CN109934783A - Image processing method, device, computer equipment and storage medium - Google Patents
Image processing method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109934783A CN109934783A CN201910158997.6A CN201910158997A CN109934783A CN 109934783 A CN109934783 A CN 109934783A CN 201910158997 A CN201910158997 A CN 201910158997A CN 109934783 A CN109934783 A CN 109934783A
- Authority
- CN
- China
- Prior art keywords
- component
- image data
- image
- pixel
- yuv
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Color Image Communication Systems (AREA)
- Image Processing (AREA)
Abstract
This application involves a kind of image processing method, device, computer equipment and storage mediums.The described method includes: obtaining YUV image data；The YUV image data are copied to graphics processor memory from central processing unit memory；The YUV image data parallel is converted into rgb image data in the graphics processor memory；Image beautification operation is carried out to the rgb image data, obtains the rgb image data by beautification；It is final YUV image data by the rgb image data Parallel transformation by beautification；Export the final YUV image data.Final YUV image data can be exported using this method, the time of the image procossing of equipment is reduced, improve the efficiency of image procossing.
Description
Technical field
This application involves field of computer technology, more particularly to a kind of image processing method, device, computer equipment and
Storage medium.
Background technique
With universal and camera function the continuous reinforcement of smart phone, mill skin, goes the images beautification technologies such as wrinkle at U.S. face
Become indispensable a kind of basic skills in most people life.And the digital camera technology bring high score graduallyd mature
The eruptive growth of resolution image and network direct broadcasting allows people to have higher demand to image beautification technology and require.
Currently, most of image beautification algorithm is by central processing unit (CPU, Central Processing Unit)
Handled, and the processing mode of CPU is serial process, once there are the problem of one of be exactly image resolution ratio increasing
Greatly, the processing time of CPU will steeply rise, if ultrahigh resolution video image on CPU realize in real time beautification etc. operation when, meeting
The processing time for increasing equipment, reduce the efficiency of image procossing.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, at the image for providing a kind of efficiency that can be improved image procossing
Manage method, apparatus, computer equipment and storage medium.
A kind of image processing method, which comprises
Obtain YUV image data；
The YUV image data are copied to graphics processor memory from central processing unit memory；
The YUV image data parallel is converted into rgb image data in the graphics processor memory；
Image beautification operation is carried out to the rgb image data, obtains the rgb image data by beautification；
It is final YUV image data by the rgb image data Parallel transformation by beautification；
Export the final YUV image data.
It is described in one of the embodiments, to turn the YUV image data parallel in the graphics processor memory
It is changed to rgb image data, comprising:
In the graphics processor memory according to the number of each pixel in the YUV image data, picture traverse and
Component length gauge calculates the index of the U component of each pixel and the index of V component；
According to the number of each pixel, the index of U component and the index of V component, it is Y points corresponding to obtain each pixel
Amount, U component and V component；
It by the corresponding Y-component of each pixel, U component and V component Parallel transformation is each according to the first preset rules
The corresponding R component of pixel, G component and B component, and the R component, G component and B component are stored into corresponding array position,
Obtain the rgb image data.
It is described in one of the embodiments, that image beautification operation is carried out to the rgb image data, it obtains by beautification
Rgb image data, comprising:
Calculate the mean value and quadratic sum of each pixel in the rgb image data；
The variance of each pixel is obtained according to the quadratic sum；
First time image co-registration is carried out according to the mean and variance of each pixel, obtains the first image co-registration image；
Face Detection is carried out for the first image blending image, obtains the skin region of the first image blending image
Domain；
Second of image co-registration is carried out for the skin area of the first image blending image, is obtained described by beautification
Rgb image data.
It is described in one of the embodiments, to be melted according to the mean and variance of each pixel progress first time image
It closes, obtains the first image co-registration image, comprising:
Obtain the first customized parameter；
According to the variance and the first customized parameter of each pixel, integration percentage coefficient is obtained；
According to the pixel value of each pixel in the mean value and the integration percentage coefficient adjustment rgb image data, obtain
The first image blending image.
It is described in one of the embodiments, to carry out Face Detection for the first image blending image, it obtains described
The skin area of first image co-registration image, comprising:
Obtain R component, G component and the B component of each pixel in the first image blending image；
When the R component, G component and B component meet preset condition, which is determined as skin area.
The skin area for the first image blending image carries out second of figure in one of the embodiments,
As fusion, the rgb image data by beautification is obtained, comprising:
Obtain the second customized parameter；
The pixel value of the skin area is adjusted according to second customized parameter, is obtained described by beautification
Rgb image data.
The rgb image data Parallel transformation by the process beautification is final YUV figure in one of the embodiments,
As data, comprising:
According to the second preset rules by it is described by beautification rgb image data each pixel R component, G component and B
Component in parallel is converted to corresponding Y-component, U component and V component；
The index of U component and the rope of V component are obtained according to the number of each pixel, picture traverse and component length
Draw；
The Y-component, U component and V component are stored to corresponding according to the index of the index of the U component and V component
Array position obtains the final YUV image data.
A kind of image processing apparatus, described device include:
Data acquisition module, for obtaining YUV image data；
Replication module, for the YUV image data to be copied to graphics processor memory from central processing unit memory；
First Parallel transformation module, for converting the YUV image data parallel in the graphics processor memory
For rgb image data；
Image beautifies module, for carrying out image beautification operation to the rgb image data, obtains the RGB by beautification
Image data；
Second Parallel transformation module, for being final YUV image by the rgb image data Parallel transformation by beautification
Data；
Output module, for exporting the final YUV image data.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
Device performs the steps of when executing the computer program
Obtain YUV image data；
The YUV image data are copied to graphics processor memory from central processing unit memory；
The YUV image data parallel is converted into rgb image data in the graphics processor memory；
Image beautification operation is carried out to the rgb image data, obtains the rgb image data by beautification；
It is final YUV image data by the rgb image data Parallel transformation by beautification；
Export the final YUV image data.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
It is performed the steps of when row
Obtain YUV image data；
The YUV image data are copied to graphics processor memory from central processing unit memory；
The YUV image data parallel is converted into rgb image data in the graphics processor memory；
Image beautification operation is carried out to the rgb image data, obtains the rgb image data by beautification；
It is final YUV image data by the rgb image data Parallel transformation by beautification；
Export the final YUV image data.
Above-mentioned image processing method, device, computer equipment and storage medium, by obtaining YUV image data；By YUV
Image data is copied to graphics processor memory from central processing unit memory；By YUV image data in graphics processor memory
Parallel transformation is rgb image data；Image beautification operation is carried out to rgb image data, obtains the RGB image number by beautification
According to；It is final YUV image data by the rgb image data Parallel transformation by beautification；Final YUV image data are exported, are reduced
The time of the image procossing of equipment improves the efficiency of image procossing.
Detailed description of the invention
Fig. 1 is a kind of applied environment figure of image processing method of one embodiment；
Fig. 2 is a kind of flow diagram of image processing method of one embodiment；
Fig. 3 is a kind of flow diagram of YUV image data parallel switch process of one embodiment；
Fig. 4 is a kind of flow diagram of rgb image data step of the acquisition of one embodiment by beautification；
Fig. 5 is a kind of flow diagram of acquisition the first image co-registration image step of one embodiment；
Fig. 6 is a kind of flow diagram of acquisition skin area step of one embodiment；
Fig. 7 is a kind of flow diagram of rgb image data step of the acquisition of one embodiment by beautification；
Fig. 8 is a kind of flow diagram of rgb image data Parallel transformation step by beautification of one embodiment；
Fig. 9 is a kind of schematic diagram of YUV image data of one embodiment；
Figure 10 is a kind of schematic diagram of final YUV image data of one embodiment；
Figure 11 is the schematic diagram of another YUV image data of one embodiment；
Figure 12 is the schematic diagram of the final YUV image data of another kind of one embodiment；
Figure 13 is a kind of structural block diagram of image processing apparatus of one embodiment；
Figure 14 is a kind of internal structure chart of computer equipment of one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Image processing method provided by the present application can be applied in application environment as shown in Figure 1.Wherein, terminal 102
It is communicated by network with server 104.Wherein, terminal 102 can be, but not limited to be various personal computers, notebook electricity
Brain, smart phone, tablet computer and portable wearable device, server 104 can be either multiple with independent server
The server cluster of server composition is realized.
In one embodiment, as shown in Fig. 2, providing a kind of image processing method, it is applied in Fig. 1 in this way
It is illustrated for server 104, comprising the following steps:
Step S201 obtains YUV image data；
In the present embodiment, which refers to the image data of yuv format, and YUV is a kind of color coded format,
It is divided into three components, and " Y " indicates brightness (Luminance or Luma), that is, gray value；And " U " and " V " is indicated then
It is coloration (Chrominance or Chroma), effect is the color and saturation degree for describing image, the color for specified pixel.
YUV is also a kind of color coded format.
In the present embodiment, server 104 may include PC (PersonalComputer, PC) server, it is big
Type machine, minicomputer, can also include Cloud Server, and the present embodiment does not limit the type and quantity of server specifically.
Further, which may include the image in video data, i.e., the YUV image data include view
The image of frequency each frame in, specifically, video data is extracted, is obtained by the available video file to yuv format
Obtain the YUV image data of every frame.
Specifically, the video file of the yuv format can be got by image collecting devices such as cameras, then will
Each frame image zooming-out in the video file of yuv format comes out, and the YUV image data can be obtained.In a kind of preferred implementation
In example, above-mentioned image collecting device can be connect with server, directly by the video file or YUV of collected yuv format
Image data is sent to server, and when detecting the video file for yuv format, server is literary by the video of the yuv format
Part is handled, and YUV image data are extracted.
YUV image data are copied to graphics processor memory from central processing unit memory by step S202；
It generally may include central processing unit (CPU, Central for the hardware structure of server
Processing Unit), graphics processor (GPU, Graphics Processing Unit), memory, hard disk and carrying it is above-mentioned
The mainboard of hardware, wherein above-mentioned central processing unit memory may include the memory connecting with central processing unit, also may include
Caching in central processing unit, caching are the scratchpad area (SPA) between CPU and memory；
On the other hand, graphics processor, which refers to aim at, executes the processor that complicated mathematics and geometry are calculated and designed,
It is the critical elements for connecting display and mainboard.The graphics processor may include image processor memory.
It is further applicable in the present embodiment, after getting YUV image data, by the YUV image data from centre
Reason device (CPU, Central Processing Unit) memory copies to graphics processor (GPU, Graphics Processing
Unit) in memory；It specifically, can be by the YUV after getting the YUV image data by image collecting devices such as cameras
Image data is sent on the storage device of server, CPU can send instruction by YUV image data be written to caching or it is interior
It deposits, then transmits it to GPU memory.
YUV image data parallel is converted to rgb image data in graphics processor memory by step S203；
It applies in the present embodiment, the process of CPU processing data is the process of a serial process, and GPU is handled
The process of data is the process of a parallel processing, in a kind of core idea of the present embodiment, passes through the spy of GPU parallel processing
Property, YUV image data parallel is converted into rgb image data in GPU memory, improves image real time transfer efficiency.
Specifically, the index of the U component of each pixel in YUV image data, V component can be identified first, according to
The index of above-mentioned U component, V component obtains the value of each component in the YUV image data, then by presetting conversion formula,
By the value of component each in YUV image data, R component, the value of G component and B component in rgb image data are converted to；Specifically
Each component in the YUV image data in multiple pixels in GPU memory parallel can be converted to R component, G component by ground
And B component, improve image processing efficiency.
It should be noted that the RGB is equally a kind of color coded format, R indicates that Red, G indicate that Green, B are indicated
Blue；Red green blue is also known as primaries, any one color can be indicated with rgb value in display device, i.e., should
Each of rgb image data pixel can be indicated with rgb value.
Step S204 carries out image beautification operation to rgb image data, obtains the rgb image data by beautification；
Further, image beautification operation can also be carried out for rgb image data in graphics processor memory, obtained
By the rgb image data of beautification；Image beautification operation may include the operation such as image detection, image co-registration, certainly, can also
With include mean filter of image etc. operation, the present embodiment to this with no restriction.
Specifically, image detection may include the operation such as edge detection, skin detection, and edge detection refers to detection image number
According to the apparent one or more pixel of middle brightness change, the algorithm of edge detection usually can be with difference edge detection algorithm etc.；
And Face Detection refers to and filters out the pixel for meeting preset threshold, using the pixel as skin area, traverses in image data
Multiple pixels filter out the skin area in image data.
Further, image co-registration can refer to that the pixel value for adjusting each pixel in rgb image data specifically can
To carry out the adjusting of pixel value according to the image after mean filter, the effect of image co-registration is realized.
In the present embodiment, YUV image data are converted into rgb image data, is handled, is mentioned for rgb image data
The colour vividness of hi-vision enhances the quality of image.
Rgb image data Parallel transformation by beautification is final YUV image data by step S205；
It applies in the present embodiment, is obtaining this after the rgb image data of beautification, it can also be in graphics process
It will be final YUV image data by the rgb image data Parallel transformation beautified in device memory, it will be by the rgb format of beautification
Image data be reconverted into the image data of yuv format, improve transfer efficiency, be convenient for image transmitting, so as to subsequent transmission
Or other operations, the efficiency of improve data transfer；Realize that yuv format turns rgb format and rgb format turns the GPU of yuv format simultaneously
Row realization process makes full use of the hardware-accelerated characteristic of GPU, improves calculating speed, reduces overall calculation time delay.
Step S206 exports final YUV image data.
It is specifically applied in the present embodiment, it, can will be final after server is converted to the final YUV image data
YUV image data are exported to terminal；For the different angle of output object, CPU can control the output of graphics processor memory
The final YUV image data send the final YUV image data to terminal, terminal exhibition into CPU memory, then through network interface
Now final YUV image data, it should be noted that when the quantity of final YUV image data is multiframe, can be broadcast with video
The mode put shows the final YUV image data of the multiframe.
In an advantageous embodiment, for the angle of application scenarios, the method for the present embodiment can also be used alone
In terminal or server, if hardware that is, in some terminal constitute in comprising GPU memory and CPU memory, can be by YUV image
Data are copied to graphics processor memory from central processing unit memory, by YUV image data parallel in graphics processor memory
Be converted to rgb image data；The rgb image data is subjected to image beautification operation again, the rgb image data after being beautified,
If the terminal includes display device, the rgb image data after beautifying can also be shown on the display apparatus, alternatively, can also incite somebody to action
Rgb image data after the beautification is converted to final YUV image data, then is shown in display device.
If being applied individually to server, export final YUV image data to server display device, in server
Show the final YUV image data in display device.
According to image processing method provided in this embodiment, YUV image data are obtained；By YUV image data from centre
Reason device memory is copied to graphics processor memory；YUV image data parallel is converted into RGB image in graphics processor memory
Data；Image beautification operation is carried out to rgb image data, obtains the rgb image data by beautification；By the RGB by beautification
Image data Parallel transformation is final YUV image data；Export final YUV image data, reduce the image procossing of equipment when
Between, improve the efficiency of image procossing.
In another embodiment, referring to Fig. 3, a kind of YUV image data parallel switch process of the present embodiment is shown
Flow diagram, YUV image data parallel is converted into rgb image data, including following son in graphics processor memory
Step:
Sub-step S11, according to the number of pixel each in YUV image data, picture traverse in graphics processor memory
And component length gauge calculates the index of the U component of each pixel and the index of V component；
Sub-step S12 obtains each pixel pair according to the index of the number of each pixel, the index of U component and V component
Y-component, U component and the V component answered；
Sub-step S13, according to the first preset rules by the corresponding Y-component of each pixel, U component and V component Parallel transformation
For the corresponding R component of each pixel, G component and B component, and R component, G component and B component are stored into corresponding array bit
It sets, obtains rgb image data.
Specifically, in the Parallel transformation process of YUV image data, it is available each into the YUV image data
Number, picture traverse and the component length of pixel, the component length may include the length of Y-component；First according to each picture
Number, picture traverse and the component length of element obtain the index of the U component of each pixel and the index of V component.
After obtaining the index of U component and the index of V component of each pixel, correspondence can be found according to above-mentioned index
Y-component, U component and the V component of pixel.
In the present embodiment, the first preset rules can also be preset, according to first preset rules by Y-component, U component
And V component is converted to R component, G component and B component, is stored at corresponding array position, the type of the array may include
Int type array.
In a kind of citing, after which copies to GPU memory from CPU memory, YUV is schemed in GPU memory
As data are converted, rgb image data is obtained.
The conversion process can be described as follows, if d_yuv indicates the yuv data array in GPU memory, w indicates that image is wide
Degree, longth indicate Y-component length, then obtain Y-component, U component, V component CUDA (Compute Unified Device
Architecture unifiedly calculates equipment framework) code may be expressed as:
Int U_idx=((i/w)/2) * (w/2)+((i%w)/2)+longth；
Int V_idx=U_idx+longth/4；
Int y_value=d_yuv [i]；
Int u_value=d_yuv [U_idx]；
Int v_value=d_yuv [V_idx]；
Wherein, i indicates the ith pixel of YUV image data；U_idx indicates the rope for corresponding to the U component of ith pixel
Draw；V_idx indicates the index for corresponding to the V component of ith pixel；Y_value, u_value, v_value respectively indicate correspondence
In the value of each component of the YUV of ith pixel, in conjunction with every in following available corresponding rgb color spaces of conversion formula (1)
The value of a pixel.The value in each each channel pixel RGB is respectively stored in an int type array element, image can be reduced
Internal storage access delay during beautification.
Y, U, V in the conversion formula (1) respectively indicate the value of the value of Y-component, the value of U component and V component；And the conversion
R, G, B in formula (1) respectively indicate the value of the value of R component, the value of G component and B component, wherein the conversion formula (1) is as
First preset rules convert the value of the value of the Y-component of each pixel, the value of U component and V component by the conversion formula
R component, G component and the B component of each pixel.
In the present embodiment, all calculating process, which are gone in GPU, to be realized, outputting and inputting for data is only responsible at the end CPU, can
It is saved while obtaining conversion effect and calculates the time.
In another embodiment, referring to Fig. 4, a kind of RGB image number of the acquisition of the present embodiment by beautification is shown
According to the flow diagram of step, image beautification operation is carried out to rgb image data, obtains the rgb image data by beautification, packet
Include following sub-step:
Sub-step S21 calculates the mean value and quadratic sum of each pixel in rgb image data；
Sub-step S22 obtains the variance of each pixel according to quadratic sum；
Sub-step S23 carries out first time image co-registration according to the mean and variance of each pixel, obtains the first image co-registration
Image；
Sub-step S24 carries out Face Detection for the first image co-registration image, obtains the skin of the first image co-registration image
Region；
Sub-step S25 carries out second of image co-registration for the skin area of the first image co-registration image, obtains by beauty
The rgb image data of change.
It is further applicable in the present embodiment, in image beautification operating process, the rgb image data can also be calculated
The mean value and quadratic sum of each pixel.
Specifically, the mean value computation formula of each pixel is as follows:
Wherein, E indicates the mean value in each components range of each pixel, fi,jIndicate the pixel value of the i-th row j column in image, r
To calculate radius.
Further, the quadratic sum calculation formula of each pixel is as follows:
Wherein, SquIndicate quadratic sum, fi,jIt indicates that the pixel value of the i-th row j column in image, r are to calculate radius, needs to illustrate
, to accelerate calculating speed, can use the high speed memory access speed advantage of the shared drive in GPU memory, to every in image
The summation of a pixel carries out decoupled method, and then progress column direction summation first carries out line direction summation, decoupled method effectively drops
Computing redundancy in low summation process.
Specifically, can also first the responsible pixel of all threads that per thread block includes be copied to before summing
In corresponding shared drive, then summation and squared and operation are carried out on the basis of shared drive to improve processing speed.Wherein,
The radius r that sums is related with the length of image and width, the conversion relation of r are as follows: r=max (long, wide) * c, c is radius factor, generally
0.01 < c < 0.02.
After calculating quadratic sum, the variance of each pixel in image, Ke Yican can also be calculated according to the quadratic sum
According to following variance calculation formula (4)；
Wherein, var indicates variance, Ssum=E* (2*r+1)2For the sum of pixel in neighborhood, E indicates each each component of pixel
Mean value in range, r are to calculate radius, W=(2*r+1)2For neighborhood window size.
It, can basis after obtaining the mean value and variance of each component of each pixel of rgb image data by above-mentioned formula
The mean value and variance carry out first time image co-registration, obtain the first image co-registration image.
After getting the first image co-registration image, the skin region of the first image co-registration image can also detect that
Domain；Second of image co-registration is carried out for the skin area, obtains the rgb image data by beautification；
It should be noted that above-mentioned image beautification operation is only a kind of citing of the present embodiment, it can also be passed through
His mode is beautified for the rgb image data, is such as carried out edge detection for the rgb image data, is detected personage
Image-region adjusts the tone and saturation degree in the character image region, the rgb image data after being beautified, the present embodiment pair
This is with no restriction.
In the present embodiment, image beautification transition of operation is realized into GPU, and propose mean filter and quadratic sum
The decoupled method process of operation, while the advantage of shared drive quickly accessed has been used, the time of image beautification is saved, it is real
The real-time beautification on high-resolution and ultrahigh resolution video image is showed.
In a kind of specific example, referring to Fig. 5, a kind of acquisition the first image co-registration image step of the present embodiment is shown
Flow diagram, according to the mean and variance of each pixel carry out first time image co-registration, obtain the first image co-registration image,
Including following sub-step:
Sub-step S231 obtains the first customized parameter；
Sub-step S232 obtains integration percentage coefficient according to the variance of each pixel and the first customized parameter；
Sub-step S233 is obtained according to the pixel value of each pixel in mean value and integration percentage coefficient adjustment rgb image data
To the first image co-registration image.
During first time image co-registration, first customized parameter can be calculated first, it first can according to this
Integration percentage coefficient t is calculated in the variance of adjustment parameter and each pixel.
Specifically, integration percentage coefficient t can be obtained by coefficient formulas, coefficient formulas is as follows；
T=var/ (var+ σ) --- --- -- coefficient formulas (5)
Wherein, σ indicates that the first customized parameter, var indicate variance, first calculating variance and the first customized parameter
With, then by variance and two parameters and ratio be determined as integration percentage coefficient, the integration percentage coefficient is for adjusting image
The integration percentage of data each section.
Further, first time image co-registration, fusion can also be carried out for rgb image data by fusion formula (6)
Formula (6) is as follows:
Inew=(1-t) * E+t*Iold--- --- -- fusion formula (6)
Wherein, the IoldIndicate that original rgb image data, E indicate the mean value in each components range of each pixel (i.e.
Value filtering image), IwenThe first image co-registration image is indicated, according to every in mean value and integration percentage coefficient adjustment rgb image data
The pixel value of a pixel obtains the first image co-registration image Inew。
It should be noted that can also make other than carrying out first time image co-registration using above-mentioned mean variance algorithm
First time image co-registration, such as Gaussian Blur image fuzzy algorithmic approach, this implementation are realized with various other image fuzzy algorithmic approaches
Example does not make excessive limitation to specific algorithm type.
In a kind of specific example, referring to Fig. 6, the process for showing a kind of acquisition skin area step of the present embodiment is shown
It is intended to, carries out Face Detection for the first image co-registration image, obtain the skin area of the first image co-registration image, including following
Sub-step:
Sub-step S241 obtains R component, G component and the B component of each pixel in the first image co-registration image；
The pixel is determined as skin area when R component, G component and B component meet preset condition by sub-step S242.
Further, after carrying out first time image co-registration, each pixel in the first image co-registration image can be identified
R component, G component and B component；It is screened for the R component of each pixel, G component and B component, that is, judges each pixel
R component, G component and B component whether meet preset condition, the pixel for meeting preset condition is determined as skin area, traverse
After each pixel in the first image co-registration image, all pictures for meeting preset condition in the first image co-registration image are filtered out
Element is used as skin area.
For example, which may include r > 95, g > 40, b > 20, r > g, r > b and5 wherein, above-mentioned r, g, and b is that the RGB of pixel is each
The value in channel, because the rgb value of the colour of skin of usually people has certain rule (i.e. preset condition), which is by statistical
What class obtained, if the value in each channel the RGB of pixel meets above-mentioned preset condition, it can be said that the bright pixel belongs to skin area
A part.
In a kind of specific example, referring to Fig. 7, a kind of RGB image number of the acquisition of the present embodiment by beautification is shown
According to flow diagram, for the first image co-registration image skin area carry out second of image co-registration, obtain by beautification
Rgb image data, including following sub-step:
Sub-step S251 obtains the second customized parameter；
Sub-step S252 is adjusted according to pixel value of second customized parameter to skin area, obtains by beautification
Rgb image data.
In a kind of preferred embodiment, the second customized parameter α can also be obtained；According to the second customized parameter to skin
The pixel value in skin region is adjusted, and obtains the rgb image data by beautification.
Specifically, second of image co-registration, fusion formula can also be carried out for skin area by fusion formula (7)
(7) as follows:
In=(Iold*α+Inew* (255- α))/256-------- fusion formula (7)
Wherein, α indicates the second customized parameter, InIndicate the rgb image data by beautification, IoldIndicate original RGB
Image data, InewIndicate the first image co-registration image；By each channel RGB of each pixel in the rgb image data by beautification
Value be stored in an int type array element, so that subsequent conversion is at YUV color space.
In a kind of specific example, referring to Fig. 8, a kind of rgb image data by beautification of the present embodiment is shown simultaneously
Rgb image data Parallel transformation by beautification is final YUV image data by the flow diagram of row switch process, including
Following sub-step:
Sub-step S31 will pass through R component, the G of each pixel of the rgb image data of beautification according to the second preset rules
Component and B component Parallel transformation are corresponding Y-component, U component and V component；
Sub-step S32 obtains the index and V component of U component according to the number of each pixel, picture traverse and component length
Index；
Sub-step S33 stores Y-component, U component and V component to correspondence according to the index of the index of U component and V component
Array position, obtain final YUV image data.
It is further applicable in the embodiment of the present invention, obtains after the rgb image data of beautification, can also be converted
It for the image of yuv format, is transmitted convenient for data, specifically, the RGB of beautification can will be passed through by the second preset rules first
R component, G component and the B component Parallel transformation of each pixel of image data are corresponding Y-component, U component and V component；
For example, which is conversion formula (2), and the conversion formula (2) is as follows
Y, U, V in conversion formula (2) respectively indicate the value of the value of Y-component, the value of U component and V component；And the conversion is public
R, G, B in formula (2) respectively indicate the value of the value of R component, the value of G component and B component；
It, can number according to each pixel, image after the value, the value of U component and the value of V component for obtaining the Y-component
Width and component length obtain the index of U component and the index of V component, and specifically, the index and V for obtaining above-mentioned U component divide
The index of amount can be realized by addressing calculation below:
Int U_idx=((i/w)/2) * (w/2)+((i%w)/2)+longth；
Int V_idx=U_idx+longth/4；
Wherein, which indicates the ith pixel of image；U_idx and V_idx respectively indicates U in the array of storage yuv data
The index of component and V component；Y-component, U component and V component are stored to correspondence according to the index of the index of U component and V component
Array position, obtain final YUV image data.
Referring to Fig. 9, a kind of schematic diagram of YUV image data of the present embodiment is shown, as shown in figure 9, the personage in figure
Skin area is unprocessed, more coarse；Referring to Fig.1 0, show a kind of signal of final YUV image data of the present embodiment
Figure, as shown in Figure 10, for the skin area of the personage in figure through image landscaping treatment, display effect is preferable.
Referring to Fig.1 1, the schematic diagram of another YUV image data of the present embodiment is shown, as shown in figure 11, in figure
Figure skin region is unprocessed, more coarse；Referring to Fig.1 2, show the final YUV image data of another kind of the present embodiment
Schematic diagram, as shown in figure 12, for the skin area of the personage in figure through image landscaping treatment, display effect is preferable.
It should be understood that although each step in the flow chart of Fig. 2-8 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-8
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in figure 13, a kind of image processing apparatus is provided, comprising: data acquisition module
301, replication module 302, the first Parallel transformation module 303, image beautification module 304, the second Parallel transformation module 305 and output
Module 306, in which:
Data acquisition module 301, for obtaining YUV image data；
Replication module 302, for being copied to the YUV image data in graphics processor from central processing unit memory
It deposits；
First Parallel transformation module 303, for turning the YUV image data parallel in the graphics processor memory
It is changed to rgb image data；
Image beautifies module 304, for carrying out image beautification operation to the rgb image data, obtains by beautification
Rgb image data；
Second Parallel transformation module 305, for being final YUV by the rgb image data Parallel transformation by beautification
Image data；
Output module 306, for exporting the final YUV image data.
The first Parallel transformation module includes: in one of the embodiments,
Computational submodule is indexed, is used in the graphics processor memory according to each picture in the YUV image data
Number, picture traverse and the component length gauge of element calculate the index of the U component of each pixel and the index of V component；
Component obtains submodule, for obtaining according to the number of each pixel, the index of U component and the index of V component
Obtain the corresponding Y-component of each pixel, U component and V component；
Rgb image data obtains submodule, for according to the first preset rules by the corresponding Y-component of each pixel, U
Component and V component Parallel transformation are the corresponding R component of each pixel, G component and B component, and by the R component, G component and B
Component is stored into corresponding array position, obtains the rgb image data.
Described image beautification module includes: in one of the embodiments,
Computational submodule, for calculating the mean value and quadratic sum of each pixel in the rgb image data；
Variance obtains submodule, for obtaining the variance of each pixel according to the quadratic sum；
First fusion submodule is obtained for carrying out first time image co-registration according to the mean and variance of each pixel
Obtain the first image co-registration image；
Face Detection submodule obtains described first for carrying out Face Detection for the first image blending image
The skin area of image co-registration image；
Second fusion submodule, carries out second of image for the skin area for the first image blending image and melts
It closes, obtains the rgb image data by beautification.
The first fusion submodule includes: in one of the embodiments,
First gain of parameter unit, for obtaining the first customized parameter；
Coefficient obtaining unit obtains integration percentage for the variance and the first customized parameter according to each pixel
Coefficient；
Blending image obtaining unit, for according in the mean value and the integration percentage coefficient adjustment rgb image data
The pixel value of each pixel obtains the first image blending image.
The Face Detection submodule includes: in one of the embodiments,
Component obtaining unit, for obtaining the R component of each pixel in the first image blending image, G component and B points
Amount；
Skin area determination unit, for when the R component, G component and B component meet preset condition, by the pixel
It is determined as skin area.
The second fusion submodule includes: in one of the embodiments,
Second gain of parameter unit, for obtaining the second customized parameter；
Unit is adjusted to obtain for the pixel value of the skin area to be adjusted according to second customized parameter
To the rgb image data by beautification.
The second Parallel transformation module includes: in one of the embodiments,
Parallel transformation submodule, for passing through each of the rgb image data beautified for described according to the second preset rules
R component, G component and the B component Parallel transformation of pixel are corresponding Y-component, U component and V component；
Index obtains submodule, for obtaining U component according to the number of each pixel, picture traverse and component length
Index and V component index；
Image data obtains submodule, for according to the index of the index of the U component and V component by the Y-component, U
Component and V component are stored to corresponding array position, obtain the final YUV image data.
Specific about image processing apparatus limits the restriction that may refer to above for image processing method, herein not
It repeats again.Modules in above-mentioned image processing apparatus can be realized fully or partially through software, hardware and combinations thereof.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, the corresponding operation of the above modules is executed in order to which processor calls.
The image processing apparatus of above-mentioned offer can be used for executing the image processing method that above-mentioned any embodiment provides, and have
Corresponding function and beneficial effect.
In one embodiment, a kind of computer equipment is provided, which can be terminal, internal structure
Figure can be as shown in figure 14.The computer equipment includes the processor connected by system bus, memory, network interface, shows
Display screen and input unit.Wherein, the processor of the computer equipment is for providing calculating and control ability.The computer equipment
Memory includes non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system and computer
Program.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The meter
The network interface for calculating machine equipment is used to communicate with external terminal by network connection.When the computer program is executed by processor
To realize a kind of image processing method.The display screen of the computer equipment can be liquid crystal display or electric ink is shown
Screen, the input unit of the computer equipment can be the touch layer covered on display screen, be also possible on computer equipment shell
Key, trace ball or the Trackpad of setting can also be external keyboard, Trackpad or mouse etc..
It will be understood by those skilled in the art that structure shown in Figure 14, only part relevant to application scheme
The block diagram of structure, does not constitute the restriction for the computer equipment being applied thereon to application scheme, and specific computer is set
Standby may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, is stored in memory
Computer program, the processor perform the steps of when executing computer program
Obtain YUV image data；
The YUV image data are copied to graphics processor memory from central processing unit memory；
The YUV image data parallel is converted into rgb image data in the graphics processor memory；
Image beautification operation is carried out to the rgb image data, obtains the rgb image data by beautification；
It is final YUV image data by the rgb image data Parallel transformation by beautification；
Export the final YUV image data.
In one embodiment, it is also performed the steps of when processor executes computer program
In the graphics processor memory according to the number of each pixel in the YUV image data, picture traverse and
Component length gauge calculates the index of the U component of each pixel and the index of V component；
According to the number of each pixel, the index of U component and the index of V component, it is Y points corresponding to obtain each pixel
Amount, U component and V component；
It by the corresponding Y-component of each pixel, U component and V component Parallel transformation is each according to the first preset rules
The corresponding R component of pixel, G component and B component, and the R component, G component and B component are stored into corresponding array position,
Obtain the rgb image data.
In one embodiment, it is also performed the steps of when processor executes computer program
Calculate the mean value and quadratic sum of each pixel in the rgb image data；
The variance of each pixel is obtained according to the quadratic sum；
First time image co-registration is carried out according to the mean and variance of each pixel, obtains the first image co-registration image；
Face Detection is carried out for the first image blending image, obtains the skin region of the first image blending image
Domain；
Second of image co-registration is carried out for the skin area of the first image blending image, is obtained described by beautification
Rgb image data.
In one embodiment, it is also performed the steps of when processor executes computer program
Obtain the first customized parameter；
According to the variance and the first customized parameter of each pixel, integration percentage coefficient is obtained；
According to the pixel value of each pixel in the mean value and the integration percentage coefficient adjustment rgb image data, obtain
The first image blending image.
In one embodiment, it is also performed the steps of when processor executes computer program
Obtain R component, G component and the B component of each pixel in the first image blending image；
When the R component, G component and B component meet preset condition, which is determined as skin area.
In one embodiment, it is also performed the steps of when processor executes computer program
Obtain the second customized parameter；
The pixel value of the skin area is adjusted according to second customized parameter, is obtained described by beautification
Rgb image data.
In one embodiment, it is also performed the steps of when processor executes computer program
According to the second preset rules by it is described by beautification rgb image data each pixel R component, G component and B
Component in parallel is converted to corresponding Y-component, U component and V component；
The index of U component and the rope of V component are obtained according to the number of each pixel, picture traverse and component length
Draw；
The Y-component, U component and V component are stored to corresponding according to the index of the index of the U component and V component
Array position obtains the final YUV image data.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Obtain YUV image data；
The YUV image data are copied to graphics processor memory from central processing unit memory；
The YUV image data parallel is converted into rgb image data in the graphics processor memory；
Image beautification operation is carried out to the rgb image data, obtains the rgb image data by beautification；
It is final YUV image data by the rgb image data Parallel transformation by beautification；
Export the final YUV image data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
In the graphics processor memory according to the number of each pixel in the YUV image data, picture traverse and
Component length gauge calculates the index of the U component of each pixel and the index of V component；
According to the number of each pixel, the index of U component and the index of V component, it is Y points corresponding to obtain each pixel
Amount, U component and V component；
It by the corresponding Y-component of each pixel, U component and V component Parallel transformation is each according to the first preset rules
The corresponding R component of pixel, G component and B component, and the R component, G component and B component are stored into corresponding array position,
Obtain the rgb image data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Calculate the mean value and quadratic sum of each pixel in the rgb image data；
The variance of each pixel is obtained according to the quadratic sum；
First time image co-registration is carried out according to the mean and variance of each pixel, obtains the first image co-registration image；
Face Detection is carried out for the first image blending image, obtains the skin region of the first image blending image
Domain；
Second of image co-registration is carried out for the skin area of the first image blending image, is obtained described by beautification
Rgb image data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Obtain the first customized parameter；
According to the variance and the first customized parameter of each pixel, integration percentage coefficient is obtained；
According to the pixel value of each pixel in the mean value and the integration percentage coefficient adjustment rgb image data, obtain
The first image blending image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Obtain R component, G component and the B component of each pixel in the first image blending image；
When the R component, G component and B component meet preset condition, which is determined as skin area.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Obtain the second customized parameter；
The pixel value of the skin area is adjusted according to second customized parameter, is obtained described by beautification
Rgb image data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
According to the second preset rules by it is described by beautification rgb image data each pixel R component, G component and B
Component in parallel is converted to corresponding Y-component, U component and V component；
The index of U component and the rope of V component are obtained according to the number of each pixel, picture traverse and component length
Draw；
The Y-component, U component and V component are stored to corresponding according to the index of the index of the U component and V component
Array position obtains the final YUV image data.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of image processing method characterized by comprising
Obtain YUV image data；
The YUV image data are copied to graphics processor memory from central processing unit memory；
The YUV image data parallel is converted into rgb image data in the graphics processor memory；
Image beautification operation is carried out to the rgb image data, obtains the rgb image data by beautification；
It is final YUV image data by the rgb image data Parallel transformation by beautification；
Export the final YUV image data.
2. the method according to claim 1, wherein it is described in the graphics processor memory by the YUV
Image data Parallel transformation is rgb image data, comprising:
According to number, picture traverse and the component of each pixel in the YUV image data in the graphics processor memory
Length gauge calculates the index of the U component of each pixel and the index of V component；
According to the number of each pixel, the index of U component and the index of V component, the corresponding Y-component of each pixel, U are obtained
Component and V component；
It by the corresponding Y-component of each pixel, U component and V component Parallel transformation is each pixel according to the first preset rules
Corresponding R component, G component and B component, and the R component, G component and B component are stored into corresponding array position, it obtains
The rgb image data.
3. the method according to claim 1, wherein described carry out image beautification behaviour to the rgb image data
Make, obtain the rgb image data by beautification, comprising:
Calculate the mean value and quadratic sum of each pixel in the rgb image data；
The variance of each pixel is obtained according to the quadratic sum；
First time image co-registration is carried out according to the mean and variance of each pixel, obtains the first image co-registration image；
Face Detection is carried out for the first image blending image, obtains the skin area of the first image blending image；
Second of image co-registration is carried out for the skin area of the first image blending image, obtains the process beautification
Rgb image data.
4. according to the method described in claim 3, it is characterized in that, described carry out according to the mean and variance of each pixel
First time image co-registration obtains the first image co-registration image, comprising:
Obtain the first customized parameter；
According to the variance and the first customized parameter of each pixel, integration percentage coefficient is obtained；
According to the pixel value of each pixel in the mean value and the integration percentage coefficient adjustment rgb image data, obtain described
First image co-registration image.
5. according to the method described in claim 3, it is characterized in that, described carry out the colour of skin for the first image blending image
Detection, obtains the skin area of the first image blending image, comprising:
Obtain R component, G component and the B component of each pixel in the first image blending image；
When the R component, G component and B component meet preset condition, which is determined as skin area.
6. according to the method described in claim 3, it is characterized in that, the skin region for the first image blending image
Domain carries out second of image co-registration, obtains the rgb image data by beautification, comprising:
Obtain the second customized parameter；
The pixel value of the skin area is adjusted according to second customized parameter, obtains the process beautification
Rgb image data.
7. the method according to claim 1, wherein described that the rgb image data by beautification is parallel
Be converted to final YUV image data, comprising:
According to the second preset rules by it is described by beautification rgb image data each pixel R component, G component and B component
Parallel transformation is corresponding Y-component, U component and V component；
The index of U component and the index of V component are obtained according to the number of each pixel, picture traverse and component length；
The Y-component, U component and V component are stored to corresponding array according to the index of the index of the U component and V component
Position obtains the final YUV image data.
8. a kind of image processing apparatus characterized by comprising
Data acquisition module, for obtaining YUV image data；
Replication module, for the YUV image data to be copied to graphics processor memory from central processing unit memory；
First Parallel transformation module, for the YUV image data parallel to be converted to RGB in the graphics processor memory
Image data；
Image beautifies module, for carrying out image beautification operation to the rgb image data, obtains the RGB image by beautification
Data；
Second Parallel transformation module, for being final YUV image number by the rgb image data Parallel transformation by beautification
According to；
Output module, for exporting the final YUV image data.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the processor realizes image processing method described in any one of claims 1 to 7 when executing the computer program
Step.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of image processing method described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910158997.6A CN109934783B (en) | 2019-03-04 | 2019-03-04 | Image processing method, image processing device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910158997.6A CN109934783B (en) | 2019-03-04 | 2019-03-04 | Image processing method, image processing device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109934783A true CN109934783A (en) | 2019-06-25 |
| CN109934783B CN109934783B (en) | 2021-05-07 |
Family
ID=66986145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910158997.6A Active CN109934783B (en) | 2019-03-04 | 2019-03-04 | Image processing method, image processing device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109934783B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112150393A (en) * | 2020-10-12 | 2020-12-29 | 深圳数联天下智能科技有限公司 | Face image buffing method and device, computer equipment and storage medium |
| CN113538215A (en) * | 2021-06-11 | 2021-10-22 | 展讯半导体(成都)有限公司 | Image format conversion method, device and system, electronic equipment and storage medium |
| CN114928730A (en) * | 2022-06-23 | 2022-08-19 | 湖南国科微电子股份有限公司 | Image processing method and image processing apparatus |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0793563A (en) * | 1993-09-21 | 1995-04-07 | Canon Inc | Image processor |
| US20160225179A1 (en) * | 2015-01-29 | 2016-08-04 | Institute Of Environmental Science And Research Limited | Three-dimensional visualization of a scene or environment |
| CN105976309A (en) * | 2016-05-03 | 2016-09-28 | 成都索贝数码科技股份有限公司 | High-efficiency and easy-parallel implementation beauty mobile terminal |
| CN105976308A (en) * | 2016-05-03 | 2016-09-28 | 成都索贝数码科技股份有限公司 | GPU-based mobile terminal high-quality beauty real-time processing method |
| CN106447606A (en) * | 2016-10-31 | 2017-02-22 | 南京维睛视空信息科技有限公司 | Rapid real-time video beautifying method |
| CN107770446A (en) * | 2017-10-31 | 2018-03-06 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and electronic equipment |
| CN107911625A (en) * | 2017-11-30 | 2018-04-13 | 广东欧珀移动通信有限公司 | Light measuring method, device, readable storage medium storing program for executing and computer equipment |
| CN108200347A (en) * | 2018-01-30 | 2018-06-22 | 努比亚技术有限公司 | A kind of image processing method, terminal and computer readable storage medium |
-
2019
- 2019-03-04 CN CN201910158997.6A patent/CN109934783B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0793563A (en) * | 1993-09-21 | 1995-04-07 | Canon Inc | Image processor |
| US20160225179A1 (en) * | 2015-01-29 | 2016-08-04 | Institute Of Environmental Science And Research Limited | Three-dimensional visualization of a scene or environment |
| CN105976309A (en) * | 2016-05-03 | 2016-09-28 | 成都索贝数码科技股份有限公司 | High-efficiency and easy-parallel implementation beauty mobile terminal |
| CN105976308A (en) * | 2016-05-03 | 2016-09-28 | 成都索贝数码科技股份有限公司 | GPU-based mobile terminal high-quality beauty real-time processing method |
| CN106447606A (en) * | 2016-10-31 | 2017-02-22 | 南京维睛视空信息科技有限公司 | Rapid real-time video beautifying method |
| CN107770446A (en) * | 2017-10-31 | 2018-03-06 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and electronic equipment |
| CN107911625A (en) * | 2017-11-30 | 2018-04-13 | 广东欧珀移动通信有限公司 | Light measuring method, device, readable storage medium storing program for executing and computer equipment |
| CN108200347A (en) * | 2018-01-30 | 2018-06-22 | 努比亚技术有限公司 | A kind of image processing method, terminal and computer readable storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112150393A (en) * | 2020-10-12 | 2020-12-29 | 深圳数联天下智能科技有限公司 | Face image buffing method and device, computer equipment and storage medium |
| CN113538215A (en) * | 2021-06-11 | 2021-10-22 | 展讯半导体(成都)有限公司 | Image format conversion method, device and system, electronic equipment and storage medium |
| CN113538215B (en) * | 2021-06-11 | 2022-12-27 | 展讯半导体(成都)有限公司 | Image format conversion method, device and system, electronic equipment and storage medium |
| CN114928730A (en) * | 2022-06-23 | 2022-08-19 | 湖南国科微电子股份有限公司 | Image processing method and image processing apparatus |
| CN114928730B (en) * | 2022-06-23 | 2023-08-22 | 湖南国科微电子股份有限公司 | Image processing method and image processing apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109934783B (en) | 2021-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9467704B2 (en) | Adaptive ratio images in HDR image representation | |
| CN107169927B (en) | Image processing system, method and display device | |
| CN109934783A (en) | Image processing method, device, computer equipment and storage medium | |
| CN106485720A (en) | Image processing method and device | |
| EP3809698A1 (en) | Video signal processing method and apparatus | |
| CN113168669B (en) | Image processing method, device, electronic equipment and readable storage medium | |
| WO2021148057A1 (en) | Method and apparatus for generating low bit width hdr image, storage medium, and terminal | |
| US20230351561A1 (en) | Method and device for obtaining a second image from a first image when the dynamic range of the luminance of the first image is greater than the dynamic range of the luminance of the second image | |
| CN111127342B (en) | Image processing method, device, storage medium and terminal equipment | |
| WO2023010750A1 (en) | Image color mapping method and apparatus, electronic device, and storage medium | |
| US20240205376A1 (en) | Image processing method and apparatus, computer device, and storage medium | |
| WO2021073330A1 (en) | Video signal processing method and apparatus | |
| CN113132696A (en) | Image tone mapping method, device, electronic equipment and storage medium | |
| CN114298889A (en) | Image processing circuit and image processing method | |
| CN114501023A (en) | Video processing method and device, computer equipment and storage medium | |
| CN111860363A (en) | Video image processing method and device, electronic equipment and storage medium | |
| WO2021073316A1 (en) | Image signal conversion method and apparatus, and terminal device | |
| CN117768774A (en) | Image processor, image processing method, photographing device and electronic device | |
| US20230171416A1 (en) | Video encoding method, video encoding device, and image processing device | |
| CN114299193B (en) | Black-white video coloring method, system, equipment and storage medium based on neural network | |
| CN113688752B (en) | Training method, device, equipment and storage medium for face color detection model | |
| KR100930483B1 (en) | Video Format Converter | |
| CN114584831A (en) | Video optimization processing method, device, equipment and storage medium for improving video definition | |
| Liu et al. | HDRC: a subjective quality assessment database for compressed high dynamic range image | |
| CN112532957B (en) | Image processing method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |