending image
Battle array；
According to the eigenmatrix extracted successively, the high-level characteristic matrix of the face key position in pending image is gradually determined；
Face key position information is determined according to the high-level characteristic matrix of the face key position.
9. the method for claim 7, which is characterized in that described by presetting convolutional neural networks to the pending figure
As before carrying out face key position identification, further including：
It is trained according to human face data set pair convolutional neural networks, obtains default convolutional neural networks, the default convolution god
It is used to determine the face key position information of the pending image through network.
10. method as claimed in claim 9, which is characterized in that described to be carried out according to human face data set pair convolutional neural networks
Training, obtaining default convolutional neural networks includes：
Each neuron on the convolutional layer of convolutional neural networks is established with the local neuron in preceding layer convolutional layer and is connected
It connects, and corresponding weights is set for every layer of convolutional layer, so that there is convolutional neural networks extraction face key position information to correspond to
Visual signature local receptor field；
The facial image that human face data is concentrated is input to convolutional neural networks to be trained, and it is corresponding to update every layer of convolutional layer
Weights obtain default convolutional neural networks.
11. a kind of image landscaping treatment device, which is characterized in that including：
Face recognition module, for determining the identification face in pending image by Face datection；Wherein, the identification face
It is at least two, the identification face includes one and the corresponding main identification face of current focusing area, other identifications
Face is secondary identification face；
Parameter determination module, for determining the corresponding first beautification parameter information and described secondary of the main identification face respectively
Identify the corresponding second beautification parameter information of face, wherein described first, which beautifies the corresponding beautification degree of parameter information, is higher than institute
State the corresponding beautification degree of the second beautification parameter information；
Landscaping treatment module, for the figure according to the corresponding beautification parameter information of each identification face to each identification face
As region carries out corresponding landscaping treatment.
12. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The image landscaping treatment method as described in any one of claim 1-10 is realized when execution.
13. a kind of terminal device, which is characterized in that including memory, processor and storage are on a memory and can be in processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1-10 when executing the computer program
Any one of them image landscaping treatment method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810217900.XA CN108491780B (en) | 2018-03-16 | 2018-03-16 | Image beautification processing method and device, storage medium and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810217900.XA CN108491780B (en) | 2018-03-16 | 2018-03-16 | Image beautification processing method and device, storage medium and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108491780A true CN108491780A (en) | 2018-09-04 |
| CN108491780B CN108491780B (en) | 2021-05-04 |
Family
ID=63339578
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810217900.XA Active CN108491780B (en) | 2018-03-16 | 2018-03-16 | Image beautification processing method and device, storage medium and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108491780B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241930A (en) * | 2018-09-20 | 2019-01-18 | 北京字节跳动网络技术有限公司 | Method and apparatus for handling supercilium image |
| CN109255814A (en) * | 2018-09-20 | 2019-01-22 | 北京字节跳动网络技术有限公司 | Method and apparatus for handling image |
| CN110188711A (en) * | 2019-06-03 | 2019-08-30 | 北京字节跳动网络技术有限公司 | Method and apparatus for output information |
| CN112651956A (en) * | 2020-12-30 | 2021-04-13 | 深圳云天励飞技术股份有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150347822A1 (en) * | 2014-05-29 | 2015-12-03 | Beijing Kuangshi Technology Co., Ltd. | Facial Landmark Localization Using Coarse-to-Fine Cascaded Neural Networks |
| CN105915782A (en) * | 2016-03-29 | 2016-08-31 | 维沃移动通信有限公司 | Picture obtaining method based on face identification, and mobile terminal |
| CN106101542A (en) * | 2016-06-28 | 2016-11-09 | 广东欧珀移动通信有限公司 | A kind of image processing method and terminal |
| CN106354303A (en) * | 2016-08-23 | 2017-01-25 | 维沃移动通信有限公司 | Photographed method of mobile terminal and mobile terminal |
| CN107124548A (en) * | 2017-04-25 | 2017-09-01 | 深圳市金立通信设备有限公司 | A kind of photographic method and terminal |
| CN107578380A (en) * | 2017-08-07 | 2018-01-12 | 北京金山安全软件有限公司 | Image processing method and device, electronic equipment and storage medium |
| US20180032797A1 (en) * | 2016-07-29 | 2018-02-01 | Samsung Electronics Co., Ltd. | Apparatus and method for processing a beauty effect |
| CN107680128A (en) * | 2017-10-31 | 2018-02-09 | 广东欧珀移动通信有限公司 | Image processing method, device, electronic equipment and computer-readable recording medium |
| CN107800965A (en) * | 2017-10-31 | 2018-03-13 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and computer equipment |
-
2018
- 2018-03-16 CN CN201810217900.XA patent/CN108491780B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150347822A1 (en) * | 2014-05-29 | 2015-12-03 | Beijing Kuangshi Technology Co., Ltd. | Facial Landmark Localization Using Coarse-to-Fine Cascaded Neural Networks |
| CN105915782A (en) * | 2016-03-29 | 2016-08-31 | 维沃移动通信有限公司 | Picture obtaining method based on face identification, and mobile terminal |
| CN106101542A (en) * | 2016-06-28 | 2016-11-09 | 广东欧珀移动通信有限公司 | A kind of image processing method and terminal |
| US20180032797A1 (en) * | 2016-07-29 | 2018-02-01 | Samsung Electronics Co., Ltd. | Apparatus and method for processing a beauty effect |
| CN106354303A (en) * | 2016-08-23 | 2017-01-25 | 维沃移动通信有限公司 | Photographed method of mobile terminal and mobile terminal |
| CN107124548A (en) * | 2017-04-25 | 2017-09-01 | 深圳市金立通信设备有限公司 | A kind of photographic method and terminal |
| CN107578380A (en) * | 2017-08-07 | 2018-01-12 | 北京金山安全软件有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN107680128A (en) * | 2017-10-31 | 2018-02-09 | 广东欧珀移动通信有限公司 | Image processing method, device, electronic equipment and computer-readable recording medium |
| CN107800965A (en) * | 2017-10-31 | 2018-03-13 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and computer equipment |
Non-Patent Citations (3)
| Title |
|---|
| ER-YANG HUAN等: "Deep Convolutional Neural Networks for Classifying Body Constitu", 《COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE》 * |
| M.EGMONT-PETERSEN等: "Image processing with neural networks—a review", 《PATTERN RECOGNITION》 * |
| 风澈VIO: "后期对焦AfterFocus Pro v2.1.0独家汉化破解版", 《简书》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241930A (en) * | 2018-09-20 | 2019-01-18 | 北京字节跳动网络技术有限公司 | Method and apparatus for handling supercilium image |
| CN109255814A (en) * | 2018-09-20 | 2019-01-22 | 北京字节跳动网络技术有限公司 | Method and apparatus for handling image |
| CN109241930B (en) * | 2018-09-20 | 2021-03-02 | 北京字节跳动网络技术有限公司 | Method and apparatus for processing eyebrow image |
| CN110188711A (en) * | 2019-06-03 | 2019-08-30 | 北京字节跳动网络技术有限公司 | Method and apparatus for output information |
| CN112651956A (en) * | 2020-12-30 | 2021-04-13 | 深圳云天励飞技术股份有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN112651956B (en) * | 2020-12-30 | 2024-05-03 | 深圳云天励飞技术股份有限公司 | Image processing method, device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108491780B (en) | 2021-05-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110118B (en) | Dressing recommendation method and device, storage medium and mobile terminal | |
| CN107995428B (en) | Image processing method, image processing device, storage medium and mobile terminal | |
| EP3700190A1 (en) | Electronic device for providing shooting mode based on virtual character and operation method thereof | |
| CN107635095A (en) | Shoot method, apparatus, storage medium and the capture apparatus of photo | |
| CN108566516B (en) | Image processing method, device, storage medium and mobile terminal | |
| CN109348135A (en) | Photographic method, device, storage medium and terminal device | |
| WO2019120029A1 (en) | Intelligent screen brightness adjustment method and apparatus, and storage medium and mobile terminal | |
| CN108681402A (en) | Identify exchange method, device, storage medium and terminal device | |
| CN108646920A (en) | Identify exchange method, device, storage medium and terminal device | |
| CN108491780A (en) | Image landscaping treatment method, apparatus, storage medium and terminal device | |
| CN107909629A (en) | Recommendation method, apparatus, storage medium and the terminal device of paster | |
| CN109547701A (en) | Image capturing method, device, storage medium and electronic equipment | |
| CN109741280A (en) | Image processing method, device, storage medium and electronic equipment | |
| US20210027513A1 (en) | Electronic device for providing avatar and operating method thereof | |
| CN109741288A (en) | Image processing method, device, storage medium and electronic equipment | |
| CN108200337B (en) | Photographing processing method, device, terminal and storage medium | |
| CN108551552B (en) | Image processing method, device, storage medium and mobile terminal | |
| CN108734002A (en) | Intelligent configuration method, device, storage medium and the mobile terminal of system resource | |
| CN108494996B (en) | Image processing method, device, storage medium and mobile terminal | |
| CN108021905A (en) | image processing method, device, terminal device and storage medium | |
| CN106980840A (en) | Shape of face matching process, device and storage medium | |
| CN109784252A (en) | Image processing method, device, storage medium and electronic equipment | |
| CN109618098A (en) | A kind of portrait face method of adjustment, device, storage medium and terminal | |
| CN110827195B (en) | Virtual article adding method and device, electronic equipment and storage medium | |
| CN108494968A (en) | Intelligent prompt method, device, storage medium and intelligent terminal |
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 | ||
| CB02 | Change of applicant information |
Address after: Changan town in Guangdong province Dongguan 523860 usha Beach Road No. 18 Applicant after: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS Corp.,Ltd. Address before: Changan town in Guangdong province Dongguan 523860 usha Beach Road No. 18 Applicant before: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS Corp.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |