CN104680518B - A kind of blue screen image cutting method based on colourity Overflow handling - Google Patents
A kind of blue screen image cutting method based on colourity Overflow handling Download PDFInfo
- Publication number
- CN104680518B CN104680518B CN201510054488.0A CN201510054488A CN104680518B CN 104680518 B CN104680518 B CN 104680518B CN 201510054488 A CN201510054488 A CN 201510054488A CN 104680518 B CN104680518 B CN 104680518B
- Authority
- CN
- China
- Prior art keywords
- point
- pixel
- color
- foreground image
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000005070 sampling Methods 0.000 claims abstract description 7
- 230000002194 synthesizing effect Effects 0.000 claims abstract description 4
- 238000011835 investigation Methods 0.000 claims description 6
- 238000000205 computational method Methods 0.000 claims description 3
- 230000005574 cross-species transmission Effects 0.000 abstract description 8
- 230000000694 effects Effects 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000006748 scratching Methods 0.000 description 2
- 230000002393 scratching effect Effects 0.000 description 2
- 241001062009 Indigofera Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Abstract
The present invention relates to a kind of blue screen image cutting method based on colourity Overflow handling, it is characterised in that comprise the following steps that：The foreground image of picture to be scratched is divided into zone of opacity, translucent area, all-transparent region, and viewing area division result, the Alpha values based on sampling are calculated；Calculate the Alpha values of each point of foreground image.The background image for synthesizing with foreground image is loaded, foreground image and background image are synthesized, calculate the RGB color value of each pixel of composograph；Each pixel for composograph illustrates colourity Overflow handling；The method only needs to be simply input before shooting, you can be automatically completed stingy picture and colourity Overflow handling, even for the more serious image of colourity spillover, can also obtain more satisfied effect.
Description
Technical field
The present invention relates to a kind of blue screen image cutting method based on colourity Overflow handling, in particularly a kind of blue screen image cutting treatment
Based on the colourity Overflow handling method of sampling, belong to computer image processing technology field.
Background technology
Blue screen image cutting refers to that foreground target is accurately separated out from the image with known color background, then with other
Image synthesizes, so as to obtain the technology of required special-effect.In blue screen image cutting, the image referred to as foreground image of picture to be scratched；For
The image synthesized with it referred to as background image；Known monochromatic background color in foreground image is referred to as foreground image key color.
The blue screen image cutting of early stage usually requires that foreground image key color is blue, therefore the technology is named as blue screen image cutting.With stingy
As the development of technology, the photographing request to foreground image are less and less, the selection of foreground image key color can be according to shooting environmental
Conditional decision, it is green, the even color such as red can be used as foreground image key color.During blue screen image cutting technology is production of film and TV
One extremely important and widely used technology.
In blue screen image cutting processing procedure, there is a kind of colourity spillover, i.e.,：In the composite image, foreground target with
There is a side for monochrome between the background of synthesis, for example：When foreground image key color is for blueness, after foreground target is extracted, edge
Have the blue side of a circle.In fact, colourity spillover has just existed in not carrying out scratching the foreground image as treatment.Occur
The color of the pixel that colourity is overflowed is together decided on by light source colour, object reflected light, color background reflected light, therefore, each
The color of colourity spill point is different, and all similar with foreground image key color.
Shooting condition is the principal element for influenceing colourity spillover, by improving the resolution ratio of capture apparatus and more effective
Ground can to a certain extent improve colourity and overflow degree, but need larger investment with the use of modes such as multiple light courcess.Work as equipment
When the shooting condition such as resolution ratio and lighting is slightly worse, may result in more serious colourity and overflow, have a strong impact on composograph
The sense of reality.Another solves the method for colourity overflow problem, is also method the most direct, is exactly manually to be located after the picture is taken
Reason.For example, stingy picture treatment software After Effect popular now etc., are required for carrying out after the picture is taken the people of complexity
Work is operated, and could obtain the effect of satisfaction.This artificial processing mode cannot meet the requirement of real-time.
The content of the invention
To solve the problems, such as to need substantial amounts of user mutual and the treatment of colourity spillover not good in existing stingy image space method, this
Goal of the invention is to provide a kind of blue screen image cutting method based on colourity Overflow handling, and the method only needs to carry out letter before shooting
Single input, you can be automatically completed stingy picture and colourity Overflow handling, even for the more serious image of colourity spillover,
Also more satisfied effect can be obtained.
In order to reach above-mentioned purpose, the technical proposal of the invention is realized in this way：It is a kind of based on colourity Overflow handling
Blue screen image cutting method, it is characterised in that comprise the following steps that：
1st, the foreground image of picture to be scratched is divided into zone of opacity, translucent area, all-transparent region, and viewing area
Division result.This step includes following sub-step：
101st, loading and display foreground image, the color space of each pixel is rgb space in image；
102nd, known monochromatic background color in foreground image is referred to as foreground image key color, user passes through paintbrush instrument
In foreground image key color regional choice key color pixel point set, it is drawn black in foreground image key color region
Color thick lines are the key color pixel point set of user's selection；
103rd, the pixel selected according to user calculates foreground image key color average value；Foreground image key color is average
Value, wherein,、、RespectivelyTri- color component values of RGB；
Calculated according to formula below；
In formula, n is the sum of the pixel of user's selection,For the point that key color pixel is concentratedR colors point
Value；、Computational methods it is identical with this；
104th, each point concentrated for key color pixel, calculate it according to formula below crucial with foreground image
Color average valueColor distance, and record maximum；
105th, setting area divides parameter k1、k2Initial value be respectively、；
106th, for each pixel p in foreground image, it and foreground image key color average value are calculatedColor
Apart from D, work as D<k1When, point p belongs to all-transparent region；As d (p, b)>k2When, point p belongs to zone of opacity；
Otherwise, point p belongs to translucent area；Wherein, color distance maximum, is designated as；
107th, a region division result images are defined, its size is identical with foreground image；By region division result images
In the RGB color value of pixel corresponding with the all-transparent region of foreground image be set to (0,0,0), by region division result images
In be set to (255,255,255) with the RGB color value of the zone of opacity pixel of foreground image, by region division result images
In be set to (128,128,128) with the RGB color value of the translucent area pixel of foreground image；Viewing area division result figure
Picture；
108th, by parameter k1、k2Shown in the way of scroll bar respectively.Each scroll bar minimum value is,
Maximum is, adjusting step-length is；
109th, user's adjustment k1、k2Parameter until image in foreground target region for white, foreground image key color region is
The junction in black, foreground target and key color region is grey, with k1、k2The adjustment of parameter, repeat step 106 and step
107 carry out region division and in real time viewing area division result image；
2nd, the Alpha values based on sampling are calculated；
Calculate the Alpha values of each point of foreground image.Wherein, the Alpha values of zone of opacity pixel are 1；Full impregnated
Area pellucida domain pixel Alpha values are 0；The Alpha of translucent area pixel is below calculated using a kind of method based on sampling
Value, specifically includes following sub-step：
201st, sample point is found；
Translucent area is divided into the fritter of some 3*3, by 9 pixels of each 3*3 fritter numbering 0,1,2,3,
4,5,6,7,8 }；For each pixel in 3*3 fritters, centered on the point, respectively to four around it
Direction finding sample point, four direction is mutually perpendicular to；Define deflectionIt is direction line and the angle of horizontal direction；Pixel
Deflection, the pixel in 3*3 fritters is represented with rectangle, represent sample investigation with four arrow lines；
9 deflections of pixel in 3*3 fritters are as shown in the figure；In pixelEach sample investigation on run into
The point of one zone of opacity is not denoted as sample point, then pixelThe sample point set representations that can be found are；
202nd, Alpha values are carried out for each sample point to estimate；
For each sample point of each pixel p of translucent area, calculated respectively according to formula below
Alpha values；
It is the basis of p pointsThe Alpha values of sample point estimation,It is the color value of p points,It is foreground image key color
Average value；
203rd, determine optimal sample point, obtain the Alpha values of pixel；
For each sample point of each pixel p, calculated by formula belowValue；
For a point q,Represent the color value and basis of the pointThe difference of the color value being calculated；Table
Show the region of the 3*3 centered on p points；WithRepresent that p points are arrived respectivelyPoint andPoint space length because
Son.Military orderIt is worth minimum sample pointThe optimal sample point of referred to as p points；Then, the Alpha values of p points；
204th, calculate in translucent area Alpha values a little；Some are there may be in translucent area special
Pixel p, any sample point cannot be found according to the method in step 201.For this kind of point p, first, it is determined that in being with p points
The rectangular area of the 5*5 of the heart, calculates the optimal sample point of each point q in the regionIt is correspondingValue,Value is minimum
Three points be designated as respectively；According to the formula in step 103, calculate respectivelyThe color of point is put down
Average and optimal sample point、Color average, be designated as respectively、；Finally, according in step 202
Formula, calculate p points Alpha values, in making formula,, then；
3rd, the background image for synthesizing with foreground image is loaded, foreground image and background image is synthesized, according to following
Formula calculates the RGB color value of each pixel of composograph respectively；
It is the color value of pixel p in composograph,WithWith p points respectively in foreground image and background image
The color value of corresponding point,It is the transparency Alpha values being calculated in step 2；
4th, colourity Overflow handling；
For each pixel of composograph, below by taking the foreground image that key color is blue as an example, colourity is illustrated
The sub-step that Overflow handling step includes：
401st, according to the foreground image key color average value obtained in step 1, by formula below calculating parameter；
WithRespectivelyBlueness and green color component；
402nd, for each point p in composograph, its RGB color component value is designated as respectively、、；When-With-Value be all higher than 0.6*, andWhen, then, point p is colourity spill point, order。
The positive effect of the present invention is the quality of composograph quality in numeral is scratched as technology, is the true of influence image
The principal element of true feeling, is the key factor for determining product success or failure.Colourity spilling is a kind of intrinsic feature of blue screen image cutting,
Scratch and just existed in the foreground image as treatment.Made by improving the resolution ratio of capture apparatus and more effectively coordinating
With modes such as multiple light courcess, can to a certain extent improve colourity and overflow degree, but need larger investment.Existing stingy image space method
In, pair be mainly with the processing mode of colourity spillover carries out substantial amounts of artificial treatment and Xiu Tu, process in post-production
It is cumbersome, waste time and energy.Blue screen image cutting method based on colourity Overflow handling proposed by the present invention, it is only necessary to which user carries out simple defeated
Enter, you can automatically complete blue screen image cutting and colourity Overflow handling.Compared with prior art, the present invention drops for scratching as in
Low man-machine interactively, solution colourity overflow problem and raising are scratched as quality aspect has good application effect；Even for colourity
The more serious image of spillover, can also obtain more satisfied effect, and this has been greatly reduced the restrictive condition for shooting.
Brief description of the drawings
Fig. 1 is that foreground image key color selects schematic diagram.As illustrated, the region pointed by 1 is foreground image key color
Region, the lines pointed by 2 are the foreground image key color pixel point set of user's selection.
Fig. 2 is region division result schematic diagram.As illustrated, white portion is zone of opacity, gray area is semi-transparent
Area pellucida domain, black region is all-transparent region.
Fig. 3 is that sample finds schematic diagram.0 ~ 8 each pixel for representing a 3*3 fritter in translucent area, it is red
Color arrow is the sample investigation of each pixel, and θ is deflection.
Fig. 4 is flow chart of the invention.This figure is also specification digest accompanying drawing.
Specific embodiment
Embodiments of the present invention are described in detail below in conjunction with accompanying drawing and specific embodiment.
The present embodiment is scratched for the image shot in a width blue background using a kind of blue screen based on colourity Overflow handling
Image space method, handling process is as shown in Figure 4.Comprise the following steps that：
1st, foreground image is divided into zone of opacity, translucent area, all-transparent region, and viewing area divides knot
Really.This step includes following sub-step：
101st, the foreground image shot in a width blue background is loaded and shows, as shown in Figure 1.The image is first by indigo plant
Color curtain is used as background, then shoots what the personage being in front of the blue curtain background obtained with camera.After image loading, each
The color space of pixel is rgb space.
102nd, user passes through paintbrush instrument in foreground image key color regional choice key color pixel point set.As shown in figure 1,
Drawn heavy black line bar is the key color pixel point set of user's selection in foreground image key color region, records selected by user
Point, and be designated as。
103rd, the pixel point set selected according to user, calculate foreground image key color average value；Foreground image
Key color average value, wherein,RespectivelyTri- color component values of RGB.
Calculated according to formula below；
N is the sum of the pixel of user's selection,For the point that key color pixel is concentratedR color component values；
、Computational methods it is identical with this.
104th, each point concentrated for key color pixel, calculate it according to formula below crucial with foreground image
Color average valueColor distance, and record maximum。
The 105th, the region division parameter k of foreground image is set1、k2Initial value, respectively、。
106th, according to region division parameter k1、k2, preliminary region division is carried out to foreground image.Specific method is：For
Each pixel p in foreground image, calculates it and foreground image key color average valueColor distance D, work as D<k1When, point p belongs to all-transparent region；As d (p, b)>k2When, point p belongs to zone of opacity；Otherwise, point p belongs to semi-transparent
Area pellucida domain.During region division, foreground image pixel and foreground image key color average value are recordedColor distance DMaximum, be designated as。
107th, a region division result images are defined, its size is identical with foreground image；By region division result images
In the RGB color value of pixel corresponding with the all-transparent region of foreground image be set to (0,0,0), by region division result images
In be set to (255,255,255) with the RGB color value of the zone of opacity pixel of foreground image, by region division result images
In be set to (128,128,128) with the RGB color value of the translucent area pixel of foreground image；And viewing area division result
Image.
108th, by parameter k1、k2Shown in the way of scroll bar respectively.Each scroll bar minimum value is,
Maximum is, adjusting step-length is。
109th, user's adjustment k1、k2Parameter；With k1、k2The adjustment of parameter, system repeats steps 106 and step 107 are entered again
Row region division and in real time viewing area division result image；User checks the region division result images for changing therewith, until
Foreground target region is white in image, and foreground image key color region is black, the handing-over in foreground target and key color region
It is grey to locate, as shown in Figure 2.
2nd, the Alpha values of each point of foreground image are calculated.Wherein, the Alpha values of zone of opacity pixel are 1；Entirely
Transparent region pixel Alpha values are 0；The Alpha values of translucent area pixel be 0 ~ 1 between decimal.Hereinafter use one
The Alpha values that the method based on sampling calculates translucent area pixel are planted, following sub-step is specifically included：
201st, for each pixel of translucent area, sample point is found in zone of opacity.Specific method：Will be partly
Transparent region is divided into the fritter of some 3*3, and 9 pixels of each 3*3 fritter are numbered into { 0,1,2,3,4,5,6,7,8 }；
For each pixel in 3*3 fritters, centered on the point, find sample to four direction around it respectively
This point, four direction is mutually perpendicular to；Define deflectionIt is direction line and the angle of horizontal direction；PixelDeflection.As shown in figure 3, representing the pixel in 3*3 fritters with rectangle, sample searching side is represented with four arrow lines
To；9 deflections of pixel in 3*3 fritters are as shown in the figure；In pixelEach sample investigation on run into
The point of first zone of opacity be not denoted as sample point, then for the pixel of translucent area, can at most find
Four sample points, are designated as。
202nd, for translucent area each pixel p each sample point, counted respectively according to formula below
Calculate Alpha values.
It is p pointsThe Alpha values of sample point,It is the color value of p points,It is foreground image key color average value.
203rd, determine optimal sample point, obtain the Alpha values of pixel.For each sample of each pixel p
Point, calculated by formula belowValue.
For a point q,Represent the color value and basis of the pointThe difference of the color value being calculated；Table
Show the region of the 3*3 centered on p points；WithRepresent that p points are arrived respectivelyPoint andPoint space length because
Son.Military orderIt is worth minimum sample pointThe optimal sample point of referred to as p points；Then, it is calculated according to optimal sample point
Alpha values are exactly the Alpha values of p points,。
204th, there are some special pixel p in translucent area, cannot be found according to the method in step 201 and appointed
What sample point；Alpha value calculating methods for this kind of point p are：First, it is determined that the rectangular area of the 5*5 centered on p points,
Calculate the optimal sample point of each point q in the regionIt is correspondingValue,Three minimum points of value are designated as respectively；According to the formula in step 103, calculate respectivelyThe color average and optimal sample of point
Point、Color average, be designated as respectively、；Finally, according to the formula in step 202, p points are calculated
Alpha values, in making formula,, then。
3rd, the background image for synthesizing with foreground image is loaded, foreground image and background image is synthesized, according to following
Formula calculates the RGB color value of each pixel of composograph respectively.
It is the color value of pixel p in composograph,WithWith p points respectively in foreground image and background image
The color value of corresponding point,It is the transparency Alpha values being calculated in step 2.
4th, colourity Overflow handling.
For each pixel of composograph, colourity Overflow handling, including following sub-step are carried out：
401st, according to the foreground image key color average value obtained in step 1, by formula below calculating parameter。
WithRespectivelyBlueness and green color component.
402nd, for each point p in composograph, its RGB color component value is designated as respectively、、；When-With-Value be all higher than 0.6*, andWhen, then, point p is colourity spill point, order。
By above step, the composograph of colourity Overflow handling can be carried out.
Claims (1)
1. a kind of blue screen image cutting method based on colourity Overflow handling, it is characterised in that comprise the following steps that：
1st, the foreground image of picture to be scratched is divided into zone of opacity, translucent area, all-transparent region, and viewing area divides
As a result, this step includes following sub-step：
101st, loading and display foreground image, the color space of each pixel is rgb space in image；
102nd, known monochromatic background color in foreground image is referred to as foreground image key color, user is by paintbrush instrument preceding
Scape image key color regional choice key color pixel point set, drawn black is thick in foreground image key color region
Lines are the key color pixel point set of user's selection；
103rd, the pixel selected according to user calculates foreground image key color average value；Foreground image key color average value, wherein,、、RespectivelyTri- color component values of RGB；
Calculated according to formula below；
In formula, n is the sum of the pixel of user's selection,For the point that key color pixel is concentratedR color component values；、Computational methods it is identical with this；
104th, each point concentrated for key color pixel, calculate it according to formula below flat with foreground image key color
AverageColor distance, and record maximum；
105th, setting area divides parameter k1、k2Initial value be respectively、；
106th, for each pixel p in foreground image, it and foreground image key color average value are calculatedColor distance
D, work as D<k1When, point p belongs to all-transparent region；As d (p, b)>k2When, point p belongs to zone of opacity；It is no
Then, point p belongs to translucent area；Wherein, color distance maximum, is designated as；
107th, a region division result images are defined, its size is identical with foreground image；By in region division result images with
The RGB color value of the corresponding pixel in all-transparent region of foreground image is set to (0,0,0), by region division result images with
The RGB color value of the zone of opacity pixel of foreground image is set to (255,255,255), by region division result images with
The RGB color value of the translucent area pixel of foreground image is set to (128,128,128)；Viewing area division result image；
108th, by parameter k1、k2Shown in the way of scroll bar respectively, each scroll bar minimum value is, maximum
For, adjusting step-length is；
109th, user's adjustment k1、k2Until foreground target region is white in image, foreground image key color region is for black for parameter
The junction in color, foreground target and key color region is grey, with k1、k2The adjustment of parameter, repeat step 106 and step 107
Carry out region division and in real time viewing area division result image；
2nd, the Alpha values based on sampling are calculated；
The Alpha values of each point of foreground image are calculated, wherein, the Alpha values of zone of opacity pixel are 1；Full impregnated area pellucida
Domain pixel Alpha values are 0；The Alpha values of translucent area pixel, tool are below calculated using a kind of method based on sampling
Body includes following sub-step：
201st, sample point is found；
Translucent area is divided into the fritter of some 3*3, by 9 pixels of each 3*3 fritter numbering 0,1,2,3,4,5,
6,7,8 }；For each pixel in 3*3 fritters, centered on the point, respectively to four direction around it
Sample point is found, four direction is mutually perpendicular to；Define deflectionIt is direction line and the angle of horizontal direction；PixelSide
To angle, the pixel in 3*3 fritters is represented with rectangle, represent sample investigation with four arrow lines；3*3
9 deflections of pixel in fritter are as shown in the figure；In pixelEach sample investigation on run into first
The point of individual zone of opacity is denoted as sample point, then pixelThe sample point set representations that can be found are；
202nd, Alpha values are carried out for each sample point to estimate；
For each sample point of each pixel p of translucent area, calculate Alpha values respectively according to formula below；
It is the basis of p pointsThe Alpha values of sample point estimation,It is the color value of p points,For foreground image key color is average
Value；
203rd, determine optimal sample point, obtain the Alpha values of pixel；
For each sample point of each pixel p, calculated by formula belowValue；
For a point q,Represent the color value and basis of the pointThe difference of the color value being calculated；Represent with p
The region of the 3*3 centered on point；WithRepresent that p points are arrived respectivelyPoint andThe space length factor of point, will
OrderIt is worth minimum sample pointThe optimal sample point of referred to as p points；Then, the Alpha values of p points；
204th, calculate in translucent area Alpha values a little；Some special pixels are there may be in translucent area
Point p, any sample point cannot be found according to the method in step 201, for this kind of point p, first, it is determined that centered on p points
The rectangular area of 5*5, calculates the optimal sample point of each point q in the regionIt is correspondingValue,The three of value minimum
Individual point is designated as respectively；According to the formula in step 103, calculate respectivelyThe color average of point,
And optimal sample point、Color average, be designated as respectively、；Finally, according to the public affairs in step 202
Formula, calculates the Alpha values of p points, in making formula,, then；
3rd, the background image for synthesizing with foreground image is loaded, foreground image and background image is synthesized, according to formula below
The RGB color value of each pixel of composograph is calculated respectively；
It is the color value of pixel p in composograph,WithIt is corresponding with p points respectively in foreground image and background image
Point color value,It is the transparency Alpha values being calculated in step 2；
4th, colourity Overflow handling；
For each pixel of composograph, below by taking the foreground image that key color is blue as an example, illustrate that colourity is overflowed
The sub-step that process step includes：
401st, according to the foreground image key color average value obtained in step 1, by formula below calculating parameter；
WithRespectivelyBlueness and green color component；
402nd, for each point p in composograph, its RGB color component value is designated as respectively、、；When-With-Value be all higher than 0.6*, andWhen, then, point p is colourity spill point, order。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510054488.0A CN104680518B (en) | 2015-02-03 | 2015-02-03 | A kind of blue screen image cutting method based on colourity Overflow handling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510054488.0A CN104680518B (en) | 2015-02-03 | 2015-02-03 | A kind of blue screen image cutting method based on colourity Overflow handling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104680518A CN104680518A (en) | 2015-06-03 |
| CN104680518B true CN104680518B (en) | 2017-06-13 |
Family
ID=53315511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510054488.0A Expired - Fee Related CN104680518B (en) | 2015-02-03 | 2015-02-03 | A kind of blue screen image cutting method based on colourity Overflow handling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104680518B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106251286B (en) * | 2015-12-30 | 2019-11-22 | 深圳超多维科技有限公司 | Image processing method, device and equipment |
| CN106952270A (en) * | 2017-03-01 | 2017-07-14 | 湖南大学 | A kind of quickly stingy drawing method of uniform background image |
| CN108896094B (en) * | 2018-06-01 | 2020-07-07 | 安图实验仪器(郑州)有限公司 | Medical microscope slide sample missing judging method |
| CN110969595B (en) * | 2019-12-02 | 2023-08-01 | 成都索贝数码科技股份有限公司 | Pure-color background matting synthesis method based on real-time background color overflow inhibition |
| CN112150348A (en) * | 2020-08-13 | 2020-12-29 | 北京旷视科技有限公司 | Method, device and equipment for generating makeup trial and makeup trial component template and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1547131A (en) * | 2003-12-03 | 2004-11-17 | 谭日辉 | Virtual reality interactive control system |
| CN1801888A (en) * | 2005-01-05 | 2006-07-12 | 英华达(上海)电子有限公司 | Method for realizing picture background transformation in digital camera |
| US8847922B1 (en) * | 2007-09-04 | 2014-09-30 | Imaging Systems Technology, Inc. | Calibrating of interactive touch system for image compositing |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7602990B2 (en) * | 2005-09-29 | 2009-10-13 | Mitsubishi Electric Research Laboratories, Inc. | Matting using camera arrays |
-
2015
- 2015-02-03 CN CN201510054488.0A patent/CN104680518B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1547131A (en) * | 2003-12-03 | 2004-11-17 | 谭日辉 | Virtual reality interactive control system |
| CN1801888A (en) * | 2005-01-05 | 2006-07-12 | 英华达(上海)电子有限公司 | Method for realizing picture background transformation in digital camera |
| US8847922B1 (en) * | 2007-09-04 | 2014-09-30 | Imaging Systems Technology, Inc. | Calibrating of interactive touch system for image compositing |
Non-Patent Citations (1)
| Title |
|---|
| 数字抠像的最新研究进展;张展鹏;《自动化学报》;20111031;1571-1584 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104680518A (en) | 2015-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104680518B (en) | A kind of blue screen image cutting method based on colourity Overflow handling | |
| CN107516319B (en) | High-precision simple interactive matting method, storage device and terminal | |
| CN103581571B (en) | Video image matting method based on three elements of color | |
| CN106897981A (en) | A kind of enhancement method of low-illumination image based on guiding filtering | |
| CN105100646B (en) | Method for processing video frequency and device | |
| CN103366364B (en) | A kind of stingy drawing method based on color distortion | |
| CN110570360B (en) | Retinex-based robust and comprehensive low-quality illumination image enhancement method | |
| CN107451969A (en) | Image processing method, device, mobile terminal and computer-readable recording medium | |
| CN103248793B (en) | The colour of skin optimization method of gamut conversion system and device | |
| CN103716547A (en) | Smart mode photographing method | |
| CN105869112B (en) | A kind of edge keeps the high dynamic range image tone mapping method of minimum | |
| CN108848325A (en) | A kind of image synthesizing method for scratching picture based on natural image | |
| CN106063249A (en) | Photographing apparatus, method of controlling the same, and computer-readable recording medium | |
| CN107087123A (en) | It is a kind of that image space method is scratched based on the real-time high-definition that high in the clouds is handled | |
| Wang et al. | Variational single nighttime image haze removal with a gray haze-line prior | |
| CN105915816A (en) | Method and equipment for determining brightness of given scene | |
| CN110163807B (en) | Low-illumination image enhancement method based on expected bright channel | |
| CN107396079B (en) | White balance adjustment method and device | |
| CN107317967A (en) | Image processing method, device, mobile terminal and computer-readable recording medium | |
| CN108140237A (en) | Image processing apparatus and image processing method | |
| CN108022223A (en) | A kind of tone mapping method based on the processing fusion of logarithmic mapping function piecemeal | |
| CN106530309A (en) | Video matting method and system based on mobile platform | |
| CN107194884A (en) | A kind of several LDR image acquisition methods with different exposure values | |
| CN106657816A (en) | ORB algorithm based multipath rapid video splicing algorithm with image registering and image fusion in parallel | |
| CN110782400A (en) | Self-adaptive uniform illumination realization method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170613 |