CN102542533A - Image blurring method and system - Google Patents
Image blurring method and system Download PDFInfo
- Publication number
- CN102542533A CN102542533A CN2010106069869A CN201010606986A CN102542533A CN 102542533 A CN102542533 A CN 102542533A CN 2010106069869 A CN2010106069869 A CN 2010106069869A CN 201010606986 A CN201010606986 A CN 201010606986A CN 102542533 A CN102542533 A CN 102542533A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- protection zone
- gray
- fuzzy
- 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
Images
Abstract
The invention relates to an image blurring method and system, belonging to the technical field of image processing. The method comprises the following steps: firstly obtaining a protection area in an original image, wherein the protection area refers to the area which does not need to be subjected to blurring treatment; than blurring treatment with edge protected is carried out on the original image, and marking the image after blurring treatment as an image M; and finally modifying pixel values of pixel points at corresponding positions in the image M according to pixel values of pixel points in the protection area in the original image. With the adoption of the image blurring method and the image blurring system, the smooth area of the image is effectively blurred, and detailed texture on the image is well maintained.
Description
Technical field
The invention belongs to technical field of image processing, be specifically related to a kind of image blurring method and system.
Background technology
What the image blurring processing at maintenance edge was commonly used at present mainly is the surface blur function among the Photoshop.Application algorithm more widely mainly is Bilateral filter (BLF) algorithm, and the surface blur function among the Photoshop also is based on this algorithm.The fuzzy algorithm that keeps the edge all is in doing Fuzzy Processing, the marginal information in the image to be added, and makes the point that is in strong edge receive less fuzzy influence.These methods all have strong edge in the image and keep effect preferably.
For example, the BLF algorithm utilizes the Gaussian filter of a two dimension that image is carried out filtering, among the Gauss respectively with color distance and space length as variable.Face a little color distortion and current point when differing greatly when certain, gaussian coefficient is just lower, and promptly this faces a little just less to the influence of current point.
Keep effect preferably though present maintenance edge fog algorithm all has the limbus in the image, because the edge of detail textures part is very short and small faint, the edge trend is very mixed and disorderly, so these methods all can't keep detail textures.The image that has a lot of detail textures like this for some, particularly to the Fuzzy Processing of human body or animal hair image, result all can be destroyed the hair details fully, and integral image obviously sensation is fuzzy, unintelligible, has a long way to go with the production demand of reality.In actual production, the technician uses the constituency usually image is protected, subarea processing.The quality that this method is image blurring is higher, but efficient is low, and cost is high.
Summary of the invention
To the defective that exists in the prior art, technical matters to be solved by this invention provides high, the effective image blurring method and system of a kind of efficient.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is following:
A kind of image blurring method may further comprise the steps:
(1) obtain protection zone in the original image, said protection zone is meant the zone that need not carry out Fuzzy Processing;
(2) original image is kept the Fuzzy Processing at edge, the image after the Fuzzy Processing is designated as image M;
(3) revise the pixel value of the pixel of same position in the said image M according to the pixel value of pixel in the said protection zone.
Aforesaid image blurring method, the acquisition methods of protection zone is following described in the step (1):
(a) original image is carried out bandpass filtering, obtain image D;
(b) image D is kept the Fuzzy Processing at edge, obtain image Ds;
(c), obtain image Dx to image Ds type of carrying out expansion process; Remaining area among the image Dx is the protection zone.
Aforesaid image blurring method, the process of type expansion process is following described in the step (c):
1. by pixel scan image Ds;
Whether the gray-scale value of 2. judging current pixel point is greater than preset threshold T; In this way, then get into step 3.; Otherwise, go to step 1.;
3. all gray-scale values in the current pixel vertex neighborhood are revised as the gray-scale value of said current pixel point greater than the gray values of pixel points of preset threshold t, go to step 1.; Said T, t ∈ [1,255], and t＜T.Preferably, T=200, t=50.
Said pixel neighborhood of a point is meant with this pixel to be the rectangular extent at center, and the length of side of said rectangle is between 3 pixel to 9 pixels.Preferably, said rectangle is that the length of side is the square of 5 pixels.
Aforesaid image blurring method, the process of revising described in the step (3) is following::
Definition is the monotonically increasing function f of variable with each gray values of pixel points among the image Dx, said functional value f ∈ [0,1];
To each pixel among the image Dx, multiply by the poor of same position gray values of pixel points in original image and the image M with the functional value f of pixel, the gray-scale value that obtains is added in the image M on the same position gray values of pixel points.
Aforesaid image blurring method, wherein, monotonically increasing function is f
j=1/8 * log
2(Dx
j+ 1); Wherein, f
jAnd Dx
jBe respectively the functional value and the j gray values of pixel points of j pixel among the image Dx.
Aforesaid image blurring method, the acquisition methods of protection zone is following described in the step (1):
At first mark is treated fuzzy region and protection zone in original image; According to said mark original image is resolved into then and treat fuzzy region and protection zone.
Image blurring method as implied above, wherein, mark treats that the method for fuzzy region and protection zone is: mark is treated the partial pixel point in each connected region in fuzzy region and the protection zone.
Aforesaid image blurring method, wherein, decompose and to treat that fuzzy region and protection zone method are following:
The pixel that is labeled as the protection zone earlier is set to 1, and the pixel that is labeled as fuzzy region is set to 0, and the pixel of zone of ignorance is set to-1;
Calculate the confidence level alpha be labeled as-1 pixel again; Alpha ∈ [0,1]; The alpha value of pixel approaches 1 more, and the possibility that belongs to the pixel in the protection zone is big more; Approach 0 more, it is big more to belong to the possibility of treating the pixel in the fuzzy region; The result images that will pass through after the above-mentioned processing is designated as image A.
Aforesaid image blurring method, the process of revising described in the step (3) is following:
With the alpha of pixel in the image A poor with same position gray values of pixel points in original image and the image M on duty, the gray-scale value that obtains is added in the image M on the same position gray values of pixel points.
A kind of image blurring system comprises the deriving means of the protection zone that is used for obtaining original image, and said protection zone is meant the zone that need not carry out Fuzzy Processing;
Be used for original image is kept the fuzzy device of the Fuzzy Processing at edge, the image after the Fuzzy Processing is designated as image M;
Be used for revising the modifier of the pixel pixel value of same position in the image M according to the pixel value of protection zone pixel.
Aforesaid image blurring system, wherein, deriving means comprises the filter unit that is used for original image is carried out bandpass filtering treatment, and the image after handling is designated as image D;
Be used for image D is kept the blur unit of the Fuzzy Processing at edge, the figure after handling is designated as image Ds;
Be used for class expansion cell, the image after handling is designated as image Dx image Ds type of carrying out expansion process.
Aforesaid image blurring system, wherein, type expansion cell comprises the scan module that is used for by pixel scan image Ds;
Whether the gray-scale value that is used to judge current pixel point is greater than the judge module of preset threshold T;
Be used for all gray-scale values in the current pixel vertex neighborhood are replaced with greater than the gray values of pixel points of preset threshold t the replacement module of the gray-scale value of said current pixel point.
Aforesaid image blurring system, wherein, deriving means comprises the indexing unit that is used for treating at the original image mark fuzzy region and protection zone; Be used for original image being resolved into the resolving cell of treating fuzzy region and protection zone according to said mark.
Can there be recovery effects preferably in the method for the invention and system to the image detail texture part after the Fuzzy Processing, and the intensity of recovering can control, and have not only improved image blurring effect effectively, and have improved treatment effeciency effectively.
Description of drawings
Fig. 1 is the structured flowchart of image blurring system among the embodiment 1;
Fig. 2 is the structured flowchart of type expansion cell among the embodiment 1;
Fig. 3 is the structured flowchart of image blurring system among the embodiment 2;
Fig. 4 is the process flow diagram of image blurring method among the embodiment 3;
Fig. 5 is a protect image zone acquisition methods process flow diagram among the embodiment 3;
Fig. 6 is the process flow diagram of type expanding processing among the embodiment 3.
Embodiment
Describe the present invention below in conjunction with embodiment and accompanying drawing.
Present embodiment is a kind of concrete implementation of image blurring system according to the invention.As shown in Figure 1, this system comprises deriving means 1, fuzzy device 2, and the modifier 3 that is connected with fuzzy device 2 with deriving means 1.Said deriving means 1 comprises filter unit 11, a blur unit 12 and a type expansion cell 13.As shown in Figure 2, said type of expansion cell 13 comprises scan module 131, judge module 132 and replacement module 133.
Present embodiment is the another kind of concrete implementation of image blurring system according to the invention.As shown in Figure 3, be that with the main difference of embodiment 1 said system deriving means 1 is different.Deriving means 1 comprises indexing unit 21 and resolving cell 22 in the present embodiment.Indexing unit 21 is used for treating fuzzy region and protection zone at the original image mark.Resolving cell 22 is used for according to mark original image being resolved into treats fuzzy region and protection zone.
Present embodiment is the method that adopts embodiment 1 said system that image is blured.As shown in Figure 4, this method comprises the steps:
(1) deriving means 1 obtains the protection zone in the original image.Said protection zone is meant the zone that need not carry out Fuzzy Processing in the original image.
For example, in people's head image, only need carry out Fuzzy Processing, and parts such as hair and eyebrow need not carried out Fuzzy Processing to people's face.The zone that needs are carried out Fuzzy Processing is called treats fuzzy region, and the zone that need not carry out Fuzzy Processing is called the protection zone.
As shown in Figure 5, the method for obtaining the protection zone in the original image in the present embodiment is following:
(a) 11 pairs of original images of filter unit carry out bandpass filtering, obtain the gray level image D identical with original image size.
Concrete band-pass filtering method can adopt the decomposition of wavelet multiresolution rate, Difference of Gaussian methods such as (DOG).Adopt the DOG method in this embodiment, promptly utilized the Gaussian Blur result of different blur radius to subtract each other and obtained the bandpass filtering result.
Because the noise in the image is generally higher than image border and detail textures frequency,, and keep details and part edge information in the image so bandpass filtering can be effectively with the most of noise filtering in the image.
(b) 12 couples of image D of blur unit keep the Fuzzy Processing at edge, obtain image Ds.
After original image carried out bandpass filtering treatment; The texture part that extracts also can be mingled with a lot of noises; Can effectively remove the noise among the image D through the processing of the maintenance edge fog in this step, and keep the characteristic at edge can effectively keep texture part wherein.
(c) type 13 couples of image Ds of expansion cell type of carrying out expansion process obtains image Dx.Remaining area among the image Dx is the protection zone.
Handle through the maintenance edge fog to image D in the step (b),, also have some losses, so the purpose of the class expansion process in this step is that the efficient recovery texture is arranged though texture part is protected.Because it is low that texture information also has been mingled with a lot of intensity, the composition that frequency is high is promptly quite similar with characteristics of noise, and the operation of the last step of process is easy to treated as noise remove.This step is passed through some stronger data separation noise and the textures in the grain details, and makes texture be able to recover.
As shown in Figure 6, the process of type expansion process is following:
1. scan module 131 is by pixel scan image Ds.
2. judge module 132 judge current pixel point gray-scale value whether greater than preset threshold T; In this way, then get into step 3.; Otherwise, go to step 1..
3. replace module 133 all gray-scale values in the current pixel vertex neighborhood are revised as the gray-scale value of said current pixel point greater than the gray values of pixel points of preset threshold t, go to step 1..
Said pixel neighborhood of a point is meant with this pixel to be the rectangular extent at center, and the length of side of rectangle is preferably the square that the length of side is 5 pixels between 3 pixel to 9 pixels.
The span of threshold value T, t is between [1,255], and t＜T.Preferably, T=200, t=50.
(2) fuzzy 2 pairs of original images of device keep the Fuzzy Processing at edge.Image after the Fuzzy Processing is designated as image M.
Fuzzy Processing to original image can adopt surface blur processing mode or other existing methods commonly used in the Photoshop software.
(3) modifier 3 is revised the pixel value of the pixel of same position in the image M according to the pixel value of pixel in the said protection zone.
Can use the pixel value of pixel among the image Dx directly to replace the pixel value of same position pixel in the image M.But this direct substitute mode may produce replacement part and the factitious problem of other part transition.Therefore, in the present embodiment, adopt following method to revise the pixel value of pixel in the image M.
At first defining with each gray values of pixel points among the image Dx is the monotonically increasing function f of variable, functional value f ∈ [0,1].In the present embodiment, said monotonically increasing function is f
j=1/8 * log
2(Dx
j+ 1); Wherein, f
jAnd Dx
jBe respectively the functional value of pixel j among the image Dx and the gray-scale value of pixel j.
Then to each pixel among the image Dx, multiply by the poor of same position gray values of pixel points in original image and the image M with the functional value f of pixel, the gray-scale value that obtains is added in the image M on the gray-scale value with this pixel position same pixel point.That is:
In the image M with image Dx in function f value * (the pixel a of original image and pixel a same position " the gray-scale value of gray-scale value-pixel a ') of gray-scale value+pixel a of modification value=pixel a ' of pixel a ' of pixel a same position.
Need to prove that also if original image is a gray-scale map, gray values of pixel points promptly is the pixel value of this pixel so, adopt the said method direct modification to get final product.If cromogram during original image then need carry out above-mentioned modification to the pixel in the gray-scale map of every kind of Color Channel.For example, the cromogram in the rgb color space need carry out the modification of above-mentioned pixel value to R passage, G passage and B passage respectively.
The pixel that adopts said method to revise in the image M is compared with direct replacing method, and the transition of replacement part and other parts is more natural.
Embodiment 4
Present embodiment is the method that adopts embodiment 2 said systems that image is blured.Be with the difference of embodiment 3 said methods: the method for obtaining the protection zone in the original image in the step (1) is different, revises the method difference of the pixel value of the pixel in the image M in the step (3).
In the present embodiment, it is following to obtain in the original image method of protection zone:
At first, utilize indexing unit 21 in original image, to mark and treat fuzzy region and protection zone.When mark is treated fuzzy region and protection zone, all pixels of respective regions needn't be accurate to, only any part that belongs to this zone need be selected.For the situation that does not belong to same connected region, need all mark.For example, need hair in the character image and eyebrow be labeled as the protection zone, then need be in a hair internal labeling part, because eyebrow is two parts separately with hair, so also need also mark is a part of on eyebrow.
Then, resolving cell 22 resolves into original image according to mark and treats fuzzy region and protection zone.Can adopt existing image disassembling method, image resolved into according to tag content automatically treat fuzzy region and protection zone.
After obtaining the protection zone, can use the pixel value of protection zone pixel directly to replace the pixel value of the pixel of same position in the image M.But this direct substitute mode may cause replacement part and the factitious problem of other part transition.
In the present embodiment, the method that adopts existing stingy nomography to disassemble as image.The pixel that is labeled as the protection zone earlier is set to 1, and the pixel that is labeled as fuzzy region is set to 0, and the pixel of zone of ignorance is set to-1, has obtained one three value image Tmap like this.Calculate the confidence level alpha that is labeled as-1 pixel again, alpha is between [0,1], and alpha representes that this pixel is the confidence level of protection zone interior pixel point.The alpha value of this pixel approaches 1 more, and the possibility that belongs to protection zone interior pixel point is big more; Otherwise approach 0 more, it is big more to belong to the possibility of treating fuzzy region interior pixel point.The image that will pass through after the above-mentioned processing is designated as image A.With the alpha of pixel in the image A poor with same position gray values of pixel points in original image and the image M on duty, the gray-scale value that obtains is added in the image M on the gray-scale value with this pixel position same pixel point.That is:
In the image M with image A in alpha value * (the pixel a of original image and pixel a same position " the gray-scale value of gray-scale value-pixel a ') of gray-scale value+pixel a of modification value=pixel a ' of pixel a ' of pixel a same position.
Need to prove that also if original image is a gray-scale map, gray values of pixel points promptly is the pixel value of this pixel so, adopt the said method direct modification to get final product.If cromogram during original image then need carry out above-mentioned modification to the pixel in the gray-scale map of every kind of Color Channel.For example, the cromogram in the rgb color space need carry out the modification of above-mentioned pixel value to R passage, G passage and B passage respectively.
The pixel that adopts said method to revise in the image M is compared with direct replacing method, and the transition of replacement part and other parts is more natural.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, belong within the scope of claim of the present invention and equivalent technology thereof if of the present invention these are revised with modification, then the present invention also is intended to comprise these changes and modification interior.
Claims (16)
1. image blurring method may further comprise the steps:
(1) obtain protection zone in the original image, said protection zone is meant the zone that need not carry out Fuzzy Processing;
(2) original image is kept the Fuzzy Processing at edge, the image after the Fuzzy Processing is designated as image M;
(3) revise the pixel value of the pixel of same position in the said image M according to the pixel value of pixel in the said protection zone.
2. image blurring method as claimed in claim 1 is characterized in that: the acquisition methods of protection zone is following described in the step (1):
(a) original image is carried out bandpass filtering, obtain image D;
(b) image D is kept the Fuzzy Processing at edge, obtain image Ds;
(c), obtain image Dx to image Ds type of carrying out expansion process; Remaining area among the image Dx is the protection zone.
3. image blurring method as claimed in claim 2 is characterized in that: the process of type expansion process is following described in the step (c):
1. by pixel scan image Ds;
Whether the gray-scale value of 2. judging current pixel point is greater than preset threshold T; In this way, then get into step 3.; Otherwise, go to step 1.;
3. all gray-scale values in the current pixel vertex neighborhood are revised as the gray-scale value of said current pixel point greater than the gray values of pixel points of preset threshold t, go to step 1.; Said T, t ∈ [1,255], and t＜T.
4. image blurring method as claimed in claim 3 is characterized in that: said T=200, t=50.
5. image blurring method as claimed in claim 3 is characterized in that: step 3. described in the pixel neighborhood of a point be meant with this pixel to be the rectangular extent at center, the length of side of said rectangle is between 3 pixel to 9 pixels.
6. image blurring method as claimed in claim 5 is characterized in that: said rectangle is that the length of side is the square of 5 pixels.
7. like each described image blurring method in the claim 2～6, it is characterized in that: the process of revising described in the step (3) is following:
Definition is the monotonically increasing function f of variable with each gray values of pixel points among the image Dx, said functional value f ∈ [0,1];
To each pixel among the image Dx, multiply by the poor of same position gray values of pixel points in original image and the image M with the functional value f of pixel, the gray-scale value that obtains is added in the image M on the same position gray values of pixel points.
8. image blurring method as claimed in claim 7 is characterized in that: said monotonically increasing function is f
j=1/8 * log
2(Dx
j+ 1); Wherein, f
jAnd Dx
jBe respectively the functional value of pixel j among the image Dx and the gray-scale value of pixel j.
9. image blurring method as claimed in claim 1 is characterized in that: the acquisition methods of protection zone is following described in the step (1):
At first mark is treated fuzzy region and protection zone in original image; According to said mark original image is resolved into then and treat fuzzy region and protection zone.
10. the image blurring method shown in claim 9, it is characterized in that: said mark treats that the method for fuzzy region and protection zone is: mark is treated the partial pixel point in each connected region in fuzzy region and the protection zone.
11. image blurring method as claimed in claim 10 is characterized in that: decompose and to treat that fuzzy region and protection zone method are following:
The pixel that is labeled as the protection zone earlier is set to 1, and the pixel that is labeled as fuzzy region is set to 0, and the pixel of zone of ignorance is set to-1;
Calculate the confidence level alpha be labeled as-1 pixel again; Alpha ∈ [0,1]; The alpha value of pixel approaches 1 more, and the possibility that belongs to the pixel in the protection zone is big more; Approach 0 more, it is big more to belong to the possibility of treating the pixel in the fuzzy region; The result images that will pass through after the above-mentioned processing is designated as image A.
12. image blurring method as claimed in claim 11 is characterized in that: the process of revising described in the step (3) is following:
With the alpha of pixel in the image A poor with same position gray values of pixel points in original image and the image M on duty, the gray-scale value that obtains is added in the image M on the same position gray values of pixel points.
13. an image blurring system comprises the deriving means (1) of the protection zone that is used for obtaining original image, said protection zone is meant the zone that need not carry out Fuzzy Processing;
Be used for original image is kept the fuzzy device (2) of the Fuzzy Processing at edge, the image after the Fuzzy Processing is designated as image M;
Be used for revising the modifier (3) of the pixel pixel value of same position in the image M according to the pixel value of protection zone pixel.
14. image blurring system as claimed in claim 13 is characterized in that: said deriving means (1) comprises the filter unit (11) that is used for original image is carried out bandpass filtering treatment, and the image after handling is designated as image D;
Be used for image D is kept the blur unit (12) of the Fuzzy Processing at edge, the figure after handling is designated as image Ds;
Be used for class expansion cell (13), the image after handling is designated as image Dx image Ds type of carrying out expansion process.
15. image blurring system as claimed in claim 14 is characterized in that: said type of expansion cell (13) comprises the scan module (131) that is used for by pixel scan image Ds;
Whether the gray-scale value that is used to judge current pixel point is greater than the judge module (132) of preset threshold T;
Be used for all gray-scale values in the current pixel vertex neighborhood are replaced with greater than the gray values of pixel points of preset threshold t the replacement module (133) of the gray-scale value of said current pixel point.
16. image blurring system as claimed in claim 13 is characterized in that: said deriving means (1) comprises the indexing unit (21) that is used for treating at the original image mark fuzzy region and protection zone; Be used for original image being resolved into the resolving cell (22) of treating fuzzy region and protection zone according to said mark.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106069869A CN102542533A (en) | 2010-12-16 | 2010-12-16 | Image blurring method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106069869A CN102542533A (en) | 2010-12-16 | 2010-12-16 | Image blurring method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102542533A true CN102542533A (en) | 2012-07-04 |
Family
ID=46349358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106069869A Pending CN102542533A (en) | 2010-12-16 | 2010-12-16 | Image blurring method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102542533A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103702024A (en) * | 2013-12-02 | 2014-04-02 | 宇龙计算机通信科技(深圳)有限公司 | Image processing device and image processing method |
| CN104933676A (en) * | 2015-07-14 | 2015-09-23 | 厦门美图之家科技有限公司 | Edge-preserved image fuzzy processing method and system, and shooting terminal |
| CN104992120A (en) * | 2015-06-18 | 2015-10-21 | 广东欧珀移动通信有限公司 | Picture encryption method and mobile terminal |
| CN106296576A (en) * | 2016-08-05 | 2017-01-04 | 厦门美图之家科技有限公司 | Image processing method and image processing apparatus |
| CN109523483A (en) * | 2018-11-14 | 2019-03-26 | 北京奇艺世纪科技有限公司 | A kind of image defogging method and device |
| CN110636331A (en) * | 2019-09-26 | 2019-12-31 | 北京百度网讯科技有限公司 | Method and apparatus for processing video |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090067740A1 (en) * | 2007-09-06 | 2009-03-12 | Li Hua Zhu | Spatial diffusion in images |
| CN101639938A (en) * | 2009-08-28 | 2010-02-03 | 浙江大学 | Image restoration method based on double-edge wave filter and margin deconvolution |
-
2010
- 2010-12-16 CN CN2010106069869A patent/CN102542533A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090067740A1 (en) * | 2007-09-06 | 2009-03-12 | Li Hua Zhu | Spatial diffusion in images |
| CN101639938A (en) * | 2009-08-28 | 2010-02-03 | 浙江大学 | Image restoration method based on double-edge wave filter and margin deconvolution |
Non-Patent Citations (1)
| Title |
|---|
| 齐欣: "数码影像技术操作篇之三十一 水墨徽州", 《照相机》, no. 7, 31 July 2006 (2006-07-31), pages 44 - 45 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103702024A (en) * | 2013-12-02 | 2014-04-02 | 宇龙计算机通信科技(深圳)有限公司 | Image processing device and image processing method |
| CN104992120A (en) * | 2015-06-18 | 2015-10-21 | 广东欧珀移动通信有限公司 | Picture encryption method and mobile terminal |
| CN104933676A (en) * | 2015-07-14 | 2015-09-23 | 厦门美图之家科技有限公司 | Edge-preserved image fuzzy processing method and system, and shooting terminal |
| CN104933676B (en) * | 2015-07-14 | 2019-01-08 | 厦门美图之家科技有限公司 | A kind of image fuzzy processing method, system and camera terminal retaining edge |
| CN106296576A (en) * | 2016-08-05 | 2017-01-04 | 厦门美图之家科技有限公司 | Image processing method and image processing apparatus |
| CN106296576B (en) * | 2016-08-05 | 2019-07-16 | 厦门美图之家科技有限公司 | Image processing method and image processing apparatus |
| CN109523483A (en) * | 2018-11-14 | 2019-03-26 | 北京奇艺世纪科技有限公司 | A kind of image defogging method and device |
| CN110636331A (en) * | 2019-09-26 | 2019-12-31 | 北京百度网讯科技有限公司 | Method and apparatus for processing video |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102542533A (en) | Image blurring method and system | |
| CN102968782B (en) | In a kind of coloured image, remarkable object takes method automatically | |
| CN103747213B (en) | A kind of real-time defogging method of the Traffic Surveillance Video based on moving target | |
| CN106251301A (en) | A kind of single image defogging method based on dark primary priori | |
| CN104331868B (en) | A kind of optimization method of image border | |
| CN110443807A (en) | A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion | |
| CN104794688A (en) | Single image defogging method and device based on depth information separation sky region | |
| WO2008091278A3 (en) | Iris data extraction | |
| CN109325498B (en) | Vein extraction method for improving Canny operator based on window dynamic threshold | |
| Moghaddam et al. | RSLDI: Restoration of single-sided low-quality document images | |
| Ning et al. | A fast single image haze removal method based on human retina property | |
| CN104134198A (en) | Method for carrying out local processing on image | |
| Xu et al. | Colour image enhancement by virtual histogram approach | |
| CN105023252A (en) | Method and system for enhancement processing of beautified image and shooting terminal | |
| CN111145105A (en) | Image rapid defogging method and device, terminal and storage medium | |
| CN111080723B (en) | Image element segmentation method based on Unet network | |
| CN111598814A (en) | Single image defogging method based on extreme scattering channel | |
| CN109118440A (en) | Single image to the fog method based on transmissivity fusion with the estimation of adaptive atmosphere light | |
| Lu et al. | A shadow removal method for tesseract text recognition | |
| Yamato et al. | Digital image watermarking method using between-class variance | |
| Fu et al. | A feature-dependent fuzzy bidirectional flow for adaptive image sharpening | |
| Demir et al. | Low-light image enhancement based on sharpening-smoothing image filter | |
| Li et al. | DLT-Net: deep learning transmittance network for single image haze removal | |
| CN110874823A (en) | Mine fog image enhancement method based on dark primary color prior and homomorphic filtering | |
| Han et al. | Locally adaptive contrast enhancement using convolutional neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20120704 |
|
| C20 | Patent right or utility model deemed to be abandoned or is abandoned |