CN107808404A - Image processing method, system, readable storage medium storing program for executing and dollying equipment - Google Patents
Image processing method, system, readable storage medium storing program for executing and dollying equipment Download PDFInfo
- Publication number
- CN107808404A CN107808404A CN201710805304.9A CN201710805304A CN107808404A CN 107808404 A CN107808404 A CN 107808404A CN 201710805304 A CN201710805304 A CN 201710805304A CN 107808404 A CN107808404 A CN 107808404A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel value
- pixel point
- current pixel
- face
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 26
- 238000001914 filtration Methods 0.000 claims abstract description 99
- 230000002146 bilateral effect Effects 0.000 claims abstract description 47
- 238000000034 method Methods 0.000 claims abstract description 16
- 230000002087 whitening effect Effects 0.000 claims description 52
- 238000009877 rendering Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 210000004709 eyebrow Anatomy 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 210000004209 hair Anatomy 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 241000404144 Pieris melete Species 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003733 optic disk Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
Classifications
-
- G06T5/70—
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20028—Bilateral filtering
-
- 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
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
Abstract
The present invention, which provides a kind of image processing method, system, readable storage medium storing program for executing and dollying equipment, this method, to be included：Original RGB image is subjected to bilateral filtering and boundary filtering processing, to obtain bilateral filtering image and boundary filtering image；When determining the pixel that any current pixel point in original RGB image is the colour of skin, judge whether pixel value of the current pixel point in boundary filtering image is less than threshold value；If it is not, then using pixel value of the current pixel point in original RGB image as U.S. face pixel value corresponding to current pixel point；If so, pixel value of the current pixel point on original RGB image and bilateral filtering image is then subjected to hybrid operation, and using the mixed pixel value calculated as U.S. face pixel value corresponding to current pixel point；The U.S. face pixel value according to corresponding to each current pixel point, original RGB image is subjected to U.S. face and rendered.Image processing method among the present invention, the border of image is preferably protected, the distortion factor is small, properer in true picture after U.S. face.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of image processing method, system, readable storage medium storing program for executing
And dollying equipment.
Background technology
With shooting and the continuous progress of image processing techniques, people also require captured while high pixel is pursued
Image U.S. face, i.e. picture pick-up device can carry out U.S. face to shot image automatically automatically.In order to meet the needs of users,
Many electronics business mens all addition of the function of automatic U.S. face in the dollying equipment produced.
However, among prior art, the processing method for realizing image U.S. face at present is that camera is shot using CPU
Whole pictures enter row format conversion (being typically converted into RGB image), the computings such as then recycling CPU is filtered, the colour of skin adjusts,
To reach automatic U.S. face effect, whole U.S. face process is serial arithmetic, and filtering algorithm does not protect surrounding environment and people
Eyebrow, border minutia, the distortion factor such as hair it is big.
The content of the invention
Based on this, it is an object of the invention to provide a kind of small image processing method of distortion factor, system, readable storage medium storing program for executing
And dollying equipment.
A kind of image processing method according to embodiments of the present invention, including：
Original RGB image is subjected to bilateral filtering and boundary filtering processing, bilateral filtering image and border are obtained with corresponding
Filtering image；
When determining the pixel that any current pixel point in the original RGB image is the colour of skin, described work as is judged
Whether pixel value of the preceding pixel point in the boundary filtering image is less than threshold value；
If it is not, then using pixel value of the current pixel point in the original RGB image as the current pixel point
Corresponding U.S. face pixel value；
If so, the then pixel value by the current pixel point on the original RGB image and the bilateral filtering image
Hybrid operation is carried out, and using the mixed pixel value calculated as the U.S. face pixel value corresponding to the current pixel point；
The U.S. face pixel value according to corresponding to each current pixel point, the original RGB image is subjected to U.S. face
Render.
In addition, a kind of image processing method according to the above embodiment of the present invention, can also have technology additional as follows
Feature：
Further, original RGB image is subjected to bilateral filtering and boundary filtering processing described, it is bilateral correspondingly to obtain
After the step of filtering image and boundary filtering image, in addition to：
Judge the current pixel point whether be the colour of skin pixel；
Work as if it is not, then performing the pixel value using the current pixel point in the original RGB image as described
Corresponding to preceding pixel point the step of U.S. face pixel value.
Further, the formula by original RGB image progress bilateral filtering processing is：
Wherein,WPFor normalizing, Ifiltered(x) refer to
Filtered image, I (x) refer to original image, xiRefer to the coordinate of the pixel of current filter processing, Ω refers to pixel point coordinates center
Point, gsIt is the convolution kernel in space, frRefer to codomain core；
Pixel value of the current pixel point on the original RGB image and the bilateral filtering image is mixed
The formula of computing is：(1-smooth)*(A(xi,yi)-B(xi,yi))+B(xi,yi),
Wherein, smooth is smoothness, A (xi,yi) it is pixel of the current pixel point in the original RGB image
Value, B (xi,yi) it is pixel value of the current pixel point on the bilateral filtering image.
Further, will be described original in the U.S. face pixel value corresponding to the basis each current pixel point
After RGB image carries out the step of U.S. face renders, in addition to：
The U.S. face pixel value corresponding to the current pixel point is subjected to whitening computing, to obtain the current pixel point
Corresponding whitening pixel value；
The whitening pixel value according to corresponding to each current pixel point, the original RGB image is subjected to whitening
Render；
Wherein, the formula by the U.S. face pixel value progress whitening computing corresponding to the current pixel point is：w(xi,yi) it is the U.S. face pixel value corresponding to the current pixel point,
v(xi,yi) it is the corresponding whitening pixel value that the current pixel point obtained after whitening computing, β is brightness enhancing system
Number.
Further, it is described to judge that the step of whether current pixel point is the pixel of the colour of skin includes：
Judge whether pixel value of the current pixel point in the original RGB image is located at the pixel value of the colour of skin
In the range of.
A kind of image processing system according to embodiments of the present invention, including：
Image processing module, it is double correspondingly to obtain for original RGB image to be carried out into bilateral filtering and boundary filtering processing
Side filtering image and boundary filtering image；
Pixel value judge module, for being the colour of skin when determining any current pixel point in the original RGB image
During pixel, judge whether pixel value of the current pixel point in the boundary filtering image is less than threshold value；
First processing module is not small for determining pixel value of the current pixel point in the boundary filtering image
When the threshold value, using pixel value of the current pixel point in the original RGB image as the current pixel point pair
The U.S. face pixel value answered；
Second processing module, it is less than for determining pixel value of the current pixel point in the boundary filtering image
During the threshold value, pixel value of the current pixel point on the original RGB image and the bilateral filtering image is carried out
Hybrid operation, and using the mixed pixel value calculated as the U.S. face pixel value corresponding to the current pixel point；
U.S. face rendering module, for the U.S. face pixel value according to corresponding to each current pixel point, by the original
Beginning RGB image carries out U.S. face and rendered.
In addition, a kind of image processing system according to the above embodiment of the present invention, can also have technology additional as follows
Feature：
Further, described image processing system also includes：
Pixel judge module, for judge the current pixel point whether be the colour of skin pixel；
When it is not the pixel of the colour of skin to determine the current pixel point, the first processing module described will be worked as
Pixel value of the preceding pixel point in the original RGB image is as U.S. face pixel value corresponding to the current pixel point.
Further, described image processing system also includes：
Whitening computing module, for the U.S. face pixel value corresponding to the current pixel point to be carried out into whitening computing, with
Obtain whitening pixel value corresponding to the current pixel point；
Whitening rendering module, for the whitening pixel value according to corresponding to each current pixel point, by the original
Beginning RGB image carries out whitening and rendered；
Wherein, the formula by the U.S. face pixel value progress whitening computing corresponding to the current pixel point is：w(xi,yi) it is the U.S. face pixel value corresponding to the current pixel point,
v(xi,yi) it is the corresponding whitening pixel value that the current pixel point obtained after whitening computing, β is brightness enhancing system
Number.
The present invention also proposes a kind of computer-readable recording medium, is stored thereon with computer program, the program is processed
Device realizes above-mentioned image processing method when performing.
The present invention also proposes a kind of dollying equipment, including memory, processor and storage are on a memory and can
The computer program run on a processor, the processor are GPU, are realized during the computing device described program above-mentioned
Image processing method.
Above-mentioned image processing method, system, readable storage medium storing program for executing and dollying equipment, by scheming to the original RGB
As carrying out boundary filtering processing, to obtain the boundary filtering image, then to each pixel in the original RGB image
Judged, when determine current pixel point be the colour of skin and its be not less than the threshold value in the pixel value of the boundary filtering image
When, the pixel is represented as the boundary pixel point in the colour of skin, then when being rendered to the U.S. face of pixel progress, using the pixel
Original pixel value in the original RGB image, when determine current pixel point be the colour of skin and its in the boundary filtering figure
When the pixel value of picture is less than the threshold value, represents the pixel as the colour of skin and be non-border pixel point, then clicked through to the pixel
When the U.S. face of row renders, using the mixing of pixel value of the pixel in the original RGB image and the bilateral filtering image
Composite pixel value, therefore described image processing method, system, readable storage medium storing program for executing and dollying equipment are carried out to the colour of skin
The border such as eyebrow, hair minutia can be preferably protected while U.S. face, and the colour of skin and non-border pixel point are carried out
Comprehensive U.S. face considers, therefore the image fault degree after U.S. face is small, properer in true picture.
Brief description of the drawings
Fig. 1 is the flow chart of the image processing method in first embodiment of the invention.
Fig. 2 is the flow chart of the image processing method in second embodiment of the invention.
Fig. 3 is the structural representation of the image processing system in third embodiment of the invention.
Main element symbol description：
| Image processing module | 11 | Pixel judge module | 12 |
| First processing module | 13 | Second processing module | 14 |
| U.S. face rendering module | 15 | Colour of skin judging unit | 121 |
| Threshold decision unit | 122 | Whitening computing module | 16 |
| Whitening rendering module | 17 | Image conversion module | 18 |
| Colour of skin judgment sub-unit | 1211 |
Detailed description below will combine above-mentioned accompanying drawing and further illustrate the present invention.
Embodiment
For the ease of understanding the present invention, the present invention is described more fully below with reference to relevant drawings.In accompanying drawing
Give some embodiments of the present invention.But the present invention can realize in many different forms, however it is not limited to this paper institutes
The embodiment of description.On the contrary, the purpose for providing these embodiments is made to the disclosure more thorough and comprehensive.
It should be noted that when element is referred to as " being fixedly arranged on " another element, it can be directly on another element
Or there may also be element placed in the middle.When an element is considered as " connection " another element, it can be directly connected to
To another element or it may be simultaneously present centering elements.Term as used herein " vertical ", " horizontal ", " left side ",
" right side " and similar statement are for illustrative purposes only.
Unless otherwise defined, all of technologies and scientific terms used here by the article is with belonging to technical field of the invention
The implication that technical staff is generally understood that is identical.Term used in the description of the invention herein is intended merely to description tool
The purpose of the embodiment of body, it is not intended that in the limitation present invention.Term as used herein " and/or " include one or more phases
The arbitrary and all combination of the Listed Items of pass.
Referring to Fig. 1, show the image processing method in first embodiment of the invention, including step S01 is to step
S03。
Step S01, carries out bilateral filtering by original RGB image and boundary filtering is handled, and bilateral filtering figure is obtained with corresponding
Picture and boundary filtering image.
Wherein, the purpose handled by bilateral filtering the original RGB image is to protect side denoising, in realization pair
While the original RGB image is handled, the edge of the original RGB image is protected.The boundary filtering processing
Claim Canny filtering process, and Canny filtering process is realized based on Canny algorithms, passes through the mesh of Canny filtering process
The border for being to search out the original RGB image.
Step S02, when determining the pixel that any current pixel point in the original RGB image is the colour of skin, sentence
Whether disconnected pixel value of the current pixel point in the boundary filtering image is less than threshold value.
Wherein, the step S02 can come sequentially to each according to putting in order for pixel in the original RGB image
Pixel is judged.
It is pointed out that the pixel value of the colour of skin is within the specific limits, by judging the pixel in the original RGB image
Whether the pixel value of point is in the range of the colour of skin (skin pixel value), you can judges whether the pixel is the colour of skin.
In addition it is also necessary to, it is noted that when the pixel in the original RGB image is in the boundary filtering image
When pixel value is not less than the threshold value, the pixel is represented in the picture to change more violent position, the pixel is side
Boundary, similarly, when pixel value of the pixel in the original RGB image in the boundary filtering image is less than the threshold value
When, represent the pixel and change in the picture uniformly, the pixel is non-border.
It is not less than institute when the step S02 determines pixel value of the current pixel point in the boundary filtering image
Threshold value is stated, represents the pixel as border (intersection of such as colour of skin and eyebrow) pixel in the colour of skin, then performs step S03,
When the step S02, which determines pixel value of the current pixel point in the boundary filtering image, is less than the threshold value,
The pixel is represented as the colour of skin and be non-border pixel point, then perform step S04.
Step S03, using pixel value of the current pixel point in the original RGB image as the current pixel point
Corresponding U.S. face pixel value.
It should be understood that the purpose of the step S03 is, it is the border in the colour of skin when determining the current pixel point
During pixel, directly rendered using the original pixel value of the pixel as follow-up U.S. face, therefore, preferably protecting the colour of skin
In border.
In addition it is also necessary to, it is noted that when it is not the pixel of the colour of skin to determine the current pixel point, then hold
The row step S03, preferably protects the surrounding environment (such as background) in the original RGB image.
Step S04, by pixel value of the current pixel point on the original RGB image and the bilateral filtering image
Hybrid operation is carried out, and using the mixed pixel value calculated as the U.S. face pixel value corresponding to the current pixel point.
Wherein, pixel value of the current pixel point on the original RGB image and the bilateral filtering image is entered
The process of row hybrid operation is：To the current pixel point in the pixel value of the original RGB image and in the bilateral filtering
Pixel value on image distributes certain weight proportion, is then carrying out hybrid operation, is finally giving a comprehensive pixel value,
As described U.S. face pixel value.
Therefore the purpose of the step S04 is, it is the colour of skin when determining the current pixel point and is non-border pixel point
When, integrated, realized to the colour of skin and non-by the pixel value after the original pixel value and bilateral filtering by the pixel
While boundary pixel point carries out U.S. face, make it proper small in time of day, the distortion factor as far as possible.
Step S05, the U.S. face pixel value according to corresponding to each current pixel point, by the original RGB image
U.S. face is carried out to render.
It should be understood that it can be obtained by the step S03 and the step S04 every among the original RGB image
The U.S. face pixel value corresponding to individual pixel, then by the way that each U.S. face pixel value is rendered in corresponding pixel
On, you can complete the U.S. face to the original RGB image.
To sum up, the image processing method among the above embodiment of the present invention, by entering row bound to the original RGB image
Filtering process, to obtain the boundary filtering image, then each pixel in the original RGB image is judged,
When it is not the colour of skin to determine current pixel point, pixel of the pixel as surrounding environment is represented, then is clicked through to the pixel
When the U.S. face of row renders, using original pixel value of the pixel in the original RGB image, it is when determining current pixel point
The colour of skin and its when the pixel value of the boundary filtering image is not less than the threshold value, represent the pixel as the border in the colour of skin
Pixel, then when being rendered to the U.S. face of pixel progress, using original pixels of the pixel in the original RGB image
Value, when determine current pixel point be the colour of skin and its be less than the threshold value in the pixel value of the boundary filtering image when, representative
The pixel is the colour of skin and is non-border pixel point, then when being rendered to the U.S. face of pixel progress, using the pixel in institute
The composite pixel value of the mixing of original RGB image and the pixel value in the bilateral filtering image is stated, therefore described image is handled
Around method, system, readable storage medium storing program for executing and dollying equipment can be protected preferably while U.S. face is carried out to the colour of skin
Environment (such as background), eyebrow, hair border minutia, and have been carried out by comprehensive U.S. face and has been examined for the colour of skin and non-border pixel point
Consider, therefore the image fault degree after U.S. face is small, it is properer in true picture.
Referring to Fig. 2, show the image processing method in second embodiment of the invention, including step S11 is to step
S18。
Step S11, the YUV image that camera obtains is converted into original RGB image.
It is pointed out that the image that camera obtains is usually YUV image, Gu need to change the YUV image of acquisition
For corresponding original RGB image, in order to which follow-up U.S. face operates.
Furthermore, it is possible to completed using existing YUV and RGB form conversion formula, the YUV image that camera is obtained
Be converted to original RGB image.
Step S12, carries out bilateral filtering by the original RGB image and boundary filtering is handled, and bilateral filter is obtained with corresponding
Ripple image and boundary filtering image.
Wherein, the formula by original RGB image progress bilateral filtering processing is：
Wherein,WPFor normalizing, Ifiltered(x) refer to
Filtered image, I (x) refer to original image, xiRefer to the coordinate of the pixel of current filter processing, Ω refers to pixel point coordinates center
Point, gsIt is the convolution kernel in space, frRefer to codomain core.
Step S13, judge any current pixel point in the original RGB image whether be the colour of skin pixel.
Wherein, it is described to judge that the step of whether current pixel point is the skin pixel point includes：
Judge whether pixel value of the current pixel point in the original RGB image is located at the pixel value of the colour of skin
In the range of.When the pixel value for determining pixel value of the current pixel point in the original RGB image and being located at the colour of skin
In the range of when, then judge the pixel that the current pixel point is the colour of skin, as skin pixel point is described when determining
When pixel value of the current pixel point in the original RGB image is not located in the pixel value range of the colour of skin, then judge
The current pixel point is not the pixel of the colour of skin, as non-skin pixel point.
When it is not the pixel of the colour of skin that the step S13, which determines the current pixel point, the pixel is represented
For surrounding environment, then perform step S14, step S17, step S18 and step S19 successively, when the step S14 determine it is described
When current pixel point is the pixel of the colour of skin, the pixel is represented as the colour of skin, then performs step S15.
Step S14, using pixel value of the current pixel point in the original RGB image as the current pixel point
Corresponding U.S. face pixel value.
Step S15, judges whether pixel value of the current pixel point in the boundary filtering image is less than threshold value.
Institute is less than when the step S15 determines pixel value of the current pixel point in the boundary filtering image
When stating threshold value, represent the pixel as the colour of skin and be non-border pixel point, then step S16 to step S19 is performed, when the step
When S15 determines pixel value of the current pixel point in the boundary filtering image and is not less than the threshold value, then represent
The pixel is the boundary pixel point in the colour of skin, then returns and perform the step S14.
Step S16, by pixel value of the current pixel point on the original RGB image and the bilateral filtering image
Hybrid operation is carried out, and using the mixed pixel value calculated as the U.S. face pixel value corresponding to the current pixel point.
Wherein, pixel value of the current pixel point on the original RGB image and the bilateral filtering image is entered
The formula of row hybrid operation is：(1-smooth)*(A(xi,yi)-B(xi,yi))+B(xi,yi),
Wherein, A (xi,yi) it is pixel value of the current pixel point in the original RGB image, B (xi,yi) for institute
Pixel value of the current pixel point on the bilateral filtering image is stated, smooth is smoothness, can be carried out according to actual conditions
Adjustment, when smoothness is higher, the image proportion after bilateral filtering is bigger, when smoothness is 1, the institute of the current pixel point
State U.S. face pixel value and directly use pixel value of the current pixel point on the bilateral filtering image, when smoothness is 0,
The U.S. face pixel value of the current pixel point directly uses its original pixel value on the original RGB image.
Step S17, the U.S. face pixel value according to corresponding to each current pixel point, by the original RGB image
U.S. face is carried out to render.
Step S18, the U.S. face pixel value corresponding to the current pixel point is subjected to whitening computing, to obtain described work as
Whitening pixel value corresponding to preceding pixel point.
Wherein, the formula by the U.S. face pixel value progress whitening computing corresponding to the current pixel point is：w(xi,yi) it is the U.S. face pixel value, v corresponding to the current pixel point
(xi,yi) it is the corresponding whitening pixel value that the current pixel point obtained after whitening computing, β is brightness enhancing system
Number, can be adjusted according to actual conditions.
Step S19, the whitening pixel value according to corresponding to each current pixel point, by the original RGB image
Whitening is carried out to render.
By the tone scale map of artwork it is two using log curves it is pointed out that the step S18 and step S19
End is slightly weak, middle slightly strong, can naturally strengthen picture brightness, reaches the effect of whitening.
To sum up, the image processing method among the above embodiment of the present invention, in implementing compared to first at central image
Reason method, it will also carry out whitening after completing to carry out U.S. face to the original RGB image to the whole image after U.S. face
Processing, further to improve the image effect after U.S. face.
Another aspect of the present invention also provides a kind of image processing system, refer to Fig. 3, show third embodiment of the invention
In image processing system, described image processing system is GPU, and described image processing system is arranged in dollying equipment
(such as mobile phone, flat board), including：
Image processing module 11, for original RGB image to be carried out into bilateral filtering and boundary filtering processing, obtained with corresponding
Bilateral filtering image and boundary filtering image；
Pixel value judge module 12, for being the colour of skin when determining any current pixel point in the original RGB image
Pixel when, judge whether pixel value of the current pixel point in the boundary filtering image is less than threshold value；
First processing module 13, for determining pixel value of the current pixel point in the boundary filtering image not
During less than the threshold value, using pixel value of the current pixel point in the original RGB image as the current pixel point
Corresponding U.S. face pixel value；
Second processing module 14 is small for determining pixel value of the current pixel point in the boundary filtering image
When the threshold value, pixel value of the current pixel point on the original RGB image and the bilateral filtering image is entered
Row hybrid operation, and using the mixed pixel value calculated as the U.S. face pixel value corresponding to the current pixel point；
U.S. face rendering module 15, for the U.S. face pixel value according to corresponding to each current pixel point, by described in
Original RGB image carries out U.S. face and rendered.
Wherein, the formula by original RGB image progress bilateral filtering processing is：
Wherein,WPFor normalizing, Ifiltered(x) refer to
Filtered image, I (x) refer to original image, xiRefer to the coordinate of the pixel of current filter processing, Ω refers to pixel point coordinates center
Point, gsIt is the convolution kernel in space, frRefer to codomain core；
Pixel value of the current pixel point on the original RGB image and the bilateral filtering image is mixed
The formula of computing is：(1-smooth)*(A(xi,yi)-B(xi,yi))+B(xi,yi),
Wherein, smooth is smoothness, A (xi,yi) it is pixel of the current pixel point in the original RGB image
Value, B (xi,yi) it is pixel value of the current pixel point on the bilateral filtering image.
Further, described image processing system also includes：
Pixel judge module 16, for judge the current pixel point whether be the colour of skin pixel；
When it is not the pixel of the colour of skin to determine the current pixel point, the first processing module 13 is by described in
Pixel value of the current pixel point in the original RGB image is as U.S. face pixel value corresponding to the current pixel point.
Further, described image processing system also includes：
Whitening computing module 17, for the U.S. face pixel value corresponding to the current pixel point to be carried out into whitening computing,
To obtain whitening pixel value corresponding to the current pixel point；
Whitening rendering module 18, for the whitening pixel value according to corresponding to each current pixel point, by described in
Original RGB image carries out whitening and rendered；
Wherein, the formula by the U.S. face pixel value progress whitening computing corresponding to the current pixel point is：w(xi,yi) it is the U.S. face pixel value, v corresponding to the current pixel point
(xi,yi) it is the corresponding whitening pixel value that the current pixel point obtained after whitening computing, β is brightness enhancing system
Number.
Further, described image processing system also includes：
Image conversion module 19, the YUV image for camera to be obtained are converted to the original RGB image.
Further, the pixel judge module 16 includes：
Pixel judgment sub-unit 161, for judging the pixel value of each pixel in the original RGB image
Whether in skin color range.
Among the present embodiment, because the image of camera acquisition is directly by GPU processing, and whole image handles and U.S.
White process is completed on GPU, compared to prior art, can preferably discharge CPU calculation resources, while also accelerate whole
Individual image processing process, so as to which the effect of real-time U.S. face can also be realized on a mobile platform.
The present invention also proposes a kind of computer-readable recording medium, is stored thereon with computer program, the program is processed
Device realizes image processing method described above when performing.
The present invention also proposes a kind of dollying equipment, including memory, processor and storage are on a memory and can
The computer program run on a processor, the processor are GPU, are realized as described above during the computing device described program
Image processing method.
It will be understood by those skilled in the art that in flow charts represent or logic described otherwise above herein and/or
Step, for example, being considered the order list of the executable instruction for realizing logic function, it may be embodied in and appoint
In what computer-readable medium, for instruction execution system, device or equipment (such as computer based system including processor
System or other can be from the system of instruction execution system, device or equipment instruction fetch and execute instruction) use, or with reference to this
A little instruction execution system, device or equipment and use.For the purpose of this specification, " computer-readable medium " can be it is any can be with
Comprising, store, communicate, propagate or transmit program and perform system for instruction execution system, device or equipment or with reference to these instructions
System, device or equipment and the device used.
The more specifically example (non-exhaustive list) of computer-readable medium includes following：Connected up with one or more
Electrical connection section (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read-only storage
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium, which can even is that, to print the paper of described program thereon or other are suitable
Medium, because can then enter edlin, interpretation or if necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In embodiment, software that multiple steps or method can be performed in memory and by suitable instruction execution system with storage
Or firmware is realized.If, and in another embodiment, can be with well known in the art for example, realized with hardware
Any one of row technology or combinations thereof are realized：With the logic gates for realizing logic function to data-signal
Discrete logic, have suitable combinational logic gate circuit application specific integrated circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example description
Point is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term not
Necessarily refer to identical embodiment or example.Moreover, specific features, structure, material or the feature of description can be any
One or more embodiments or example in combine in an appropriate manner.
Embodiment described above only expresses the several embodiments of the present invention, and its description is more specific and detailed, but simultaneously
Therefore the limitation to the scope of the claims of the present invention can not be interpreted as.It should be pointed out that for one of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the guarantor of the present invention
Protect scope.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (10)
- A kind of 1. image processing method, it is characterised in that including：Original RGB image is subjected to bilateral filtering and boundary filtering processing, bilateral filtering image and boundary filtering are obtained with corresponding Image；When determining the pixel that any current pixel point in the original RGB image is the colour of skin, the current picture is judged Whether pixel value of the vegetarian refreshments in the boundary filtering image is less than threshold value；If it is not, the pixel value then using the current pixel point in the original RGB image is corresponding as the current pixel point U.S. face pixel value；If so, then pixel value of the current pixel point on the original RGB image and the bilateral filtering image is carried out Hybrid operation, and using the mixed pixel value calculated as the U.S. face pixel value corresponding to the current pixel point；The U.S. face pixel value according to corresponding to each current pixel point, the original RGB image is subjected to U.S. face and rendered.
- 2. image processing method according to claim 1, it is characterised in that it is described original RGB image carried out it is bilateral Filtering and boundary filtering processing, after corresponding the step of obtaining bilateral filtering image and boundary filtering image, in addition to：Judge the current pixel point whether be the colour of skin pixel；If it is not, the pixel value using the current pixel point in the original RGB image is then performed as the current picture Corresponding to vegetarian refreshments the step of U.S. face pixel value.
- 3. image processing method according to claim 1, it is characterised in that the original RGB image is subjected to bilateral filter Ripple processing formula be：Wherein,WPFor normalizing, Ifiltered(x) filtering is referred to Image afterwards, I (x) refer to original image, xiRefer to the coordinate of the pixel of current filter processing, Ω refers to pixel point coordinates central point, gs It is the convolution kernel in space, frRefer to codomain core；Pixel value of the current pixel point on the original RGB image and the bilateral filtering image is subjected to hybrid operation Formula be：(1-smooth)*(A(xi,yi)-B(xi,yi))+B(xi,yi),Wherein, smooth is smoothness, A (xi,yi) it is pixel value of the current pixel point in the original RGB image, B (xi,yi) it is pixel value of the current pixel point on the bilateral filtering image.
- 4. image processing method according to claim 1, it is characterised in that in each current pixel point of the basis The corresponding U.S. face pixel value, after the original RGB image is carried out into the step of U.S. face renders, in addition to：The U.S. face pixel value corresponding to the current pixel point is subjected to whitening computing, it is corresponding to obtain the current pixel point Whitening pixel value；The whitening pixel value according to corresponding to each current pixel point, original RGB image progress whitening is rendered；Wherein, the formula by the U.S. face pixel value progress whitening computing corresponding to the current pixel point is：w(xi,yi) it is the U.S. face pixel value, v corresponding to the current pixel point (xi,yi) it is the corresponding whitening pixel value that the current pixel point obtained after whitening computing, β is brightness enhancing system Number.
- 5. image processing method according to claim 2, it is characterised in that it is described judge the current pixel point whether be The step of pixel of the colour of skin, includes：Judge whether pixel value of the current pixel point in the original RGB image is located at the pixel value range of the colour of skin It is interior.
- A kind of 6. image processing system, it is characterised in that including：Image processing module, for original RGB image to be carried out into bilateral filtering and boundary filtering processing, bilateral filter is obtained with corresponding Ripple image and boundary filtering image；Pixel value judge module, for working as the pixel that any current pixel point determined in the original RGB image is the colour of skin During point, judge whether pixel value of the current pixel point in the boundary filtering image is less than threshold value；First processing module, it is not less than institute for determining pixel value of the current pixel point in the boundary filtering image When stating threshold value, using pixel value of the current pixel point in the original RGB image as corresponding to the current pixel point U.S. face pixel value；Second processing module, for determining pixel value of the current pixel point in the boundary filtering image less than described During threshold value, pixel value of the current pixel point on the original RGB image and the bilateral filtering image is mixed Computing, and using the mixed pixel value calculated as the U.S. face pixel value corresponding to the current pixel point；U.S. face rendering module, for the U.S. face pixel value according to corresponding to each current pixel point, by the original RGB Image carries out U.S. face and rendered.
- 7. image processing system according to claim 6, it is characterised in that described image processing system also includes：Pixel judge module, for judge the current pixel point whether be the colour of skin pixel；When it is not the pixel of the colour of skin to determine the current pixel point, the first processing module is by the current picture Pixel value of the vegetarian refreshments in the original RGB image is as U.S. face pixel value corresponding to the current pixel point.
- 8. image processing system according to claim 6, it is characterised in that described image processing system also includes：Whitening computing module, for the U.S. face pixel value corresponding to the current pixel point to be carried out into whitening computing, to obtain Whitening pixel value corresponding to the current pixel point；Whitening rendering module, for the whitening pixel value according to corresponding to each current pixel point, by the original RGB Image carries out whitening and rendered；Wherein, the formula by the U.S. face pixel value progress whitening computing corresponding to the current pixel point is：w(xi,yi) it is the U.S. face pixel value, v corresponding to the current pixel point (xi,yi) it is the corresponding whitening pixel value that the current pixel point obtained after whitening computing, β is brightness enhancing system Number.
- 9. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is held by processor The method as described in claim 1-5 is any is realized during row.
- 10. a kind of dollying equipment, including memory, processor and storage can be run on a memory and on a processor Computer program, it is characterised in that the processor is GPU, realizes that right such as will during the computing device described program Seek any described methods of 1-5.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710805304.9A CN107808404A (en) | 2017-09-08 | 2017-09-08 | Image processing method, system, readable storage medium storing program for executing and dollying equipment |
| PCT/CN2017/116362 WO2019047409A1 (en) | 2017-09-08 | 2017-12-15 | Image processing method and system, readable storage medium and mobile camera device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710805304.9A CN107808404A (en) | 2017-09-08 | 2017-09-08 | Image processing method, system, readable storage medium storing program for executing and dollying equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107808404A true CN107808404A (en) | 2018-03-16 |
Family
ID=61576562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710805304.9A Pending CN107808404A (en) | 2017-09-08 | 2017-09-08 | Image processing method, system, readable storage medium storing program for executing and dollying equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107808404A (en) |
| WO (1) | WO2019047409A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109639982A (en) * | 2019-01-04 | 2019-04-16 | Oppo广东移动通信有限公司 | A kind of image denoising method, device, storage medium and terminal |
| CN110363717A (en) * | 2019-06-28 | 2019-10-22 | 北京字节跳动网络技术有限公司 | A kind of method, apparatus, medium and electronic equipment handling face-image |
| CN111462003A (en) * | 2020-03-20 | 2020-07-28 | 稿定（厦门）科技有限公司 | Face image processing method, medium, device and apparatus |
| CN111669611A (en) * | 2020-06-19 | 2020-09-15 | 广州繁星互娱信息科技有限公司 | Image processing method, device, terminal and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101123683A (en) * | 2007-08-27 | 2008-02-13 | 北京航空航天大学 | Restraint method for SAR image spot noise combining visual light image information |
| CN106920211A (en) * | 2017-03-09 | 2017-07-04 | 广州四三九九信息科技有限公司 | U.S. face processing method, device and terminal device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101348596B1 (en) * | 2008-01-22 | 2014-01-08 | 삼성전자주식회사 | Apparatus and method for immersive generation |
| CN101930601B (en) * | 2010-09-01 | 2012-02-08 | 浙江大学 | Edge information-based multi-scale blurred image blind restoration method |
| US9105088B1 (en) * | 2013-10-04 | 2015-08-11 | Google Inc. | Image blur with preservation of detail |
| CN105787888A (en) * | 2014-12-23 | 2016-07-20 | 联芯科技有限公司 | Human face image beautifying method |
| CN106550243A (en) * | 2016-12-09 | 2017-03-29 | 武汉斗鱼网络科技有限公司 | Live video processing method, device and electronic equipment |
-
2017
- 2017-09-08 CN CN201710805304.9A patent/CN107808404A/en active Pending
- 2017-12-15 WO PCT/CN2017/116362 patent/WO2019047409A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101123683A (en) * | 2007-08-27 | 2008-02-13 | 北京航空航天大学 | Restraint method for SAR image spot noise combining visual light image information |
| CN106920211A (en) * | 2017-03-09 | 2017-07-04 | 广州四三九九信息科技有限公司 | U.S. face processing method, device and terminal device |
Non-Patent Citations (2)
| Title |
|---|
| LAVIEWPBT: "对皮肤美白算法的一些研究", 《博客园》 * |
| 天空之城: "美妆算法---磨皮算法研究汇总", 《微像素》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109639982A (en) * | 2019-01-04 | 2019-04-16 | Oppo广东移动通信有限公司 | A kind of image denoising method, device, storage medium and terminal |
| CN109639982B (en) * | 2019-01-04 | 2020-06-30 | Oppo广东移动通信有限公司 | Image noise reduction method and device, storage medium and terminal |
| CN110363717A (en) * | 2019-06-28 | 2019-10-22 | 北京字节跳动网络技术有限公司 | A kind of method, apparatus, medium and electronic equipment handling face-image |
| CN110363717B (en) * | 2019-06-28 | 2021-07-23 | 北京字节跳动网络技术有限公司 | Method, device, medium and electronic equipment for processing face image |
| CN111462003A (en) * | 2020-03-20 | 2020-07-28 | 稿定（厦门）科技有限公司 | Face image processing method, medium, device and apparatus |
| CN111462003B (en) * | 2020-03-20 | 2022-08-23 | 稿定（厦门）科技有限公司 | Face image processing method, medium, device and apparatus |
| CN111669611A (en) * | 2020-06-19 | 2020-09-15 | 广州繁星互娱信息科技有限公司 | Image processing method, device, terminal and storage medium |
| CN111669611B (en) * | 2020-06-19 | 2022-02-22 | 广州繁星互娱信息科技有限公司 | Image processing method, device, terminal and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019047409A1 (en) | 2019-03-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Galdran et al. | Automatic red-channel underwater image restoration | |
| Chen et al. | Robust image and video dehazing with visual artifact suppression via gradient residual minimization | |
| Guo | LIME: A method for low-light image enhancement | |
| CN107808404A (en) | Image processing method, system, readable storage medium storing program for executing and dollying equipment | |
| AU2011253980B2 (en) | Method, apparatus and system for identifying distracting elements in an image | |
| CN105046658B (en) | A kind of low-light (level) image processing method and device | |
| WO2016206087A1 (en) | Low-illumination image processing method and device | |
| US9881202B2 (en) | Providing visual effects for images | |
| CN109658330B (en) | Color development adjusting method and device | |
| CN109300101A (en) | A kind of more exposure image fusion methods based on Retinex theory | |
| CN107194869A (en) | A kind of image processing method and terminal, computer-readable storage medium, computer equipment | |
| CN111489322A (en) | Method and device for adding sky filter to static picture | |
| CN109859217A (en) | The dividing method in pore region and calculating equipment in facial image | |
| CN107516302A (en) | A kind of method of the mixed image enhancing based on OpenCV | |
| Xu et al. | Deep retinex decomposition network for underwater image enhancement | |
| Liu et al. | Single color image dehazing based on digital total variation filter with color transfer | |
| Yu et al. | Hue-preserving unsharp-masking for color image enhancement | |
| US11501404B2 (en) | Method and system for data processing | |
| Yan et al. | A natural-based fusion strategy for underwater image enhancement | |
| CN112200730B (en) | Image filtering processing method, device, equipment and storage medium | |
| CN113450282A (en) | Method and system for beautifying image | |
| EP3062288A1 (en) | Method, apparatus and computer program product for reducing chromatic aberrations in deconvolved images | |
| Kurihara et al. | Low-light image enhancement via adaptive shape and texture prior | |
| CN107492078B (en) | Method for removing black noise in image and computing equipment | |
| Zhou et al. | Saliency preserving decolorization |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180316 |