CN105898151A - Image processing method and device - Google Patents
Image processing method and device Download PDFInfo
- Publication number
- CN105898151A CN105898151A CN201510782728.9A CN201510782728A CN105898151A CN 105898151 A CN105898151 A CN 105898151A CN 201510782728 A CN201510782728 A CN 201510782728A CN 105898151 A CN105898151 A CN 105898151A
- Authority
- CN
- China
- Prior art keywords
- grid
- image
- value
- denoising
- carries out
- 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 24
- 238000000034 method Methods 0.000 claims description 39
- 206010034960 Photophobia Diseases 0.000 claims description 15
- 208000013469 light sensitivity Diseases 0.000 claims description 15
- 230000003287 optical effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 8
- 230000035945 sensitivity Effects 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000001965 increasing effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
- H04N25/62—Detection or reduction of noise due to excess charges produced by the exposure, e.g. smear, blooming, ghost image, crosstalk or leakage between pixels
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/73—Circuitry for compensating brightness variation in the scene by influencing the exposure time
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
- Studio Devices (AREA)
Abstract
The invention provides an image processing method and device. An object in an image to be processed is identified, and mesh generation is carried out. The sensitivity ISO value of each mesh in a region covered by each object is acquired. The average value of the ISO value of each mesh in the region covered by each object is calculated. The number of first meshes in the region covered by each object is counted. When the preset number of meshes is exceeded, an exposure index corresponding to each first mesh is used to denoise an image corresponding to the first mesh. An exposure index corresponding to the average value is used to denoise images corresponding to other meshes except the first mesh. A target image is acquired. According to the invention, the image denoising effect is great through local dynamic denoising.
Description
Technical field
The present invention relates to electronic technology field, particularly relate to a kind of image processing method and device.
Background technology
Smart mobile phone has gradually been dissolved among people's daily life at present, not only becomes daily communication
Equipment, also becomes the daily amusement equipment being easy to carry about with one.In smart mobile phone, the configuration of camera is increasingly
Height, as being provided with the performances such as large aperture, high pixel and optical anti-vibration, and based on smart mobile phone
Portability, user increasingly likes being taken pictures by the camera on smart mobile phone.Although constantly
Ground carries out configuration optimization to the camera on smart mobile phone, but there is noise still on the image of shooting
Unavoidable.
At present, mobile phone based on Android system can use high pass algorithm that image is carried out denoising, tool
Every image to be captured is all resolved into the grid of 64 × 48 by body ground, calculates all these grid
Average brightness, to determine exposure index, compensates by clear zone, and brightness value exceeds necessarily than meansigma methods
The smaller area of quantity can be left in the basket when determining exposure index.This high pass algorithm is according to different increasings
Denoising parameter is divided into the region that 6 groups i.e. 6 are different by benefit (gain) value, and these 6 regions are by mutually
Phase interpolation covers the brightest to the darkest all scenes.First one pictures is judged to its digital phase
Machine light sensitivitys quantifies regulation (International Standardization Organization is called for short ISO)
It is how much to be exposure index, then carries out denoising according to the denoising parameter of corresponding region.High pass algorithm
Go to judge simply by whole picture, then carry out denoising by unified denoising parameter, in image
Bright its denoising effect of the part secretly differed greatly is poor.
Summary of the invention
The present invention provides a kind of image processing method and device, for solving by high pass algorithm figure
Exist when the problem that in image, bright its denoising effect of the part secretly differed greatly is poor during as carrying out denoising.
To achieve these goals, the invention provides a kind of image processing method, including:
Identify the object comprised in pending image and carry out stress and strain model；
Obtain the light sensitivitys ISO value of each grid in each object institute overlay area；
The meansigma methods of each grid ISO value in calculating each object overlay area；
The number of the first grid in the statistics each object area of coverage of acquisition；Wherein, described first grid is
The described ISO value that in each object area of coverage, grid is corresponding exceeds preset value with the difference of described meansigma methods
All grids in any one；
If described number is beyond the meshes number preset, obtain each described first grid corresponding
Described exposure value number；
Use the described exposure index that each described first grid is corresponding corresponding to described first grid
Image section carries out denoising, and uses described exposure index corresponding to described meansigma methods to except described
The image section that other grids outside first grid are corresponding carries out denoising, obtains described target image.
To achieve these goals, the invention provides a kind of image processing apparatus, including:
Identify and divide module, for identifying the object comprised in pending image and carrying out stress and strain model；
First acquisition module, light sensitivitys ISO of each grid in obtaining each object institute overlay area
Value；
Computing module, the meansigma methods of each grid ISO value in calculating each object overlay area；
Second acquisition module, the number of the first grid in adding up each object area of coverage of acquisition；Its
In, described first grid be grid is corresponding in each object area of coverage described ISO value with described averagely
The difference of value is beyond any one in all grids of preset value；
3rd acquisition module, if for described number beyond the meshes number preset, obtaining each
The exposure value number that described first grid is corresponding；
Denoising module, for using described exposure index that each described first grid is corresponding to described
The image section that first grid is corresponding carries out denoising, and uses described exposure corresponding to described meansigma methods
Optical index carries out denoising to the image section that other grids in addition to described first grid are corresponding,
To target image.
The image processing method of the present invention and device, by identifying the object comprised in pending image
And carry out stress and strain model, obtain the light sensitivitys ISO value of each grid in each object institute overlay area,
The meansigma methods of each grid ISO value in calculating each object overlay area, statistics obtains each object and covers
The number of the first grid in district, when beyond the meshes number preset, uses each described first net
The described exposure index that lattice are corresponding carries out denoising to the image section that described first grid is corresponding, and
Use described exposure index corresponding to described meansigma methods to other grids in addition to described first grid
Corresponding image section carries out denoising, obtains target image.The present invention has evaded original to whole figure
Picture only makes the problem that bright dark place noise level is inconsistent with an exposure index denoising, and leads to
Crossing local dynamic station denoising makes the denoising effect of image preferable.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the image processing method of the embodiment of the present invention one；
Fig. 2 is the schematic flow sheet of the image processing method of the embodiment of the present invention two；
Fig. 3 is the schematic flow sheet of the image processing method of the embodiment of the present invention three；
Fig. 4 is the structural representation of the image processing apparatus of the embodiment of the present invention four.
Detailed description of the invention
The image processing method and the device that there is provided the embodiment of the present invention below in conjunction with the accompanying drawings are carried out
Describe in detail.
Embodiment one
As it is shown in figure 1, the schematic flow sheet of the image processing method that it is the embodiment of the present invention one,
This image processing method includes:
Step 101, identify the object comprised in pending image and carry out stress and strain model.
Before pending image is carried out denoising, image can be carried out object identification.Specifically,
First pass through the substantial amounts of subject image with obvious Ha Er (Haar) feature, use mould
Formula is known method for distinguishing and is trained obtaining grader, permissible by haar feature based on this grader
Identify objects different in pending image.After different objects in identifying pending image,
This image division can be become the grid of 64 × 48.
Step 102, obtain the light sensitivitys ISO value of each grid in each object institute overlay area.
Step 103, calculate each object overlay area in the meansigma methods of each grid ISO value.
The number of the first grid in step 104, the statistics each object area of coverage of acquisition.
Wherein, described ISO value that in described first grid is each object area of coverage, grid is corresponding and institute
State the difference of meansigma methods beyond any one in all grids of preset value.
In calculating each object overlay area after the meansigma methods of each grid ISO value, by each object
In overlay area, the ISO value of each grid and meansigma methods do difference, if the difference that a grid is corresponding
This grid, beyond preset value, is become the first grid by value.Wherein, the first grid is that each object covers
Any beyond in all grids of preset value of the ISO value that in district, grid is corresponding and the difference of meansigma methods
One.Further, the number of the first grid in each object is added up.
If the described number of step 105 is beyond the meshes number preset, obtain each described first net
The described exposure value number that lattice are corresponding.
The number that statistics gets the first grid in each object covers is entered with the meshes number preset
Row compares, and when this number is beyond default meshes number, illustrates that these first grids are corresponding
Image section is relatively big with the luminance difference of other grids, needs to obtain the exposure that each first grid is corresponding
Optical index, in the present embodiment, chooses corresponding exposure according to the ISO value that each first grid is corresponding
Index, the exposure index being then based on each first grid corresponding carries out list to corresponding image section
Solely exposure.
Step 106, use described exposure index that each described first grid is corresponding to described first net
The image section that lattice are corresponding carries out denoising, and uses described exposure index corresponding to described meansigma methods
The image section that other grids in addition to described first grid are corresponding is carried out denoising, obtains described
Target image.
In order to the denoising effect making pending image is preferable, the present embodiment is to the first grid with except first
Other grids outside grid carry out denoising respectively, to obtain target image.Specifically, make
The exposure index corresponding with each first grid carries out denoising to the image section that the first grid is corresponding,
And use the image that the meansigma methods calculated by each grid ISO value is corresponding to other grids
Part carries out denoising, obtains target image.
The image processing method that the present embodiment provides, by identifying the object comprised in pending image
And carry out stress and strain model, obtain the light sensitivitys ISO value of each grid in each object institute overlay area,
The meansigma methods of each grid ISO value in calculating each object overlay area, statistics obtains each object and covers
The number of the first grid in district, when beyond the meshes number preset, uses each described first net
The described exposure index that lattice are corresponding carries out denoising to the image section that described first grid is corresponding, and
Use described exposure index corresponding to described meansigma methods to other grids in addition to described first grid
Corresponding image section carries out denoising, obtains target image.The present invention has evaded original to whole figure
As only making the inconsistent problem of bright dark place noise level with an exposure index denoising, and pass through
Local dynamic station denoising makes the denoising effect of image preferable.
Embodiment two
As in figure 2 it is shown, the schematic flow sheet of the image processing method that it is the embodiment of the present invention two,
This image processing method includes:
Step 201, identify the object comprised in pending image and carry out stress and strain model.
Step 202, obtain the light sensitivitys ISO value of each grid in each object institute overlay area.
Step 203, calculate each object overlay area in the meansigma methods of each grid ISO value.
The number of the first grid in step 204, the statistics each object area of coverage of acquisition.
Step 101～step during in step 201～step 204, related content can be found in above-described embodiment one
Record in 104, step repeats herein.
If the described number of step 205 is without departing from default meshes number, obtain each object corresponding
The exposure index that described meansigma methods is corresponding.
Step 206, use the described exposure index of each object that each self-corresponding image section is gone
Make an uproar, obtain target image.
Further, in the number of the first grid without departing from default network number, illustrate pending
The image section that in image, the first grid is corresponding is less, in order to improve the efficiency that image denoising processes,
Each self-corresponding image section is entered by the exposure index that the meansigma methods of each object ISO can be used corresponding
Row denoising, obtains target image.
The image processing method that the present embodiment provides, by identifying the object comprised in pending image
And carry out stress and strain model, obtain the light sensitivitys ISO value of each grid in each object institute overlay area,
The meansigma methods of each grid ISO value in calculating each object overlay area, statistics obtains each object and covers
The number of the first grid in district, when without departing from default meshes number, uses and obtains each object pair
The described exposure index that the described meansigma methods answered is corresponding, uses the described exposure index of each object to respectively
Self-corresponding image section carries out denoising, obtains target image.The present embodiment is at pending image
In the case of bright dark difference is less, it is possible to use the exposure index that the ISO meansigma methods that gets is corresponding
Image is carried out denoising, is possible not only to obtain preferable denoising effect, and processing speed is relatively
Hurry up.
Embodiment three
As it is shown on figure 3, the schematic flow sheet of the image processing method that it is the embodiment of the present invention three,
This image processing method includes:
Step 301, pending image is carried out stress and strain model.
Step 302, obtain the light sensitivitys ISO value of each grid.
Step 303, described ISO value according to each grid choose corresponding described exposure index.
Step 304, the described exposure value number using each grid corresponding carry out denoising to described image and obtain
To described target image.
In the present embodiment, whole pending image is divided into the grid of multiple 64 × 48, then calculates
The ISO value of each grid, after getting the ISO value of each grid, selects according to this ISO value
Corresponding exposure index.After getting the exposure index of each grid, use this exposure index to often
The image section that individual grid is corresponding carries out denoising, to obtain target image.In the present embodiment,
No longer carrying out calculating the average brightness of whole pending image, each grid is according to the exposure of oneself
Index carries out denoising respectively, it is to avoid in picture, the denoising of Liang Chu region causes the most by force obscuring, dark place district
Denoising the most weak noise in territory is big, and such denoising effect is finer and smoother.
Embodiment four
As shown in Figure 4, it is the structural representation of image processing apparatus of the embodiment of the present invention four.
This device includes: identifies and divides module the 11, first acquisition module 12, computing module 13, second
Acquisition module the 14, the 3rd acquisition module 15 and denoising module 16.
Specifically, identify and divide module 11, for identifying that the object comprised in pending image is gone forward side by side
Row stress and strain model.
First acquisition module 12, the light sensitivitys of each grid in obtaining each object institute overlay area
ISO value.
Computing module 13, the meansigma methods of each grid ISO value in calculating each object overlay area.
Second acquisition module 14, the number of the first grid in adding up each object area of coverage of acquisition.
Wherein, the described ISO value that in described first grid is each object area of coverage, grid is corresponding is flat with described
The difference of average is beyond any one in all grids of preset value.
3rd acquisition module 15, if for described number beyond the meshes number preset, obtains every
The exposure value number that individual described first grid is corresponding.
Denoising module 16, for using described exposure index that each described first grid is corresponding to institute
State image section corresponding to the first grid and carry out denoising, and use corresponding described of described meansigma methods
Exposure index carries out denoising to the image section that other grids in addition to described first grid are corresponding,
Obtain target image.
Further, described 3rd acquisition module 15, if being additionally operable to described number without departing from described
Meshes number, obtains the described exposure value number that described meansigma methods corresponding to each object is corresponding.
Described denoising module 16, is also used for the described exposure index of each object to each self-corresponding
Image section carries out denoising, obtains described target image.
Further, described first acquisition module 12, it is additionally operable to obtaining each object institute overlay area
After the light sensitivitys ISO value of interior each grid, choose accordingly according to the described ISO value of each grid
Described exposure index.
Described denoising module 16, is also used for described exposure value number corresponding to each grid to described
Image carries out denoising and obtains described target image.
Further, described identification divides module 11, specifically for using Ha Er Haar feature to treat
Process image, to get the object comprised in described pending image, by described pending image
It is divided into the grid of 64 × 48
Each functional module of the image processing apparatus that the present embodiment provides can be used for performing Fig. 1, figure
The flow process of the image processing method shown in 2 and Fig. 3, its specific works principle repeats no more, in detail
See the description of embodiment of the method.
The image processing apparatus that the present embodiment provides, is entered target to be captured by imageing sensor
Row data acquisition, identifies each scenery included in target from the data gathered, and obtains each
The profile information that scenery is corresponding, carries out region division according to the profile information of each scenery to target, adjusts
Distance between whole imageing sensor and target, obtains the one-tenth that definition corresponding to each region is optimal
Image position, extracts the area image on the image space corresponding to each region, by each region institute
Corresponding area image merges, and obtains the target image of target.The present embodiment is taken turns by scenery
Exterior feature carries out region division to target, and the area image the most clearly in each region takes out synthesis target
Final image so that being ultimately imaged and become apparent from sharp keen, and the scenery of target internal all can be clear
Present.
One of ordinary skill in the art will appreciate that: realize the whole of above-mentioned each method embodiment
Or part steps can be completed by the hardware that programmed instruction is relevant.Aforesaid program is permissible
It is stored in a computer read/write memory medium.This program upon execution, performs to include
State the step of each method embodiment；And aforesaid storage medium includes: ROM, RAM, magnetic
The various medium that can store program code such as dish or CD.
It is last it is noted that various embodiments above is only in order to illustrate technical scheme,
It is not intended to limit；Although the present invention being described in detail with reference to foregoing embodiments,
It will be understood by those within the art that: foregoing embodiments still can be remembered by it
The technical scheme carried is modified, or carries out the most some or all of technical characteristic
With replacing；And these amendments or replacement, do not make the essence of appropriate technical solution depart from this
Invent the scope of each embodiment technical scheme.
Claims (8)
1. an image processing method, it is characterised in that including:
Identify the object comprised in pending image and carry out stress and strain model；
Obtain the light sensitivitys ISO value of each grid in each object institute overlay area；
The meansigma methods of each grid ISO value in calculating each object overlay area；
The number of the first grid in the statistics each object area of coverage of acquisition；Wherein, described first grid is
The described ISO value that in each object area of coverage, grid is corresponding exceeds preset value with the difference of described meansigma methods
All grids in any one；
If described number is beyond the meshes number preset, obtain each described first grid corresponding
Described exposure value number；
Use the described exposure index that each described first grid is corresponding corresponding to described first grid
Image section carries out denoising, and uses described exposure index corresponding to described meansigma methods to except described
The image section that other grids outside first grid are corresponding carries out denoising, obtains described target image.
Image processing method the most according to claim 1, it is characterised in that also include:
If described number without departing from default meshes number, obtain each object corresponding described averagely
The described exposure value number that value is corresponding；
The described exposure index using each object carries out denoising to each self-corresponding image section, obtains
Target image.
Image processing method the most according to claim 1, it is characterised in that described acquisition is each
In object institute overlay area after the light sensitivitys ISO value of each grid, also include:
Described ISO value according to each grid chooses corresponding described exposure index；
The described exposure value number using each grid corresponding carries out denoising to described image and obtains described mesh
Logo image.
4. according to the arbitrary described image processing method of claim 1-3, it is characterised in that described
The region being covered object carries out stress and strain model and includes:
Use Ha Er Haar feature to pending image, comprise to get in described pending image
Object；
Described pending image division is become the grid of 64 × 48.
5. an image processing apparatus, it is characterised in that including:
Identify and divide module, for identifying the object comprised in pending image and carrying out stress and strain model；
First acquisition module, light sensitivitys ISO of each grid in obtaining each object institute overlay area
Value；
Computing module, the meansigma methods of each grid ISO value in calculating each object overlay area；
Second acquisition module, the number of the first grid in adding up each object area of coverage of acquisition；Its
In, described first grid be grid is corresponding in each object area of coverage described ISO value with described averagely
The difference of value is beyond any one in all grids of preset value；
3rd acquisition module, if for described number beyond the meshes number preset, obtaining each
The exposure value number that described first grid is corresponding；
Denoising module, for using described exposure index that each described first grid is corresponding to described
The image section that first grid is corresponding carries out denoising, and uses described exposure corresponding to described meansigma methods
Optical index carries out denoising to the image section that other grids in addition to described first grid are corresponding,
To target image.
Image processing apparatus the most according to claim 5, it is characterised in that the described 3rd obtains
Delivery block, if being additionally operable to described number without departing from described meshes number, obtains each object corresponding
The described exposure value number that described meansigma methods is corresponding；
Described denoising module, is also used for the described exposure index of each object to each self-corresponding figure
As part carries out denoising, obtain described target image.
Image processing apparatus the most according to claim 6, it is characterised in that described first obtains
Delivery block, be additionally operable in obtaining each object institute overlay area the light sensitivitys ISO value of each grid it
After, choose corresponding described exposure index according to the described ISO value of each grid；
Described denoising module, is also used for described exposure value number corresponding to each grid to described figure
Described target image is obtained as carrying out denoising.
8. according to the image processing apparatus described in any one of claim 5-7, it is characterised in that institute
State identification and divide module, specifically for use Ha Er Haar feature to pending image, to get
The object comprised in described pending image, becomes the grid of 64 × 48 by described pending image division.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510782728.9A CN105898151A (en) | 2015-11-15 | 2015-11-15 | Image processing method and device |
| PCT/CN2016/089024 WO2017080236A1 (en) | 2015-11-15 | 2016-07-07 | Image processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510782728.9A CN105898151A (en) | 2015-11-15 | 2015-11-15 | Image processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105898151A true CN105898151A (en) | 2016-08-24 |
Family
ID=57001805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510782728.9A Pending CN105898151A (en) | 2015-11-15 | 2015-11-15 | Image processing method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105898151A (en) |
| WO (1) | WO2017080236A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111131716A (en) * | 2019-12-31 | 2020-05-08 | 联想(北京)有限公司 | Image processing method and electronic device |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111754385B (en) * | 2019-03-26 | 2024-06-11 | 深圳中科飞测科技股份有限公司 | Data point model processing method and system, detection method and system and readable medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100066868A1 (en) * | 2008-09-12 | 2010-03-18 | Canon Kabushiki Kaisha | Image processing apparatus and method of processing image |
| US20110069881A1 (en) * | 2009-09-24 | 2011-03-24 | Canon Kabushiki Kaisha | Image processing apparatus and image processing method |
| CN104202524A (en) * | 2014-09-02 | 2014-12-10 | 三星电子（中国）研发中心 | Method and device for backlight filming |
| CN104902143A (en) * | 2015-05-21 | 2015-09-09 | 广东欧珀移动通信有限公司 | Resolution-ratio-based image de-noising method and device |
| CN105005973A (en) * | 2015-06-30 | 2015-10-28 | 广东欧珀移动通信有限公司 | Fast image denoising method and apparatus |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6267502B2 (en) * | 2013-12-10 | 2018-01-24 | キヤノン株式会社 | IMAGING DEVICE, IMAGING DEVICE CONTROL METHOD, AND PROGRAM |
-
2015
- 2015-11-15 CN CN201510782728.9A patent/CN105898151A/en active Pending
-
2016
- 2016-07-07 WO PCT/CN2016/089024 patent/WO2017080236A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100066868A1 (en) * | 2008-09-12 | 2010-03-18 | Canon Kabushiki Kaisha | Image processing apparatus and method of processing image |
| US20110069881A1 (en) * | 2009-09-24 | 2011-03-24 | Canon Kabushiki Kaisha | Image processing apparatus and image processing method |
| CN104202524A (en) * | 2014-09-02 | 2014-12-10 | 三星电子（中国）研发中心 | Method and device for backlight filming |
| CN104902143A (en) * | 2015-05-21 | 2015-09-09 | 广东欧珀移动通信有限公司 | Resolution-ratio-based image de-noising method and device |
| CN105005973A (en) * | 2015-06-30 | 2015-10-28 | 广东欧珀移动通信有限公司 | Fast image denoising method and apparatus |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111131716A (en) * | 2019-12-31 | 2020-05-08 | 联想(北京)有限公司 | Image processing method and electronic device |
| CN111131716B (en) * | 2019-12-31 | 2021-06-15 | 联想(北京)有限公司 | Image processing method and electronic device |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2017080236A1 (en) | 2017-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9330446B2 (en) | Method and apparatus for processing image | |
| CN105989367B (en) | Target Acquisition method and apparatus | |
| WO2017076040A1 (en) | Image processing method and device for use during continuous shooting operation | |
| CN107944403B (en) | Method and device for detecting pedestrian attribute in image | |
| CN108986097B (en) | Lens fogging state detection method, computer device and readable storage medium | |
| EP2709062B1 (en) | Image processing device, image processing method, and computer readable medium | |
| CN109961078B (en) | Image matching and splicing method, device, system and readable medium | |
| JP6135283B2 (en) | Image processing apparatus, image processing method, program, and recording medium | |
| CN108924427A (en) | A kind of video camera focus method, device and video camera | |
| US20190206117A1 (en) | Image processing method, intelligent terminal, and storage device | |
| CN104182721A (en) | Image processing system and image processing method capable of improving face identification rate | |
| CN108961302A (en) | Image processing method, device, mobile terminal and computer readable storage medium | |
| US11417003B2 (en) | Method and apparatus for tracking eyes of user and method of generating inverse-transform image | |
| CN107172354A (en) | Method for processing video frequency, device, electronic equipment and storage medium | |
| CN113065645A (en) | Twin attention network, image processing method and device | |
| CN113379609B (en) | Image processing method, storage medium and terminal equipment | |
| CN107147851A (en) | Photo processing method, device, computer-readable recording medium and electronic equipment | |
| CN105898151A (en) | Image processing method and device | |
| CN111800568B (en) | Light supplement method and device | |
| CN116563190B (en) | Image processing method, device, computer equipment and computer readable storage medium | |
| CN111292331B (en) | Image processing method and device | |
| CN105893578A (en) | Method and device for selecting photos | |
| JP4750758B2 (en) | Attention area extraction method, attention area extraction device, computer program, and recording medium | |
| CN111147693B (en) | Noise reduction method and device for full-size photographed image | |
| CN108062741B (en) | Binocular image processing method, imaging device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160824 |
|
| WD01 | Invention patent application deemed withdrawn after publication |