CN103793150B - Image-selecting method and system - Google Patents
Image-selecting method and system Download PDFInfo
- Publication number
- CN103793150B CN103793150B CN201210428391.8A CN201210428391A CN103793150B CN 103793150 B CN103793150 B CN 103793150B CN 201210428391 A CN201210428391 A CN 201210428391A CN 103793150 B CN103793150 B CN 103793150B
- Authority
- CN
- China
- Prior art keywords
- pixel
- image
- described image
- region
- msub
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
A kind of image-selecting method includes：It is K0 to mark all pixels point in image；Pixel mark is changed to K1 in the region that record selection brush is inswept on image, the region；The region for being will be marked to amplify preset ratio in image, and the pixel in the region in image after amplification is changed to K2 marked as K0；Color-spatial distribution of the statistics labeled as K1 pixel；Color-spatial distribution of the statistics labeled as K2 pixel；Energy distribution model is set up to the pixel that mark is；According to energy distribution model, when calculating obtains energy function minimum, labeled as the indicia distribution of K2 each pixel, indicia distribution is K0 or K1；According to indicia distribution, it is or K1 by the mark labeled as K2 each pixel.The present invention also provides a kind of correspondence system and autgmentability image selection is accurately realized there is provided more convenient and quicker.
Description
Technical field
The present invention relates to image processing techniques, the more particularly to system of selection of picture material and system.
Background technology
Image processing techniques is often used in the life of people, and such as PS softwares are exactly that the most frequently used image procossing is soft
Part.Before the Local treatment to image, the face image of personage is such as handled, that is accomplished by choosing the facial regions in image first
Domain.Conventional image selection mode mainly has following several：
A kind of is the mode of fixed circular brush, and user drags mouse pointer on the image of original image or partial enlargement
Or finger is dragged on the touchscreen, click all choose a circular region every time, in order to select the detail section of image, lead to
Normal way is the radius for changing circular brush, and is repeatedly selected.
Another is magic wand tool, is clicked on behind a position in image, system will be automatically selected and the position phase
Connection and the same or similar region of color.
However, when using fixed brush selection image detail region, user must be accurately carried out point each time
Choosing or dragging operation, it is necessary to interactive operation it is excessive, although higher selection accuracy can be obtained, the efficiency of selection is very
It is low.
Although magic wand tool, which can be automated, chooses certain color, however, by taking face as an example, its face-image is not
One single color, can not realize selection by the instrument.
The content of the invention
Based on this, it is necessary to provide a kind of efficiently and accurately and image-selecting method and system with extension selection function.
A kind of image-selecting method, comprises the following steps：
It is K0 to mark all pixels point in described image；
Receive and respond user's operation, record selection brush inswept region on the image will should in described image
Pixel mark in region is changed to K1, and records the position of the end point of the selection brush；
The region for being will be marked to amplify preset ratio in described image, and by the region in described image after amplification
Pixel be changed to K2 marked as K0；
Count in described image in the region of the default size of the position of the selection brush mobile end point, mark
For the Color-spatial distribution of K1 pixel, predetermined number color cluster center is obtained using clustering method；
The Color-spatial distribution for the pixel for being is marked in statistics described image, predetermined number is obtained using clustering method
Individual color cluster center；
The predetermined number color cluster center of K1 pixel is designated as according to the region internal standard of the default size,
With the predetermined number color cluster center that the pixel for being is marked in described image, to marking the picture for being in described image
Vegetarian refreshments sets up energy distribution model；
According to the energy distribution model, calculating obtain energy function it is minimum when, be each is marked in described image
The indicia distribution of pixel, the indicia distribution is K0 or K1；
According to the indicia distribution, it is or K1 by the mark for marking each pixel for being in described image.
In one of embodiment, also comprise the following steps：The pixel for being will be marked to be set in described image
Selection state.The selection brush is circle.
In one of embodiment, the predetermined number is more than one.
In one of embodiment, the energy function of the energy distribution model is as follows：
Wherein：
Ed(xp)=xpmin|Ip-Cf|+(1-xp)min|Ip-Cb|
Ec(xp,xq)=|xp-xq|(Ip-Iq)-1
Cf is designated as the predetermined number color cluster center of K1 pixel for the region internal standard of the default size；
Cb is the predetermined number color cluster center that the pixel for being is marked in described image；
K0=0, K1=1, K2=2, Ip, IqFor the adjacent p in position, the pixel color of q points.Ed(xp) represent described image middle position
The pixel for being set to p is labeled as xpWhen energy expenditure, xp=K0 or K1, Ec(xp,xq) illustrate the adjacent point marks of p q two in position
Note is respectively Ip, IpWhen energy expenditure, xq=0 or 1.
In one of embodiment, the indicia distribution is obtained using energy distribution model described in maxflow Algorithm for Solving
Arrive.
A kind of image selection system, including：
All pixels point is K0 in image shows module, mark image；
Logging modle, for receiving and responding user's operation, record selection selection brush inswept area on the image
Domain, is changed to K1, and record the position of the end point of selection brush by the pixel mark in the region of this in described image；
Expansion module, for will be marked in described image be region amplify preset ratio, and by described image
The pixel in region after amplification is changed to K2 marked as K0；
First statistical module, for counting in described image close to the default of the position of the selection brush mobile end point
In the region of size, labeled as the Color-spatial distribution of K1 pixel, predetermined number color is obtained using clustering method and gathered
Class center；
Second statistical module, the Color-spatial distribution for the pixel for being is marked for counting in described image, using poly-
Class method obtains predetermined number color cluster center；
Modeling module, the predetermined number of the pixel for being designated as K1 according to the region internal standard of the default size
Color cluster center, and the predetermined number color cluster center for the pixel for being is marked in described image, to described image
It is middle to mark the pixel for being to set up energy distribution model；
Processing module, according to the energy distribution model, calculating obtain energy function it is minimum when, in described image mark for
The indicia distribution of K2 each pixel, the indicia distribution is K0 or K1；
Mark module, for according to the indicia distribution, the mark for each pixel for being being marked in described image
For K0 or K1.
In one of embodiment, in addition to：Selecting module, for the pixel for being will to be marked to set in described image
To have selected state.
In one of embodiment, the predetermined number is more than one.
In one of embodiment, the energy function for the energy distribution model that the modeling module is set up is as follows：
Wherein：
Ed(xp)=xpmin|Ip-Cf|+(1-xp)min|Ip-Cb|
Ec(xp,xq)=|xp-xq|(Ip-Iq)-1
Cf is designated as the predetermined number color cluster center of K1 pixel for the region internal standard of the default size；
Cb is the predetermined number color cluster center that the pixel for being is marked in described image；
K0=0, K1=1, K2=2, Ip, IqFor the adjacent p in position, the pixel color of q points.Ed(xp) represent described image middle position
The pixel for being set to p is labeled as xpWhen energy expenditure, xp=K0 or K1, Ec(xp,xq) illustrate the adjacent point marks of p q two in position
Note is respectively Ip, IpWhen energy expenditure, xq=0 or 1.
In one of embodiment, the processing module is obtained using energy distribution model described in maxflow Algorithm for Solving
To the indicia distribution.
For the mode of above-mentioned image-selecting method and the relatively conventional fixed brush of system, user can wish selection
Image-region is arbitrarily clicked, and algorithm real-time and accurately can expand to selection region the proximity similar to selection region
Domain, so as to improve efficiency and accuracy that user selects image-region.For the selection mode of magic wand tool, magic
Rod can only carry out monochrome image extension to some pixel, and to one polyenergetic image of selection, such as face is to use
Magic wand tool selection.And user is selected and extended to combine by this case methods described, disposable extension selection polyenergetic is realized
Image, can also count multiple color cluster centers from the region of selection, so as to carry out multiple color simultaneously default
In the range of extend so that efficiency of selection is higher, more accurately.
Brief description of the drawings
Fig. 1 is the step flow chart of the image-selecting method of an embodiment；
Form of expression schematic diagram when Fig. 2 is the image-selecting method selection picture material shown in Fig. 1；
Fig. 3 is the functional block diagram of the image selection system of an embodiment.
Embodiment
The image-selecting method of the embodiment of this case one is the selection area extension selected in the picture will to be operated based on user
Default size, be expanded region；Then at least part pixel in extended area is included into selection area, it is described to include choosing
Determine the pixel in region and the color distance of selection area, the color distance of pixel with not including selection area and with phase
Adjoint point marks whether that identical feature meets preparatory condition jointly.
Realize the segmentation of above-mentioned extended area and partial pixel is included into selection area, can be using the most frequently used base of industry
In the image partition method of energy function.
As shown in figure 1, it is the step flow chart of the image-selecting method of an embodiment, comprise the following steps：
Step 101, all pixels point is K0 in mark image.
The K0 can be Arabic numerals, letter, symbol etc., for playing in separator effect, the present embodiment, K0
Digital " 0 " is taken, here labeled as the non-selected pixel of expression of " 0 ", image initial is also.
Step 102, receive and respond user's operation, record selection brush inswept region on the image will be described
Pixel mark in the region of this in image is changed to K1, and records the position of the end point of selection brush.
The selection brush can be arbitrary shape, generally circular or square, and the present embodiment is circular brush.The choosing
It can be that user clicks mouse or pins mouse dragging action to select operation, can also be user's finger or stylus in touch-screen
On click or sliding action.The K1 can be Arabic numerals, letter, symbol etc., and in the present embodiment, K1 takes " 1 ".
Step 103, will be marked in described image be region amplify preset ratio, and by described image after amplification
Region in pixel be changed to K2 marked as K0.
The K2 can be Arabic numerals, letter, symbol etc..In the present embodiment, K2 takes " 2 ".Now, in image, mark
The K0 non-selected region of expression is designated as, labeled as K1 expression selection region, the expression labeled as K2 extends selection region, but
Selection region, which may not be extended, can all be chosen.
Step 104, the area of the default size of the position of the close selection brush mobile end point in described image is counted
In domain, labeled as the Color-spatial distribution of K1 pixel, predetermined number color cluster center is obtained using clustering method.
In the present embodiment, the region of the default size of the position close to the selection brush mobile end point is with institute
State selection brush mobile end point position centered on default size region.
In the present embodiment, the predetermined number color that the region internal standard of the default size is designated as K1 pixel is gathered
Class center is denoted as Cf.
Step 105, the Color-spatial distribution for the pixel for being is marked in statistics described image, is obtained using clustering method
Predetermined number color cluster center.
The predetermined number color cluster center for the pixel for being is marked to be denoted as Cb in the present embodiment, in described image.
Above-mentioned two predetermined number can be any, at least one, and predetermined number is more than one in the present embodiment.
Step 106, the predetermined number color of K1 pixel is designated as according to the region internal standard of the default size
The predetermined number color cluster center for the pixel for being is marked in cluster centre, and described image, described image is got the bid
The pixel for being designated as K2 sets up energy distribution model.
It is described to marking the pixel for being to set up energy distribution model in described image, refer to by the predetermined number
Color cluster centre data is brought into default energy function, and solution is calculated to carry out next step.
In the present embodiment, K0=0, K1=1, K2=2, the energy function of the energy distribution model is as follows：
Wherein：
Ed(xp)=xpmin|Ip-Cf|+(1-xp)min|Ip-Cb|
Ec(xp,xq)=|xp-xq|(Ip-Iq)-1
Ip, IqFor the adjacent p in position, the pixel color of q points.Ed(xp) represent that described image middle position is set to p pixel mark
It is designated as xpWhen energy expenditure, xp=0 or 1, Ec(xp,xq) illustrate adjacent 2 points of marks of the pq respectively x in positionp, xqWhen energy
Amount consumption, xq=0 or 1.
In other embodiment, if labeled as letter or other irregular marks, need to only set two energy function difference
The identical and different two kinds of situations of corresponding adjacent pixel piont mark.
Marked step 107, according to the energy distribution model, when calculating obtains energy function minimum, in described image
The indicia distribution of K2 each pixel, the indicia distribution is K0 or K1.
Permutation and combination is carried out equivalent to marking the mark for each pixel for being to be changed to K0 or K1 in described image,
The value for obtaining energy function during various permutation and combination is calculated respectively, retains the corresponding permutation and combination method of energy function minimum value.
In the present embodiment, the indicia distribution is obtained using energy distribution model described in maxflow Algorithm for Solving.
Step 108, according to the indicia distribution, be by the mark for marking each pixel for being in described image or
Person K1.
Step 109, the pixel for being will be marked to be set to select state in described image.
It is described to have selected state surround the pixel for being labeled as K1 with dotted line frame, or with certain color with semi-transparent
Bright mode overlay marks are K1 pixel etc., as long as being easy to user's difference selection and non-selected content.Now use
Family can see oneself this time drag selected region have been realized in extension.
For the mode of the above-mentioned relatively conventional fixed brush of image-selecting method, user can wish the image district of selection
Domain is arbitrarily clicked, and algorithm real-time and accurately can expand to selection region the adjacent domain similar to selection region, so that
Improve efficiency and accuracy that user selects image-region.For the selection mode of magic wand tool, Magic wand can only
Monochrome image extension is carried out to some pixel, to one polyenergetic image of selection, such as face is can not to use Magic wand
Instrument selection.And user is selected and extended to combine by this case methods described, disposable extension selection polyenergetic image is realized, also
Multiple color cluster centers can be counted from the region of selection, are expanded within a preset range so as to carry out multiple color simultaneously
Exhibition so that efficiency of selection is higher, more accurately.
Sometimes user can carry out contracting/putting after display to image, then carry out image selection.Therefore, in other embodiment, institute
Stating step 103 preset ratio can be according to the contracting of described image/put ratio correspondence change, and such as preset ratio and image contract/put ratio
It is proportional.
Please refer to Fig. 2, form of expression signal when it is the image-selecting method selection picture material shown in Fig. 1
Figure.
In Fig. 2, user selects picture material by circular brush in image 20, during initialization, each picture of image 20
Vegetarian refreshments is labeled as " 0 ".Circular brush is dragged to B points by user from A points, obtains selection region 31, and mark in extended area 32
Pixel is " 1 ".Now, system can amplify selection region 31 automatically, and be expanded region 32, and marks in extended area 32
Pixel is " 2 ".Then the color space using the mark in the default size area 33 in the B points center of circle as 1 " pixel is counted
Distribution, predetermined number color cluster center is obtained using clustering method；The color of the pixel in extended area 32 is counted again
Spatial distribution, predetermined number color cluster center is obtained using clustering method.Calculated again by energy distribution model and obtain energy
When flow function is minimum, the indicia distribution of each pixel in extended area 32, and set according to indicia distribution in extended area 32
The mark of each pixel.The pixel that all marks be 1 " in final image 20 is to select during content, its scope be probably Fig. 2
The part that dotted line frame 34 is surrounded.
As shown in figure 3, its functional block diagram for the image selection system 40 of an embodiment, including：
Image shows module 401, is K0 for marking all pixels point in described image.
Logging modle 402, for receiving and responding user's operation, record selection selection brush is inswept on the image
Region, is changed to K1, and record the position of the end point of selection brush by the pixel mark in the region of this in described image.
Expansion module 403, for the region for being will to be marked to amplify preset ratio in described image, and by described image
The pixel in region after amplification is changed to K2 marked as K0.
First statistical module 404, for counting in described image close to the position of the selection brush mobile end point
In the region of default size, labeled as the Color-spatial distribution of K1 pixel, predetermined number face is obtained using clustering method
Color cluster centre.
Second statistical module 405, the Color-spatial distribution for the pixel for being is marked for counting in described image, used
Clustering method obtains predetermined number color cluster center.
Modeling module 406, the present count of the pixel for being designated as K1 according to the region internal standard of the default size
Amount color cluster center, and the predetermined number color cluster center for the pixel for being is marked in described image, to described
The pixel for being is marked to set up energy distribution model in image.
Processing module 407, according to the energy distribution model, when calculating obtains energy function minimum, described image acceptance of the bid
The indicia distribution of K2 each pixel is designated as, the indicia distribution is K0 or K1.
Mark module 408, for according to the indicia distribution, the mark for each pixel for being being marked in described image
It is designated as K0 or K1.
Selecting module 409, for the pixel for being will to be marked to be set to select state in described image.
Described K0, K1 and K2 can be Arabic numerals, letter, symbol etc., for playing separator effect.
Other features of above-mentioned image selection system 40 can be identical with above-mentioned image-selecting method, and here is omitted.
For the mode of the above-mentioned relatively conventional fixed brush of image selection system 40, user can wish the image of selection
Region is arbitrarily clicked, and algorithm real-time and accurately can expand to selection region the adjacent domain similar to selection region, from
And improve efficiency and accuracy that user selects image-region.For the selection mode of magic wand tool, Magic wand is only
Monochrome image extension can be carried out to some pixel, to one polyenergetic image of selection, such as face is can not to use magic
The selection of rod instrument.And user is selected and extended to combine by this case methods described, disposable extension selection polyenergetic image is realized,
Multiple color cluster centers can be also counted from the region of selection, so as to carry out multiple color within a preset range simultaneously
Extension so that efficiency of selection is higher, more accurately.
Embodiment described above only expresses the several embodiments of the present invention, and it describes more specific and detailed, but simultaneously
Therefore the limitation to the scope of the claims of the present invention can not be interpreted as.It should be pointed out that for one of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the guarantor of the present invention
Protect scope.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (10)
1. a kind of image-selecting method, it is characterised in that comprise the following steps：
It is K0 to mark all pixels point in described image；
Receive and respond user's operation, record selection brush inswept region on the image, by the region of this in described image
Interior pixel mark is changed to K1, and records the position of the end point of the selection brush；
The region for being will be marked to amplify preset ratio in described image, and by the picture in the region in described image after amplification
Vegetarian refreshments is changed to K2 marked as K0；
Count in described image in the region of the default size of the position of the selection brush mobile end point, labeled as K1
Pixel Color-spatial distribution, predetermined number color cluster center is obtained using clustering method；
The Color-spatial distribution for the pixel for being is marked in statistics described image, predetermined number face is obtained using clustering method
Color cluster centre；
The predetermined number color cluster center of K1 pixel, and institute are designated as according to the region internal standard of the default size
The predetermined number color cluster center that the pixel for being is marked in image is stated, to marking the pixel for being in described image
Set up energy distribution model；
According to the energy distribution model, when calculating obtains energy function minimum, each pixel for being is marked in described image
The indicia distribution of point, the indicia distribution is K0 or K1；
According to the indicia distribution, it is or K1 by the mark for marking each pixel for being in described image.
2. image-selecting method according to claim 1, it is characterised in that described according to the indicia distribution, will be described
The mark that each pixel for being is marked in image be or the step of K1 after, also comprise the following steps：
The pixel for being will be marked to be set to select state in described image.
3. image-selecting method according to claim 1, it is characterised in that the predetermined number is more than one.
4. image-selecting method according to claim 1, it is characterised in that the energy function of the energy distribution model is such as
Under：
<mrow>
<mi>E</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>p</mi>
</munder>
<msub>
<mi>E</mi>
<mi>d</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>p</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>+</mo>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>p</mi>
<mo>,</mo>
<mi>q</mi>
</mrow>
</munder>
<msub>
<mi>E</mi>
<mi>c</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>p</mi>
</msub>
<mo>,</mo>
<msub>
<mi>x</mi>
<mi>q</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
Wherein：
Ed(xp)=xp min|Ip-Cf|+(1-xp)min|Ip-Cb|
Ec(xp,xq)=| xp-xq|(Ip-Iq)-1
Cf is designated as the predetermined number color cluster center of K1 pixel for the region internal standard of the default size；
Cb is the predetermined number color cluster center that the pixel for being is marked in described image；
K0=0, K1=1, K2=2, Ip, IqFor the adjacent p in position, the pixel color of q points；Ed(xp) represent position in described image
X is labeled as p pixelpWhen energy expenditure, xp=K0 or K1, Ec(xp,xq) illustrate the adjacent 2 points of marks of p, q in position
Respectively Ip, IqWhen energy expenditure, xq=0 or 1.
5. image-selecting method according to claim 4, it is characterised in that the indicia distribution is calculated using maxflow
Method solves the energy distribution model and obtained.
6. a kind of image selection system, it is characterised in that including：
Image shows module, is K0 for opening all pixels point in simultaneously display image, mark described image；
Logging modle, for receiving and responding user's operation, record selection selection brush inswept region on the image will
Pixel mark in the region of this in described image is changed to K1, and records the position of the end point of selection brush；
Preset ratio is amplified in expansion module, the region for being for that will be marked in described image, and will amplified in described image
The pixel in region afterwards is changed to K2 marked as K0；
First statistical module, for counting the default size in described image close to the position of the selection brush mobile end point
Region in, labeled as the Color-spatial distribution of K1 pixel, obtained using clustering method in predetermined number color cluster
The heart；
Second statistical module, the Color-spatial distribution for the pixel for being is marked for counting, using cluster side in described image
Method obtains predetermined number color cluster center；
Modeling module, the predetermined number color of the pixel for being designated as K1 according to the region internal standard of the default size
The predetermined number color cluster center for the pixel for being is marked in cluster centre, and described image, described image is got the bid
The pixel for being designated as K2 sets up energy distribution model；
Processing module, according to the energy distribution model, when calculating obtains energy function minimum, is marked in described image
The indicia distribution of each pixel, the indicia distribution is K0 or K1；
Mark module, for according to the indicia distribution, be by the mark for marking each pixel for being in described image
Or K1.
7. image selection system according to claim 6, it is characterised in that also include：Selecting module, for the mark
After module is called, the pixel for being will be marked to be set to select state in described image.
8. image selection system according to claim 6, it is characterised in that the predetermined number is more than one.
9. image selection system according to claim 6, it is characterised in that the energy point that the modeling module is set up
The energy function of cloth model is as follows：
<mrow>
<mi>E</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>p</mi>
</munder>
<msub>
<mi>E</mi>
<mi>d</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>p</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>+</mo>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>p</mi>
<mo>,</mo>
<mi>q</mi>
</mrow>
</munder>
<msub>
<mi>E</mi>
<mi>c</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>p</mi>
</msub>
<mo>,</mo>
<msub>
<mi>x</mi>
<mi>q</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
Wherein：
Ed(xp)=xpmin|Ip-Cf|+(1-xp)min|Ip-Cb|
Ec(xp,xq)=| xp-xq|(Ip-Iq)-1
Cf is designated as the predetermined number color cluster center of K1 pixel for the region internal standard of the default size；
Cb is the predetermined number color cluster center that the pixel for being is marked in described image；
K0=0, K1=1, K2=2, Ip, IqFor the adjacent p in position, the pixel color of q points；Ed(xp) represent position in described image
X is labeled as p pixelpWhen energy expenditure, xp=K0 or K1, Ec(xp,xq) illustrate the adjacent 2 points of marks of p, q in position
Respectively Ip, IpWhen energy expenditure, xq=0 or 1.
10. image selection system according to claim 9, it is characterised in that the processing module is calculated using maxflow
Method solves the energy distribution model and obtains the indicia distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210428391.8A CN103793150B (en) | 2012-10-31 | 2012-10-31 | Image-selecting method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210428391.8A CN103793150B (en) | 2012-10-31 | 2012-10-31 | Image-selecting method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103793150A CN103793150A (en) | 2014-05-14 |
| CN103793150B true CN103793150B (en) | 2017-08-25 |
Family
ID=50668888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210428391.8A Active CN103793150B (en) | 2012-10-31 | 2012-10-31 | Image-selecting method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103793150B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112051959B (en) * | 2020-09-02 | 2022-05-27 | 北京字节跳动网络技术有限公司 | Method, device and equipment for generating image drawing process and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398930A (en) * | 2008-10-24 | 2009-04-01 | 清华大学 | Interactive edition method and system |
| CN101599124A (en) * | 2008-06-03 | 2009-12-09 | 汉王科技股份有限公司 | A kind of from video image the method and apparatus of separating character |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7602991B2 (en) * | 2001-10-24 | 2009-10-13 | Nik Software, Inc. | User definable image reference regions |
-
2012
- 2012-10-31 CN CN201210428391.8A patent/CN103793150B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101599124A (en) * | 2008-06-03 | 2009-12-09 | 汉王科技股份有限公司 | A kind of from video image the method and apparatus of separating character |
| CN101398930A (en) * | 2008-10-24 | 2009-04-01 | 清华大学 | Interactive edition method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103793150A (en) | 2014-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104572735B (en) | A kind of picture mark words recommending method and device | |
| CN108227912A (en) | Apparatus control method and device, electronic equipment, computer storage media | |
| EP3058514B1 (en) | Adding/deleting digital notes from a group | |
| CN106372648A (en) | Multi-feature-fusion-convolutional-neural-network-based plankton image classification method | |
| CN103793683B (en) | Gesture recognition method and electronic device | |
| CN105069754B (en) | System and method based on unmarked augmented reality on the image | |
| JP2014502399A (en) | Handwriting input method by superimposed writing | |
| CN108550107A (en) | A kind of image processing method, picture processing unit and mobile terminal | |
| CN107209631A (en) | User terminal and its method for displaying image for display image | |
| CN104679388B (en) | The method and its mobile terminal of application program are opened by icon duplicate | |
| CN108596955A (en) | A kind of image detecting method, image detection device and mobile terminal | |
| CN106888236A (en) | Conversation managing method and session management device | |
| CN107463331A (en) | Gesture path analogy method, device and electronic equipment | |
| CN110046941A (en) | A kind of face identification method, system and electronic equipment and storage medium | |
| CN107340964A (en) | The animation effect implementation method and device of a kind of view | |
| CN105068735B (en) | The method of adjustment and device of user interface layout | |
| CN101685368A (en) | Method for displaying and browsing layered information | |
| CN109509257A (en) | Indoor floor rank components pattern forming method, terminal and storage medium | |
| CN103839074B (en) | Image classification method based on matching of sketch line segment information and space pyramid | |
| CN110377659A (en) | A kind of intelligence chart recommender system and method | |
| CN104376038A (en) | Position associated text information visualization method based on label cloud | |
| CN106933454A (en) | Display methods and system | |
| CN104281850A (en) | Character area identification method and device | |
| CN107734138A (en) | The display methods and device of notification message, computer installation and storage medium | |
| CN106971201A (en) | Multi-tag sorting technique based on integrated study |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |