CN108492348A - Image processing method, image processing device, electronic equipment and storage medium - Google Patents
Image processing method, image processing device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN108492348A CN108492348A CN201810277457.5A CN201810277457A CN108492348A CN 108492348 A CN108492348 A CN 108492348A CN 201810277457 A CN201810277457 A CN 201810277457A CN 108492348 A CN108492348 A CN 108492348A
- Authority
- CN
- China
- Prior art keywords
- area
- layer
- color
- image
- initial pictures
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/40—Filling a planar surface by adding surface attributes, e.g. colour or texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/001—Texturing; Colouring; Generation of texture or colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Abstract
The embodiment of the invention provides an image processing method, an image processing device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring an initial image; generating a layer to be processed containing a first area through the initial image, wherein the first area is the area to be processed in the initial image; determining a color to be filled in the first area; at least coloring a first area in the layer to be processed according to the color to be filled and the coloring rule to obtain a colored layer; and fusing the colored image layer and the initial image to generate a processed image. In the embodiment of the invention, the layer to be processed including the first region in the initial image can be generated firstly, then the layer to be processed is colored through the color to be filled to obtain the colored layer, and the colored layer is fused with the initial image, so that the purpose of adjusting the color of the initial image is achieved.
Description
Technical field
The present invention relates to video technique fields, more particularly to a kind of image processing method, device, electronic equipment and storage
Medium.
Background technology
With the development of image acquisition technology, more and more images can be got by image capture device.However,
After getting image, the color of part or whole region may not be what user wanted in image;Alternatively, in image part or
The color effects of whole region are bad to cause whole image quality poor.Such as, when image includes personage, the color development of personage,
Clothes color etc. may be unsatisfactory for the expectation or ineffective of user.In this case, it needs to some or all of in image
Region carries out color adjustment.
Therefore, how to carry out color adjustment to image becomes a urgent problem to be solved.
Invention content
The embodiment of the present invention is designed to provide a kind of image processing method, device, electronic equipment and storage medium, with
The technical issues of solving how to carry out color adjustment to image.
To achieve the goals above, in a first aspect, an embodiment of the present invention provides a kind of image processing method, the method
Including:
Obtain initial pictures;
By the initial pictures, the pending figure layer for including first area is generated, the first area is described initial
Pending area in image;
Determine the color to be filled of the first area;And according to the color to be filled and coloring rule, at least to institute
The first area stated in pending figure layer carries out coloring treatment, obtains colored figure layer;
The colored figure layer is merged with the initial pictures, generates processed image.
Optionally, the color to be filled of the determination first area, including:
According to position of the first area in the initial pictures, the color to be filled of the first area is determined;
Or
According to the color for presetting second area in the initial pictures, the color to be filled of the first area is determined;Or
According to presetting second area in position of the first area in the initial pictures and the initial pictures
Color, determine the color to be filled of the first area.
Optionally, described according to the color to be filled and coloring rule, at least to first in the pending figure layer
Region carries out coloring treatment, obtains colored figure layer, including:
According to the color to be filled, at least the first area in the pending figure layer is carried out at gradual change type coloring
Reason, obtains colored figure layer.
Optionally, described according to the color to be filled, at least the first area in the pending figure layer is carried out gradually
Variant coloring treatment obtains colored figure layer, including:
Obtain the corresponding color value of the color to be filled;
At least for each pixel of first area in the pending figure layer, existed according to the pixel position
The coordinate value of preset coordinate axis, the color value and preset transformational relation determine the corresponding color of object of the pixel
Value, and the color value of the pixel is revised as the target color values.
Optionally, described to generate the pending figure layer for including first area by the initial pictures, including:
By neural network image semantic segmentation model, at least one first area in the initial pictures is identified, and raw
At the target image for including the first area;
According to preset erosion ratio, edge erosion processing is carried out to the first area in the target image;
Edge emergence processing is carried out to the first area in the target image, and generates after processing that image is corresponding waits locating
Manage figure layer.
Optionally, before the acquisition initial pictures, the method further includes:
Obtain sample image, wherein the sample image includes at least one marked region;
Preset neural network image semantic segmentation model is trained using the sample image, obtains meeting default
The neural network image semantic segmentation model of condition.
Optionally, described to merge the colored figure layer with the initial pictures, generate processed image, packet
It includes:
The initial pictures are converted into the first tone saturation degree lightness HSV images, the colored figure layer is converted to
2nd HSV images;
By the tone H and saturation degree S components of each pixel in first area described in the first HSV images, replace with described
H the and S components of each pixel in first area corresponding position, obtain processed image described in 2nd HSV images.
Second aspect, an embodiment of the present invention provides a kind of image processing apparatus, described device includes:
Image collection module, for obtaining initial pictures;
Figure layer generation module, for by the initial pictures, generating the pending figure layer for including first area, described the
One region is the pending area in the initial pictures;
Figure layer process module, the color to be filled for determining the first area;And according to the color to be filled and
Coloring rule, at least carries out coloring treatment to the first area in the pending figure layer, obtains colored figure layer;
Image co-registration module generates processed figure for merging the colored figure layer with the initial pictures
Picture.
Optionally, the figure layer process module, is specifically used for:
According to position of the first area in the initial pictures, the color to be filled of the first area is determined;
Or
According to the color for presetting second area in the initial pictures, the color to be filled of the first area is determined;Or
According to presetting second area in position of the first area in the initial pictures and the initial pictures
Color, determine the color to be filled of the first area.
Optionally, the figure layer process module, is specifically used for:According to the color to be filled, at least to described pending
First area in figure layer carries out gradual change type coloring treatment, obtains colored figure layer.
Optionally, the figure layer process module, including:
Color value acquisition submodule, for obtaining the corresponding color value of preset color of object;
Color value determination sub-module, for each pixel at least for pending area in the figure layer, according to institute
Pixel position is stated in the coordinate value, the color value and preset transformational relation of preset direction, determines the pixel
The corresponding target color values of point, and the color value of the pixel is revised as the target color values.
Optionally, the figure layer generation module, including:
Submodule is identified, for by neural network image semantic segmentation model, identifying at least one in the initial pictures
A first area, and generate the target image for including the first area;
Submodule is corroded, for according to preset erosion ratio, edge to be carried out to the first area in the target image
Erosion is handled;
Emergence submodule for carrying out edge emergence processing to the first area in the target image, and generates processing
The corresponding pending figure layer of image afterwards.
Optionally, described device further includes:
Sample acquisition module, for obtaining sample image, wherein the sample image includes at least one marked region;
Model training module, for being carried out to preset neural network image semantic segmentation model using the sample image
Training, obtains the neural network image semantic segmentation model for meeting preset condition.
Optionally, described image Fusion Module, including:
Image transform subblock, for the initial pictures to be converted to the first tone saturation degree lightness HSV images, by institute
It states colored figure layer and is converted to the 2nd HSV images;
Component replaces submodule, is used for the tone H and saturation of each pixel in first area described in the first HSV images
S components are spent, H the and S components of each pixel in first area corresponding position described in the 2nd HSV images is replaced with, obtains
Handle image.
The third aspect, an embodiment of the present invention provides a kind of electronic equipment, including processor, communication interface, memory and
Communication bus, wherein the processor, the communication interface and the memory are completed each other by the communication bus
Communication;
The memory, for storing computer program;
The processor when for executing the program stored on memory, realizes the side as described in above-mentioned first aspect
Method step.
Fourth aspect, an embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage
Dielectric memory contains computer program, and the side as described in above-mentioned first aspect is realized when the computer program is executed by processor
Method step.
5th aspect, an embodiment of the present invention provides a kind of computer program product, the computer program product is being counted
When being run on calculation machine, the method and step as described in above-mentioned first aspect is realized.
An embodiment of the present invention provides a kind of image processing method, device, electronic equipment and storage medium, the method packets
It includes:Obtain initial pictures;By the initial pictures, the pending figure layer for including first area is generated, the first area is
Pending area in the initial pictures;Determine the color to be filled of the first area;And according to the color to be filled
With coloring rule, coloring treatment at least is carried out to the first area in the pending figure layer, obtains colored figure layer;It will be described
Colored figure layer is merged with the initial pictures, generates processed image.
In the embodiment of the present invention, the pending figure layer for including first area in initial pictures, Jin Ertong can be firstly generated
Color to be filled is crossed, carrying out coloring treatment to pending figure layer obtains colored figure layer, and by colored figure layer and initial pictures
It is merged, achievees the purpose that carry out color adjustment to initial pictures.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
It obtains it is clear that being emerged from by implementing the present invention.The purpose of the present invention and other advantages can by specification,
Specifically noted structure is realized and is obtained in claims and attached drawing.
Certainly, implement any of the products of the present invention or method it is not absolutely required at the same reach all the above excellent
Point.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Obtain other attached drawings according to these attached drawings.
Fig. 1 is a kind of image processing method flow chart of the embodiment of the present invention;
Fig. 2 is the effect diagram that figure layer is colored as to fused image when same color value of the embodiment of the present invention;
Fig. 3 is to carry out block partition effect diagram to figure layer;
Fig. 4 is a kind of another flow chart of image processing method of the embodiment of the present invention;
Fig. 5 is the image processing effect schematic diagram of the embodiment of the present invention;
Fig. 6 is the target image schematic diagram of the embodiment of the present invention generated according to initial pictures;
Fig. 7 is the effect diagram to target image progress edge erosion and after sprouting wings of the embodiment of the present invention;
Fig. 8 is a kind of image processing apparatus structural schematic diagram of the embodiment of the present invention;
Fig. 9 is a kind of electronic equipment structural schematic diagram of the embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In the embodiment of the present invention, color adjustment can be carried out to image, to meet the expectation of user or improve the matter of image
Amount.For example, color adjustment can be carried out to the different zones of image, it such as can be to the hair zones of personage, face area in image
Domain, eye areas, lip region, clothes region etc. carry out color adjustment.
It can be appreciated that being directed to different regions, same or analogous method may be used, color adjustment is carried out to it.For
Convenient for description, the embodiment of the present invention to the hair zones of image for carrying out color adjustment, illustrating the embodiment of the present invention
Image processing method, when being handled for other regions, the method similar with the embodiment of the present invention may be used, the present invention
Embodiment is to this without repeating.
As shown in Figure 1, an embodiment of the present invention provides a kind of image processing method, this method comprises the following steps:
S101 obtains initial pictures;
Method provided in an embodiment of the present invention can be applied to electronic equipment.For example, can be arbitrary specific image procossing
The equipment of function, such as desktop computer, portable computer, intelligent mobile terminal.
In embodiments of the present invention, electronic equipment can obtain initial pictures namely pending image.For example, electronics
Equipment can receive the image of external equipment input, as initial pictures;Alternatively, electronic equipment can show the figure being locally stored
Picture, and according to selection instruction input by user, the selected image of user is determined as initial pictures, the embodiment of the present invention is to this
It does not limit.
S102 generates the pending figure layer for including first area, the first area is described by the initial pictures
Pending area in initial pictures;
In embodiments of the present invention, electronic equipment can be adjusted the personage's hair color occurred in initial pictures.
It is appreciated that the personage's hair occurred in initial pictures usually only occupies the subregion of initial pictures, in the embodiment of the present invention,
Region where personage's hair can be known as pending area.
Personage's hair color in initial pictures is adjusted, namely color adjustment is carried out to pending area.In this hair
In bright embodiment, in order to be accurately adjusted to the personage's hair color occurred in initial pictures, electronic equipment is got just
After beginning image, the pending figure layer for including first area can be generated, wherein above-mentioned first area is first by initial pictures
Pending area in beginning image.
Figure layer is like to be stacked together, combine in order a sheet by a sheet containing the film of the elements such as word or figure
Form the final effect of the page.Wherein, each figure layer is made of many pixels, and figure layer further through being superimposed up and down
Mode forms whole image.A metaphor is beaten, as soon as it is transparent " glass " that each figure layer, which just seems, and figure layer content is drawn
On these " glass ", if " glass " there is nothing, here it is a fully transparent empty graph layers, when each " glass " has figure
When picture, All Layers are overlooked from top to bottom, to form the final display effect of image.
Features described above based on figure layer it is found that figure layer handled compared to image carry out processing want much easier,
The case where especially for only handling subregion.Therefore, in the embodiment of the present invention, after getting initial pictures, electronics
Equipment can generate the pending figure layer comprising first area.For example, DPM (deformable may be used in electronic equipment
Parts model, deformable member model) etc. target detections class algorithm, to detect the hair zones in initial pictures, as
One region, and determine the coordinate information of first area, such as determine the coordinate value of each boundary position in first area;And then according to first
The coordinate information in region generates the pending figure layer for including first area.
S103 determines the color to be filled of the first area;And according to the color to be filled and coloring rule, at least
Coloring treatment is carried out to the first area in the pending figure layer, obtains colored figure layer;
After obtaining the pending figure layer comprising first area, electronic equipment can to the first area in pending figure layer into
Row color adjusts, to achieve the purpose that carry out color adjustment to first area in initial pictures.
Specifically, electronic equipment can determine the color to be filled of first area, preset color can be such as obtained,
As color to be filled.And then according to color to be filled and coloring rule, at least the first area in pending figure layer is carried out
Coloring treatment obtains colored figure layer.
That is, when being coloured to pending figure layer, only the first area of pending figure layer can be coloured,
The whole region of pending figure layer can also be coloured, this is all reasonable.
Wherein, above-mentioned color to be filled can be one or more colors, this is all reasonable.Above-mentioned coloring rule can be with
For each pixel value of the painted areas (first area or whole region) of pending figure layer is colored as identical color;
Alternatively, can multiple subregions be divided into the painted areas of pending figure layer, different subregions is colored as to different face
Color, the embodiment of the present invention are not construed as limiting this.
S104 merges the colored figure layer with the initial pictures, generates processed image.
Pending figure layer is carried out after coloring treatment obtains colored figure layer, electronic equipment can by colored figure layer with just
Beginning image is merged, and processed image is obtained, and completes to adjust the color of initial pictures.
For example, the first area of initial pictures can be replaced with the first area of colored figure layer by electronic equipment, obtain
Processed image.
In the embodiment of the present invention, the pending figure layer for including pending area in initial pictures can be firstly generated, in turn
By color to be filled, coloring treatment is carried out to pending figure layer and obtains colored figure layer, and by colored figure layer and initial graph
As being merged, achieve the purpose that carry out color adjustment to initial pictures.
As a kind of embodiment of the embodiment of the present invention, electronic equipment in the color to be filled for determining first area,
Different modes may be used to determine color to be filled.
In one implementation, electronic equipment can determine first according to position of the first area in initial pictures
The color to be filled in region can carry out color adjustment to realize to different zones.
Such as, electronic equipment can prestore the corresponding color to be filled of different zones, can such as store hair zones pair
Color to be filled, eye areas corresponding color, the corresponding color to be filled in face region and the lip region to be filled answered
Corresponding color to be filled etc..It, can be according to first area in initial pictures in the color to be filled for determining first area
Position namely first area be hair zones, eye areas, face region or lip region, to determine waiting for for first area
Fill Color.
In another implementation, electronic equipment can be determined according to the color for presetting second area in initial pictures
The color to be filled of first area.Such as, above-mentioned second area can be face region, that is to say, that can be directed to different skins
Color determines different colors to be filled, according to personage's feature, to realize and be adjusted to the personalization of image, enhancing image procossing effect
Fruit.
Such as, when carrying out color adjustment to hair zones, electronic equipment can prestore that the different colours of skin are corresponding to be waited filling out
Fill color., can be according to the color in face region in initial pictures in the color to be filled for determining first area, determination waits filling out
Fill color.
In another implementation, electronic equipment can be according to position of the first area in initial pictures, Yi Jichu
The color that second area is preset in beginning image, determines the color to be filled of first area, can root to different zones with realization
Color to be filled is determined according to the colour of skin, improves the applicability of color of image adjustment.
When carrying out color adjustment to personage's hair zones, if the first area in pending figure layer is colored as same face
Color value, i.e., the color value all same of each pixel in first area, will lead to final processing result image in pending figure layer
There are the feeling that shade differs, image effect bad.
As shown in Fig. 2, after generating figure layer 220 according to initial pictures 210, figure layer 220 is colored as same color value, most
Eventually in the image 230 after fusion, color development is too single, becomes unnatural.
As a kind of embodiment of the embodiment of the present invention, in embodiments of the present invention, in order to ensure image after handling
Effect, when being coloured to pending figure layer, can according to color to be filled, at least to the first area in pending figure layer into
Row gradual change type coloring treatment.Also can be different colors by the different pixels Point Coloring in first area in pending figure layer
Value.
When carrying out gradual change type coloring treatment to first area, color change trend can be by by force to the change in a weak direction
Change trend;Can be either it is first weak after it is strong weak again or first strong after weak a variety of variation tendencies strong again;Or for lip region,
Eye areas, or central area is strong, and the weak variation tendency of surrounding, the embodiment of the present invention is not construed as limiting this.
For example, when carrying out color adjustment to hair zones, color change trend is to be become by the variation by force to a weak direction
When gesture, first area in pending figure layer can be divided into multiple blocks by electronic equipment along preset direction, and then by each area
Block is colored as different color values.When determining the color value of different blocks, in one implementation, electronic equipment can be to each
Block is numbered successively, and then the color value after either block is coloured is determined as:Block number * preset ratio values * is waited for
Fill Color value.
As shown in figure 3, longitudinal 310 equal proportion of total length in pending figure layer can be divided into default number (such as 20 parts,
50 parts, 100 parts etc.), wherein every portion location block is 320 (being not entirely shown in figure);Then it is directed to each block successively,
It is colored as the certain proportion value of color value to be filled.
As shown in figure 4, according to color to be filled, at least the first area in pending figure layer is carried out at gradual change type coloring
Reason, obtains the process of colored figure layer, may comprise steps of:
S401 obtains the corresponding color value of the color to be filled;
In embodiments of the present invention, when being coloured to the first area of pending figure layer, electronic equipment can obtain first
Take the corresponding color value of color to be filled.Wherein, above-mentioned color to be filled can be at least two, with realize to first area into
Row gradual change type colours.Correspondingly, the corresponding color value of color to be filled is also at least two.
For example, electronic equipment can prestore the correspondence of each color and color value, in turn, when determining face to be filled
After color, the corresponding color value of color to be filled can be found in the correspondence of pre-stored each color and color value.
S402, at least for each pixel of first area in the pending figure layer, according to where the pixel
Position determines the corresponding mesh of the pixel in the coordinate value, the color value and preset transformational relation of preset coordinate axis
Color value is marked, and the color value of the pixel is revised as the target color values.
After determining the corresponding color value of color to be filled, electronic equipment can carry out face to the first area of pending figure layer
Color adjustment.For example, when determining color value includes two kinds of color values, electronic equipment can determine in pending figure layer the first
Length value of one region in preset coordinate axis direction.In turn, the corresponding target of any pixel point can be calculated by following formula
Color value C (y):
Wherein, y is coordinate value of the pixel in preset coordinate axis, and C (y) is the corresponding target color values of the pixel, L
It is first area in pending figure layer in the length value of preset coordinate axis direction, C0And C1For identified two kinds of color values.
When determining color value includes multiple color value, when such as n, n >=2 can be by the first area of pending figure layer
It is divided into n-1 block along preset coordinate axis direction, and by following formula, determines the corresponding mesh of any pixel point in each block
Mark color value:
Wherein, yiIt is the pixel in the coordinate value of preset coordinate axis, Ci(yi) it is the corresponding color of object of the pixel
Value, LiFor i-th of block preset coordinate axis direction length value,It is i-th of block initial position in preset coordinate axis
Coordinate value, CiFor i-th kind of color value, Ci+1Color value is planted for (i+1).
For example, coordinate system can be established in pending figure layer, it such as can be to be established along vertical in pending figure layer shown in Fig. 3
To, origin is in first area least significant end or the y-axis coordinate system of top.It may thereby determine that the seat of each pixel in first area
Scale value y calculates the corresponding target color values of each pixel according further to above-mentioned formula, and the color value of each pixel is changed
For corresponding target color values.
As shown in figure 5, being coloured using the pending figure layer corresponding to initial pictures 510 of the method described in the present embodiment
The colored figure layer obtained afterwards as depicted 520, the processed figure obtained after colored figure layer 520 is merged with initial pictures 510
As being image 530.As shown in Figure 5, in the image 530 obtained after fusion, personage's hair color compare naturally, image effect compared with
It is excellent.
In the present embodiment, each pixel can be clicked through according to the coordinate value of each pixel in first area in pending figure layer
Row coloring treatment, so that evenly, fade effect is apparent for each pixel color value in first area in processed figure layer,
And then cause the effect of fused image more excellent.
ENet is a quick neural network model, efficient can be split to picture.ENet belongs to
Encoder-decoder frameworks, main effect are the feature extraction to each classification and classify to pixel.Also, make
Feature extraction and pixel classifications can be accurately carried out with ENet.
As a kind of embodiment of the embodiment of the present invention, the process that electronic equipment generates pending figure layer may include with
Lower step:
A identifies at least one first area in initial pictures, and generate by neural network image semantic segmentation model
Include the target image of first area;
It, can be by neural network image semantic segmentation model, if ENet is to first after electronic equipment gets initial pictures
Beginning image carries out feature extraction, identifies at least one first area in initial pictures, and generate the target figure for including first area
Picture.
Such as, the sample training ENet for having marked personage's hair zones can be first passed through in advance, it when performing image processing, can be with
The target image for including first area in initial pictures is generated using trained ENet.
As shown in fig. 6, image 610 is the initial pictures of input, this image does not need additional processing.Through ENet's
Segmentation will produce and export target image 620.
Wherein, the size of target image 620 is consistent with initial pictures 610.Each pixel is all marked as one in image 620
A classification.620 black picture element of image in Fig. 6 is background, white portion is hair.
B carries out edge erosion processing according to preset erosion ratio to the first area in target image;
C, in target image first area carry out edge emergence processing, and generate processing after image it is corresponding pending
Figure layer.
It is appreciated that in the embodiment of the present invention, color adjustment only is carried out to the first area in initial pictures, other areas
Domain is not adjusted, and this may cause in the processed image ultimately generated, the marginal position of first area and other regions
There is color saltus step in place, bad so as to cause image effect.
Emergence is processing picture tool important in the image processing tools such as ps.Emergence principle enables the inside and outside linking in constituency
Part blurs, and plays the role of gradual change to achieve the effect that natural sparse model.It therefore, can be to target image in the present embodiment
First area carry out emergence processing, to which obtained processed image not will produce first area marginal portion color after fusion
Saltus step ensure that image effect.
However, when the first area to target image carries out edge emergence, the range that may result in first area exceeds
Original boundary.It therefore, can be according to preset erosion ratio before carrying out edge emergence to the first area of target image
(such as 5%, 6%, 8%) carries out edge erosion processing to the first area in target image.Also it can reduce target image
The size of middle first area.
As shown in fig. 7,710 be target image, obtain image 720 after edge erosion processing is carried out to it, to image 720 into
Image 730 is obtained after the emergence processing of row edge.As can be seen that the size of first area and the firstth area in image 710 in image 730
The size in domain is almost the same.
After carrying out edge emergence processing to target image, electronic equipment can generate the corresponding pending figure of image after processing
Layer.In one implementation, can include two figure layers, one of figure layer packet in the pending figure layer that electronic equipment generates
Containing first area, another figure layer includes other regions, and final display effect is the image for including first area and other regions.
Also, in order to be distinguished to first area and other regions, the color in first area and other regions can be different.
In another implementation, electronic equipment can only generate a figure layer, and shape, size are and first area
Identical, final display effect as only includes the image of first area.
In the present embodiment, edge erosion processing first is carried out to first area in target image, then obtained after being sprouted wings
First area size it is almost the same with the size of first area in initial pictures, so as to avoid first area beyond original
The problem of boundary.
As a kind of embodiment of the embodiment of the present invention, electronic equipment can train to obtain nerve net by following steps
Network image, semantic parted pattern:
D obtains sample image, wherein sample image includes at least one marked region;
In the present embodiment, electronic equipment can obtain sample image, and neural network image is trained to use sample image
Semantic segmentation model.Wherein, include at least one marked region in every sample image that electronic equipment obtains.Wherein, on
It can be one or more in hair zones, face region, eye areas, lip region to state marked region for example.
F is trained preset neural network image semantic segmentation model using sample image, obtains meeting default item
The neural network image semantic segmentation model of part.
After getting sample image, electronic equipment can utilize sample image to preset neural network image semantic segmentation
Model is trained, and known any method such as may be used and instructed to preset neural network image semantic segmentation model
Practice, it is not limited in the embodiment of the present invention.
Above-mentioned preset condition for example can be:The neural network image semanteme point that sample image input training in part is obtained
After cutting model, the similarity between the marked region marked in the first area of each image of output, and corresponding sample image is big
In predetermined threshold value, such as 80%, 85%, 90%.
In the present embodiment, can train to obtain neural network image semantic segmentation model by sample image, into
When row color of image adjusts, the first area in initial pictures can be accurately identified by the model that training obtains, in turn
Color adjustment accurately is carried out to first area.
As a kind of embodiment of the embodiment of the present invention, electronic equipment carries out colored figure layer and the initial pictures
Fusion, generates the process of processed image, may comprise steps of:
Initial pictures are converted to the first tone saturation degree lightness HSV images by step 1, and colored figure layer is converted to
Two HSV images;
In the present embodiment, initial pictures can be converted to the first HSV (Hue, Saturation, Value, tone saturation
Spend lightness) image, colored figure layer is converted into the 2nd HSV images.
For example, image conversion techniques may be used, initial pictures and colored figure layer are converted to the domains HSV respectively, are obtained
The corresponding first HSV images of initial pictures and the corresponding 2nd HSV images of colored figure layer.
Step 2 replaces with H the and S components of each pixel in first area in the first HSV images in the 2nd HSV images
H the and S components of the one each pixel in region corresponding position, the image that obtains that treated.
In order to keep the lightness feature of initial pictures, in the present embodiment, colored figure layer and initial pictures are melted
When conjunction, the firstth area in the 2nd HSV images can be replaced with only by H the and S components of each pixel in first area in the first HSV images
H the and S components of each pixel in domain corresponding position keep the V component of first area in initial pictures constant, obtain processed figure
Picture.
It, can be by converting initial pictures and colored figure layer to the domains HSV in the present embodiment, and then only replace initial graph
H the and S components of each pixel in first area as in can retain just while carrying out color adjustment to initial pictures in this way
The lightness feature of beginning image.
Corresponding to above method embodiment, the embodiment of the invention also discloses a kind of image processing apparatus, as shown in figure 8,
The device includes:
Image collection module 810, for obtaining initial pictures;
Figure layer generation module 820, for by the initial pictures, generating the pending figure layer for including first area, institute
It is the pending area in the initial pictures to state first area;
Figure layer process module 830, the color to be filled for determining the first area;And according to the color to be filled
With coloring rule, coloring treatment at least is carried out to the first area in the pending figure layer, obtains colored figure layer;
Image co-registration module 840 generates processed for merging the colored figure layer with the initial pictures
Image.
In the embodiment of the present invention, the pending figure layer for including first area in initial pictures, Jin Ertong can be firstly generated
Color to be filled is crossed, carrying out coloring treatment to pending figure layer obtains colored figure layer, and by colored figure layer and initial pictures
It is merged, achievees the purpose that carry out color adjustment to initial pictures.
As a kind of embodiment of the embodiment of the present invention, the figure layer process module 830 is specifically used for:
According to position of the first area in the initial pictures, the color to be filled of the first area is determined;
Or
According to the color for presetting second area in the initial pictures, the color to be filled of the first area is determined;Or
According to presetting second area in position of the first area in the initial pictures and the initial pictures
Color, determine the color to be filled of the first area.
As a kind of embodiment of the embodiment of the present invention, the figure layer process module 830 is specifically used for:According to described
Color to be filled at least carries out gradual change type coloring treatment to the first area in the pending figure layer, obtains colored figure layer.
As a kind of embodiment of the embodiment of the present invention, the figure layer process module 830, including:
Color value acquisition submodule, for obtaining the corresponding color value of preset color of object;
Color value determination sub-module, for each pixel at least for pending area in the figure layer, according to institute
Pixel position is stated in the coordinate value, the color value and preset transformational relation of preset direction, determines the pixel
The corresponding target color values of point, and the color value of the pixel is revised as the target color values.
As a kind of embodiment of the embodiment of the present invention, the figure layer generation module 820, including:
Submodule is identified, for by neural network image semantic segmentation model, identifying at least one in the initial pictures
A first area, and generate the target image for including the first area;
Submodule is corroded, for according to preset erosion ratio, edge to be carried out to the first area in the target image
Erosion is handled;
Emergence submodule for carrying out edge emergence processing to the first area in the target image, and generates processing
The corresponding pending figure layer of image afterwards.
As a kind of embodiment of the embodiment of the present invention, described device further includes:
Sample acquisition module, for obtaining sample image, wherein the sample image includes at least one marked region;
Model training module, for being carried out to preset neural network image semantic segmentation model using the sample image
Training, obtains the neural network image semantic segmentation model for meeting preset condition.
As a kind of embodiment of the embodiment of the present invention, described image Fusion Module 840, including:
Image transform subblock, for the initial pictures to be converted to the first tone saturation degree lightness HSV images, by institute
It states colored figure layer and is converted to the 2nd HSV images;
Component replaces submodule, is used for the tone H and saturation of each pixel in first area described in the first HSV images
S components are spent, H the and S components of each pixel in first area corresponding position described in the 2nd HSV images is replaced with, obtains
Handle image.
Based on technical concept identical with embodiment of the method, the embodiment of the present invention additionally provides a kind of electronic equipment, such as Fig. 9
It is shown, including processor 91, communication interface 92, memory 93 and communication bus 94, wherein processor 91, is deposited at communication interface 92
Reservoir 93 completes mutual communication by communication bus 94,
Memory 93, for storing computer program;
Processor 91 when for executing the program stored on memory 93, realizes one described in above method embodiment
Kind image processing method.
The communication bus 94 that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral
Component Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..The communication bus 94 can be divided into address bus, data/address bus, controlling bus etc..For
Convenient for indicating, only indicated with a thick line in Fig. 9, it is not intended that an only bus or a type of bus.
Above-mentioned communication interface 92 is for the communication between above-mentioned electronic equipment and other equipment.
Above-mentioned memory 93 may include random access memory (Random Access Memory, RAM), can also wrap
Include nonvolatile memory (non-volatile memory, NVM), for example, at least a magnetic disk storage.Optionally, it stores
Device can also be at least one storage device for being located remotely from aforementioned processor.
Above-mentioned processor 91 can be general processor, including central processing unit (Central Processing
Unit, abbreviation CPU), network processing unit (Ne twork Processor, NP) etc.;It can also be digital signal processor
(Digital Signal Processing, DSP), application-specific integrated circuit (Applica tion Specific Integrated
Circuit, ASIC), field programmable gate array (Field-Programmable Gate Array, FPGA) or other can
Programmed logic device, discrete gate or transistor logic, discrete hardware components.
Above-mentioned electronic equipment includes but not limited to smart mobile phone, computer, personal digital assistant etc..
In the embodiment of the present invention, the pending figure layer for including first area in initial pictures, Jin Ertong can be firstly generated
Color to be filled is crossed, carrying out coloring treatment to pending figure layer obtains colored figure layer, and by colored figure layer and initial pictures
It is merged, achievees the purpose that carry out color adjustment to initial pictures.
Based on technical concept identical with embodiment of the method, the embodiment of the present invention additionally provides a kind of computer-readable storage
Medium.It is stored with computer program in the computer readable storage medium, is realized when computer program is executed by processor above-mentioned
A kind of image processing method in embodiment of the method.
Above computer readable storage medium storing program for executing can include but is not limited to random access memory (RAM), dynamic random is deposited
Access to memory (DRAM), static RAM (SRAM), read-only memory (ROM), programmable read only memory
(PROM), Erarable Programmable Read only Memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory (example
Such as, NOR type flash memory or NAND-type flash memory), Content Addressable Memory (CAM), polymer memory is (for example, ferroelectric polymers
Memory), phase transition storage, ovonic memory, silicon-oxide-nitride silicon-silica-silicon (Silicon-
Oxide-Nitride-Oxide-Silicon, SONOS) memory, magnetic card or light-card, also or any other appropriate type
Computer readable storage medium.
In the embodiment of the present invention, the pending figure layer for including first area in initial pictures, Jin Ertong can be firstly generated
Color to be filled is crossed, carrying out coloring treatment to pending figure layer obtains colored figure layer, and by colored figure layer and initial pictures
It is merged, achievees the purpose that carry out color adjustment to initial pictures.
Based on technical concept identical with embodiment of the method, the embodiment of the present invention additionally provides a kind of computer program production
Product when the computer program product is run on computers, realize a kind of image processing method in above method embodiment.
In the embodiment of the present invention, the pending figure layer for including first area in initial pictures, Jin Ertong can be firstly generated
Color to be filled is crossed, carrying out coloring treatment to pending figure layer obtains colored figure layer, and by colored figure layer and initial pictures
It is merged, achievees the purpose that carry out color adjustment to initial pictures.
It should be noted that herein, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also include other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is all made of relevant mode and describes, the highlights of each of the examples are with
The difference of other embodiment, the same or similar parts between the embodiments can be referred to each other.Especially for device,
For electronic equipment, storage medium, since it is substantially similar to the method embodiment, so description is fairly simple, related place
Illustrate referring to the part of embodiment of the method.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (10)
1. a kind of image processing method, which is characterized in that the method includes:
Obtain initial pictures;
By the initial pictures, the pending figure layer for including first area is generated, the first area is the initial pictures
In pending area;
Determine the color to be filled of the first area;And it according to the color to be filled and coloring rule, at least waits for described
The first area handled in figure layer carries out coloring treatment, obtains colored figure layer;
The colored figure layer is merged with the initial pictures, generates processed image.
2. according to the method described in claim 1, it is characterized in that, the color to be filled of the determination first area, packet
It includes:
According to position of the first area in the initial pictures, the color to be filled of the first area is determined;Or
According to the color for presetting second area in the initial pictures, the color to be filled of the first area is determined;Or
According to the face for presetting second area in position of the first area in the initial pictures and the initial pictures
Color determines the color to be filled of the first area.
3. according to the method described in claim 1, it is characterized in that, it is described according to the color to be filled and coloring rule, until
Few first area in the pending figure layer carries out coloring treatment, obtains colored figure layer, including:
According to the color to be filled, gradual change type coloring treatment at least is carried out to the first area in the pending figure layer, is obtained
To colored figure layer.
4. according to the method described in claim 3, it is characterized in that, described according to the color to be filled, at least wait for described
The first area handled in figure layer carries out gradual change type coloring treatment, obtains colored figure layer, including:
Obtain the corresponding color value of the color to be filled;
At least for each pixel of first area in the pending figure layer, according to the pixel position default
The coordinate value of reference axis, the color value and preset transformational relation determine the corresponding target color values of the pixel,
And the color value of the pixel is revised as the target color values.
5. according to claim 1-4 any one of them methods, which is characterized in that it is described by the initial pictures, generate packet
Pending figure layer containing first area, including:
By neural network image semantic segmentation model, at least one first area in the initial pictures is identified, and generate packet
Target image containing the first area;
According to preset erosion ratio, edge erosion processing is carried out to the first area in the target image;
Edge emergence processing is carried out to the first area in the target image, and generates the corresponding pending figure of image after processing
Layer.
6. according to the method described in claim 5, it is characterized in that, before the acquisition initial pictures, the method is also wrapped
It includes:
Obtain sample image, wherein the sample image includes at least one marked region;
Preset neural network image semantic segmentation model is trained using the sample image, obtains meeting preset condition
The neural network image semantic segmentation model.
7. according to claim 1-4 any one of them methods, which is characterized in that described by the colored figure layer and described first
Beginning image is merged, and processed image is generated, including:
The initial pictures are converted into the first tone saturation degree lightness HSV images, the colored figure layer is converted to second
HSV images;
By the tone H and saturation degree S components of each pixel in first area described in the first HSV images, described second is replaced with
H the and S components of each pixel in first area corresponding position described in HSV images, obtain processed image.
8. a kind of image processing apparatus, which is characterized in that described device includes:
Image collection module, for obtaining initial pictures;
Figure layer generation module, for by the initial pictures, generating the pending figure layer for including first area, firstth area
Domain is the pending area in the initial pictures;
Figure layer process module, the color to be filled for determining the first area;And according to the color to be filled and coloring
Rule at least carries out coloring treatment to the first area in the pending figure layer, obtains colored figure layer;
Image co-registration module generates processed image for merging the colored figure layer with the initial pictures.
9. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein described
Processor, the communication interface and the memory complete mutual communication by the communication bus;
The memory, for storing computer program;
The processor when for executing the program stored on memory, realizes any method steps of claim 1-7
Suddenly.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium
Program realizes claim 1-7 any method and steps when the computer program is executed by processor.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810277457.5A CN108492348A (en) | 2018-03-30 | 2018-03-30 | Image processing method, image processing device, electronic equipment and storage medium |
PCT/CN2019/078094 WO2019184715A1 (en) | 2018-03-30 | 2019-03-14 | Image processing method and device, electronic device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810277457.5A CN108492348A (en) | 2018-03-30 | 2018-03-30 | Image processing method, image processing device, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108492348A true CN108492348A (en) | 2018-09-04 |
Family
ID=63317238
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810277457.5A Pending CN108492348A (en) | 2018-03-30 | 2018-03-30 | Image processing method, image processing device, electronic equipment and storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108492348A (en) |
WO (1) | WO2019184715A1 (en) |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109345602A (en) * | 2018-09-28 | 2019-02-15 | Oppo广东移动通信有限公司 | Image processing method and device, storage medium, electronic equipment |
CN109360261A (en) * | 2018-09-28 | 2019-02-19 | 北京达佳互联信息技术有限公司 | Image processing method, device, electronic equipment and storage medium |
CN109376618A (en) * | 2018-09-30 | 2019-02-22 | 北京旷视科技有限公司 | Image processing method, device and electronic equipment |
CN109726302A (en) * | 2018-12-10 | 2019-05-07 | 广州景骐科技有限公司 | Mask method, device, computer equipment and the storage medium of image |
CN109739414A (en) * | 2018-12-29 | 2019-05-10 | 努比亚技术有限公司 | A kind of image processing method, mobile terminal, computer readable storage medium |
CN109785408A (en) * | 2018-12-29 | 2019-05-21 | 北京金山安全软件有限公司 | Map filling method and device and electronic equipment |
CN110033500A (en) * | 2018-12-29 | 2019-07-19 | 北京金山安全软件有限公司 | Graph filling method and device and electronic equipment |
CN110276735A (en) * | 2019-06-28 | 2019-09-24 | 北京金山安全软件有限公司 | Method, device and equipment for generating image color retention effect and storage medium |
WO2019184715A1 (en) * | 2018-03-30 | 2019-10-03 | 北京金山安全软件有限公司 | Image processing method and device, electronic device and storage medium |
CN110348535A (en) * | 2019-07-17 | 2019-10-18 | 北京金山数字娱乐科技有限公司 | A kind of vision Question-Answering Model training method and device |
CN110379358A (en) * | 2019-07-04 | 2019-10-25 | 南京宇丰晔禾信息科技有限公司 | LED display image broadcast control method, device |
CN110458921A (en) * | 2019-08-05 | 2019-11-15 | 腾讯科技(深圳)有限公司 | A kind of image processing method, device, terminal and storage medium |
CN110597589A (en) * | 2019-09-11 | 2019-12-20 | 北京达佳互联信息技术有限公司 | Page coloring method and device, electronic equipment and storage medium |
CN111080746A (en) * | 2019-12-10 | 2020-04-28 | 中国科学院计算技术研究所 | Image processing method, image processing device, electronic equipment and storage medium |
CN111091510A (en) * | 2019-12-16 | 2020-05-01 | 中国医科大学附属第一医院 | Image processing method and device, electronic equipment and storage medium |
CN111145306A (en) * | 2019-12-25 | 2020-05-12 | 北京奇艺世纪科技有限公司 | Picture processing method and device, electronic equipment and computer readable storage medium |
CN111340687A (en) * | 2020-02-21 | 2020-06-26 | 北京字节跳动网络技术有限公司 | Image processing method and device and electronic equipment |
CN111612870A (en) * | 2019-02-25 | 2020-09-01 | 阿里巴巴集团控股有限公司 | Image processing method and device |
CN111724449A (en) * | 2019-03-19 | 2020-09-29 | 北京京东尚科信息技术有限公司 | Image processing method, device and equipment |
CN112489169A (en) * | 2020-12-17 | 2021-03-12 | 脸萌有限公司 | Portrait image processing method and device |
CN113139923A (en) * | 2020-01-20 | 2021-07-20 | 北京达佳互联信息技术有限公司 | Image fusion method and device, electronic equipment and storage medium |
CN113467680A (en) * | 2021-06-28 | 2021-10-01 | 网易(杭州)网络有限公司 | Drawing processing method, drawing processing device, electronic equipment and storage medium |
WO2023273111A1 (en) * | 2021-06-29 | 2023-01-05 | 北京市商汤科技开发有限公司 | Image processing method and apparatus, and computer device and storage medium |
WO2023138441A1 (en) * | 2022-01-19 | 2023-07-27 | 北京字跳网络技术有限公司 | Video generation method and apparatus, and device and storage medium |
CN111080746B (en) * | 2019-12-10 | 2024-04-26 | 中国科学院计算技术研究所 | Image processing method, device, electronic equipment and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101764913A (en) * | 2008-12-10 | 2010-06-30 | 新奥特(北京)视频技术有限公司 | Color replacement method base on HSV space |
CN104581103A (en) * | 2013-10-21 | 2015-04-29 | 腾讯科技(深圳)有限公司 | Image processing method and device |
US20160117849A1 (en) * | 2014-03-28 | 2016-04-28 | Huawei Device Co., Ltd. | Method, apparatus, and terminal device for determining color of interface control |
CN105654437A (en) * | 2015-12-24 | 2016-06-08 | 广东迅通科技股份有限公司 | Enhancement method for low-illumination image |
CN105787878A (en) * | 2016-02-25 | 2016-07-20 | 杭州格像科技有限公司 | Beauty processing method and device |
CN107341763A (en) * | 2017-06-30 | 2017-11-10 | 北京金山安全软件有限公司 | Image processing method and device, electronic equipment and storage medium |
CN107358573A (en) * | 2017-06-16 | 2017-11-17 | 广东欧珀移动通信有限公司 | Image U.S. face treating method and apparatus |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108492348A (en) * | 2018-03-30 | 2018-09-04 | 北京金山安全软件有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
-
2018
- 2018-03-30 CN CN201810277457.5A patent/CN108492348A/en active Pending
-
2019
- 2019-03-14 WO PCT/CN2019/078094 patent/WO2019184715A1/en active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101764913A (en) * | 2008-12-10 | 2010-06-30 | 新奥特(北京)视频技术有限公司 | Color replacement method base on HSV space |
CN104581103A (en) * | 2013-10-21 | 2015-04-29 | 腾讯科技(深圳)有限公司 | Image processing method and device |
US20160117849A1 (en) * | 2014-03-28 | 2016-04-28 | Huawei Device Co., Ltd. | Method, apparatus, and terminal device for determining color of interface control |
CN105654437A (en) * | 2015-12-24 | 2016-06-08 | 广东迅通科技股份有限公司 | Enhancement method for low-illumination image |
CN105787878A (en) * | 2016-02-25 | 2016-07-20 | 杭州格像科技有限公司 | Beauty processing method and device |
CN107358573A (en) * | 2017-06-16 | 2017-11-17 | 广东欧珀移动通信有限公司 | Image U.S. face treating method and apparatus |
CN107341763A (en) * | 2017-06-30 | 2017-11-10 | 北京金山安全软件有限公司 | Image processing method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
Title |
---|
杨怀义等: "《Photoshop高级案例制作教程》", 28 February 2009, 天津科学技术出版社 * |
林剑楚 等: "《一种高保真人脸图像妆容移植方法》", 《计算机应用与软件》 * |
Cited By (34)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019184715A1 (en) * | 2018-03-30 | 2019-10-03 | 北京金山安全软件有限公司 | Image processing method and device, electronic device and storage medium |
CN109360261B (en) * | 2018-09-28 | 2019-12-31 | 北京达佳互联信息技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
CN109360261A (en) * | 2018-09-28 | 2019-02-19 | 北京达佳互联信息技术有限公司 | Image processing method, device, electronic equipment and storage medium |
CN109345602A (en) * | 2018-09-28 | 2019-02-15 | Oppo广东移动通信有限公司 | Image processing method and device, storage medium, electronic equipment |
CN109376618A (en) * | 2018-09-30 | 2019-02-22 | 北京旷视科技有限公司 | Image processing method, device and electronic equipment |
CN109726302A (en) * | 2018-12-10 | 2019-05-07 | 广州景骐科技有限公司 | Mask method, device, computer equipment and the storage medium of image |
CN109739414A (en) * | 2018-12-29 | 2019-05-10 | 努比亚技术有限公司 | A kind of image processing method, mobile terminal, computer readable storage medium |
CN109785408A (en) * | 2018-12-29 | 2019-05-21 | 北京金山安全软件有限公司 | Map filling method and device and electronic equipment |
CN110033500A (en) * | 2018-12-29 | 2019-07-19 | 北京金山安全软件有限公司 | Graph filling method and device and electronic equipment |
CN110033500B (en) * | 2018-12-29 | 2023-09-22 | 北京金山安全软件有限公司 | Graph filling method and device and electronic equipment |
CN109739414B (en) * | 2018-12-29 | 2021-12-14 | 努比亚技术有限公司 | Picture processing method, mobile terminal and computer readable storage medium |
CN111612870A (en) * | 2019-02-25 | 2020-09-01 | 阿里巴巴集团控股有限公司 | Image processing method and device |
CN111612870B (en) * | 2019-02-25 | 2023-06-20 | 阿里巴巴集团控股有限公司 | Image processing method and device |
CN111724449A (en) * | 2019-03-19 | 2020-09-29 | 北京京东尚科信息技术有限公司 | Image processing method, device and equipment |
CN110276735A (en) * | 2019-06-28 | 2019-09-24 | 北京金山安全软件有限公司 | Method, device and equipment for generating image color retention effect and storage medium |
CN110379358B (en) * | 2019-07-04 | 2021-03-30 | 南京宇丰晔禾信息科技有限公司 | LED display screen image playing and controlling method and device |
CN110379358A (en) * | 2019-07-04 | 2019-10-25 | 南京宇丰晔禾信息科技有限公司 | LED display image broadcast control method, device |
CN110348535A (en) * | 2019-07-17 | 2019-10-18 | 北京金山数字娱乐科技有限公司 | A kind of vision Question-Answering Model training method and device |
CN110458921A (en) * | 2019-08-05 | 2019-11-15 | 腾讯科技(深圳)有限公司 | A kind of image processing method, device, terminal and storage medium |
CN110458921B (en) * | 2019-08-05 | 2021-08-03 | 腾讯科技(深圳)有限公司 | Image processing method, device, terminal and storage medium |
CN110597589A (en) * | 2019-09-11 | 2019-12-20 | 北京达佳互联信息技术有限公司 | Page coloring method and device, electronic equipment and storage medium |
CN110597589B (en) * | 2019-09-11 | 2023-05-02 | 北京达佳互联信息技术有限公司 | Page coloring method and device, electronic equipment and storage medium |
CN111080746B (en) * | 2019-12-10 | 2024-04-26 | 中国科学院计算技术研究所 | Image processing method, device, electronic equipment and storage medium |
CN111080746A (en) * | 2019-12-10 | 2020-04-28 | 中国科学院计算技术研究所 | Image processing method, image processing device, electronic equipment and storage medium |
CN111091510A (en) * | 2019-12-16 | 2020-05-01 | 中国医科大学附属第一医院 | Image processing method and device, electronic equipment and storage medium |
CN111145306A (en) * | 2019-12-25 | 2020-05-12 | 北京奇艺世纪科技有限公司 | Picture processing method and device, electronic equipment and computer readable storage medium |
CN113139923A (en) * | 2020-01-20 | 2021-07-20 | 北京达佳互联信息技术有限公司 | Image fusion method and device, electronic equipment and storage medium |
CN113139923B (en) * | 2020-01-20 | 2024-04-30 | 北京达佳互联信息技术有限公司 | Image fusion method, device, electronic equipment and storage medium |
CN111340687A (en) * | 2020-02-21 | 2020-06-26 | 北京字节跳动网络技术有限公司 | Image processing method and device and electronic equipment |
CN112489169A (en) * | 2020-12-17 | 2021-03-12 | 脸萌有限公司 | Portrait image processing method and device |
CN112489169B (en) * | 2020-12-17 | 2024-02-13 | 脸萌有限公司 | Portrait image processing method and device |
CN113467680A (en) * | 2021-06-28 | 2021-10-01 | 网易(杭州)网络有限公司 | Drawing processing method, drawing processing device, electronic equipment and storage medium |
WO2023273111A1 (en) * | 2021-06-29 | 2023-01-05 | 北京市商汤科技开发有限公司 | Image processing method and apparatus, and computer device and storage medium |
WO2023138441A1 (en) * | 2022-01-19 | 2023-07-27 | 北京字跳网络技术有限公司 | Video generation method and apparatus, and device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
WO2019184715A1 (en) | 2019-10-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108492348A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
RU2693303C2 (en) | Method and device for changing background colour of tool panel according to dominant colour of image | |
CN108961303A (en) | A kind of image processing method, device, electronic equipment and computer-readable medium | |
US9100678B2 (en) | Image display method, server, and image display system | |
CN108701351B (en) | Image display enhancement method and device | |
CN108734749A (en) | The visual style of image converts | |
CN104076928B (en) | A kind of method for adjusting text importing image | |
CN105574866A (en) | Image processing method and apparatus | |
Berezhnoy et al. | Computer analysis of van Gogh’s complementary colours | |
CN105468617B (en) | Adjust the method and device of web page contents color | |
CN108288253A (en) | HDR image generation method and device | |
CN105608716B (en) | Information processing method and electronic equipment | |
CN106023276B (en) | Pencil drawing method for drafting and device based on image procossing | |
CN106663311B (en) | System and method for increasing the locating depth of image | |
CN110163832A (en) | Face fusion method, apparatus and terminal | |
US20210035336A1 (en) | Augmented reality display method of simulated lip makeup | |
CN109359499A (en) | A kind of method and apparatus for face classifier | |
CN108564526A (en) | Image processing method and device, electronic equipment and medium | |
CN104765600A (en) | Page processing method | |
US20200125880A1 (en) | Machine guided photo and video composition | |
CN109803172A (en) | A kind of processing method of live video, device and electronic equipment | |
CN110084871A (en) | Image typesetting process and device, electric terminal | |
CN111090368B (en) | Color matching method and device for determining page characters | |
CN110442313B (en) | Display attribute adjusting method and related equipment | |
CN109949248A (en) | Modify method, apparatus, equipment and the medium of the color of vehicle in the picture |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20201201 Address after: Room 115, area C, 1 / F, building 8, yard 1, yaojiayuan South Road, Chaoyang District, Beijing 100123 Applicant after: Beijing LEMI Technology Co.,Ltd. Address before: 100123 Building 8, Huitong Times Square, 1 South Road, Chaoyang District, Beijing. Applicant before: BEIJING KINGSOFT INTERNET SECURITY SOFTWARE Co.,Ltd. |
|
TA01 | Transfer of patent application right | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180904 |
|
RJ01 | Rejection of invention patent application after publication |