CN107103635A - Image layout color matching method - Google Patents
Image layout color matching method Download PDFInfo
- Publication number
- CN107103635A CN107103635A CN201710165372.3A CN201710165372A CN107103635A CN 107103635 A CN107103635 A CN 107103635A CN 201710165372 A CN201710165372 A CN 201710165372A CN 107103635 A CN107103635 A CN 107103635A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- template
- image
- color matching
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000000007 visual effect Effects 0.000 claims abstract description 17
- 238000006116 polymerization reaction Methods 0.000 claims description 13
- 239000003086 colorant Substances 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000001815 facial effect Effects 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 101100323284 Mus musculus Ankrd27 gene Proteins 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 208000035126 Facies Diseases 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000000379 polymerizing effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
The present invention relates to a kind of image layout color matching method, comprise the following steps：Step 01, optimal text point region is chosen for character image in original image, the optimal text point region is the relatively low band of position of visual saliency；Step 02, it is word color matching in the character image；Step 03, output includes the original image of the character image after color matching.In the present invention, realize and easily export satisfied picture while lifting the aesthetic quality of picture.
Description
Technical field
The invention belongs to digital image processing techniques field, be related to the word of computer graphics and Digital Image Processing with
A kind of mixed composition technology of image, and in particular to image layout color matching method.
Background technology
As user is constantly lifted to the individual requirement that image is propagated, image is fused into propagate at present with text
New model.Automatically being added for user picture turns into a new direction --- the vision of image procossing with watermark word attractive in appearance
The Automated Design of media.
The mixed composition of word and image is one of Automatic Typesetting technology urgent problem to be solved.Work main sides in the past
Overweight the Automatic Typesetting problem to character block and image under solid background.This kind of typesetting problem is based on grid (Grid) in design
Concept.Worked with forward part and solve the problems, such as to add the typesetting of other elements in the uniform background region of image.But these work
It is to handle the aesthstic typesetting problem that single text is covered on the picture of uniform background by setting up computation model.
The content of the invention
In order to solve above mentioned problem of the prior art, the present invention proposes a kind of image layout color matching method, to realize
Satisfied picture is easily exported while lifting the aesthetic quality of picture.
This method comprises the following steps：
Step 01, optimal text point region, the optimal text point area are chosen for character image in original image
Domain is the relatively low band of position of visual saliency；
Step 02, it is word color matching in the character image；
Step 03, output includes the original image of the character image after color matching.
Preferably, when choosing optimal text point region, the calculation formula of visual saliency is：
HOFsal(p)=max (1, Hsal(p)+Obj(p)+Face(p))
Wherein, HOFsal(p) it is visual saliency, p is a pixel in any original image, Hsal(p) it is each
The layering notable figure of pixel, Obj(p) it is the object area sign picture of each pixel, Face(p) it is the facial zone table of each pixel
Diagram, Hsal(p)、ObjAnd F (p)ace(p) span is between 0~1.
Preferably, when choosing the optimal text point region for character image, the formula for calculating text point is：
Wherein, R*For optimal text point region, (x, y) is that the rectangular area upper left corner pinpoints coordinate, and w is rectangle frame width
Degree, h is rectangle frame height, Tsp(i, j) is the space importance in (i, j) place pixel.
Preferably, the form and aspect histogram calculation formula of the visual saliency is：
Wherein, M*For form and aspect histogram, the form and aspect histogram M*For original image in chrominance space all pixels point p
Saturation degree and visual saliency value significance weighted distribution, H represents color component,.Represent two matrix correspondence positions
The product of element.
Preferably, after normalized, the histogrammic formula of form and aspect is：
Wherein, M is the form and aspect histogram after normalized.
Preferably, the step 02 is specially：
Select with the immediate template of original image color substep, the original image is carried out according to the template harmonious
Change is handled, and is matched colors accordingly for the word.
Preferably, in modeling plate, harmonious loss of the original image under each template is calculated, formula is：
Wherein, F represents harmonious loss, and I represents original image, and T represents template, and H represents color component, and // // represents certain
It is of the same colour be transferred to the interval minimum colour wheel of given template away from；Tsp(p) pixel p space importance is represented.
Preferably, in modeling plate, harmonious loss of the original image under each template is calculated, formula is：
Wherein, F represents harmonious loss, and I represents original image, TmExpression template, m ∈ { X, Y, V, T }, α represents template
Any anglec of rotation.
Preferably, the selection strategy of the template is：
B (I)=(m, α)
Preferably, the selection strategy of the template includes polymerization and divided, and is specially：
Division：T templates are split into X templates, and V templates are split into Y templates, and X, Y template are without division；
Polymerization：T matrix polymerizations are V templates, and X matrix polymerizations are Y templates, and V, Y template are without polymerization.
Compared with prior art, the present invention at least has advantages below：
By image layout color matching method in the present invention, realize and easily export satisfied picture while lifting picture
Aesthetic quality.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of image layout color matching method provided by the present invention.
Embodiment
The preferred embodiment of the present invention described with reference to the accompanying drawings.It will be apparent to a skilled person that this
A little embodiments are used only for explaining the technical principle of the present invention, it is not intended that limit the scope of the invention.
The method of the present invention is broadly divided into two parts：Character area Selection Strategy and text color Selection Strategy.Using base
When vision significantly theoretical determination text point with aesthstic composition, need to consider picture material (avoiding being blocked by word) with
Typesetting composition skill.Piece image is given, is best suitable for placing and specifies the region of size word to be defined as optimal text point
Region.The region should disturb the expression of artwork content as few as possible, and designing picture composition principle is met as much as possible.View-based access control model
When contrast and harmonious in colourization principle determine text color, it need to ensure that artwork color is harmonious as far as possible, word color and artwork
Produce contrast as big as possible.
It will be detailed below for the specific algorithm of each step.
The invention discloses a kind of image layout color matching method, as shown in figure 1, comprising the following steps：
Step 01, optimal text point region is chosen in original image for character image.
The optimal text point region is the relatively low band of position of visual saliency.In order to ensure that word is as few as possible
The expression of artwork content is disturbed, character area should not be positioned over the region that artwork observer may be interested, in order to avoid cause to regard
Feel is obscured, if this obscure inevitable, should try one's best reduction, i.e., should be positioned over the relatively low region of vision significance.Conventional regards
Contrast of the pixel with periphery background in terms of color, brightness, direction being considered feel conspicuousness more.In addition, such as " people in image
Face ", " human body " or other familiar objects also should be by particular concerns, because being likely to result in when word is positioned over into these regions
" visual noise " and then the effective transmission for influenceing information.
When choosing optimal text point region, the calculation formula of visual saliency is：
HOFsal(p)=max (1, Hsal(p)+Obj(p)+Face(p))
Wherein, HOFsal(p) it is visual saliency, p is a pixel in any original image, Hsal(p) it is each
The layering notable figure of pixel, Obj(p) it is the object area sign picture of each pixel, Face(p) it is the facial zone table of each pixel
Diagram, Hsal(p)、ObjAnd F (p)ace(p) span is between 0~1.When conventional object or people are not present in image
During face region, ObjAnd F (p)ace(p) it is 0；When three adds up more than 1, it is believed that with stronger conspicuousness, by 1 processing,
So HOFsal(p) span is also 0~1, and each notable figure concrete meaning and computational methods are as follows：
1) it is layered notable figure：The present invention reduces small yardstick and high-contrast area to conspicuousness using multi-level simulation tool method
Detect the influence caused.Image is divided into 3 layers by this method, will be processed into different size of hyperfractionated image block, shape per tomographic image
Into tree structure, then the method inferred by level draws Saliency maps.Each element span 0~1 in notable figure is layered,
Value is bigger, and the expression region significance is higher.
2) object area mark figure：Conventional object present in diagram is used to refer to, is extracted using Faster RCNN common
Object area information, then obtains object area mark figure with reference to Grab Cut.Handling process is one width figure of input, is used
Faster RCNN obtain the outer encirclement frame of object area, and the region representation in inframe portion may be prospect or background, and outer frame portion is
Non- prospect, using this frame as Grab Cut, input can obtain object area mark figure.Each element takes in object area mark figure
It is worth scope 0~1, value is bigger to represent that the region there is a possibility that object is higher.
3) facial zone mark figure：Face datection is carried out, if detecting human face region, using Grab Cut according to face
Regional structure Face Map, flow is similar with above-mentioned object area mark figure.Each element value model in facial zone mark figure
0~1 is enclosed, value is bigger to represent that the region there is a possibility that face is higher.
The optimal position of superposition word is calculated automatic, not only to consider that the table of artwork content should be disturbed as few as possible
Reach, designing picture composition principle is also met as much as possible.The spatial locality metric template that the present invention is proposed using khan.The template
Based on three points of composition methods, it is believed that the point in transverse and longitudinal 3-point line is important, in addition, special consideration image reform and up and down
Three quantiles.Brighter area identification relevant position is more notable in template, and darker place represents more inessential., will during calculating
The template zooms to identical with artwork size, and each element is zoomed into 0~1, uses Tsp(i, j) is represented in (i, j) place point
Space importance.
Summary, for any original image Im*nIf optimal text point region is designated as R*, (x, y) is rectangular area
The upper left corner pinpoints coordinate, and w is rectangle width of frame, and h is rectangle frame height.
Above mentioned problem can be reduced to find the maximum submatrix problem of the dimension matrix of non-negative 2, can quickly be asked with dynamic programming algorithm
Solution.
For textcolor Selection Strategy, the present invention is primarily based on the HSV form and aspect histograms that above-mentioned significance calculates image,
Comprise the following steps that：
Form and aspect histogram M* represents original image Im*nAll pixels point p saturation degree and vision is notable in chrominance space
The significance weighted distribution of the value of degree, is defined as：
In above-mentioned formula, H represents color component,Represent the product of two matrix correspondence position elements.
After normalized, the histogrammic formula of form and aspect is：
M is the form and aspect histogram after normalized.
Step 02, it is word color matching in character image.
Specifically, select with the immediate template of original image color substep, according to the template to the original image
Harmonization processing is carried out, and is matched colors accordingly for the word.
Secondly, the present invention also to select with the immediate template of artwork COLOR COMPOSITION THROUGH DISTRIBUTION, then utilize matched template
The harmonization of artwork is completed, and is matched colors on this basis for word, step is as follows：
According to the definition to image harmony degree, with reference to significance proposed above, by a width original image Im*nIn template T
Under harmonious loss be defined as：
Wherein, F represents harmonious loss, and I represents original image, and T represents template, and H represents color component, |/|/represent certain
It is of the same colour be transferred to the interval minimum colour wheel of given template away from；Tsp(p) pixel p space importance is represented.
In modeling plate, harmonious loss of the original image under each template is calculated, formula is：
Wherein, F represents harmonious loss, and I represents original image, TmExpression template, m ∈ { X, Y, V, T }, α represents template
Any anglec of rotation.
Harmonious degree of the image I under each template is provided, a kind of obvious template Selection Strategy is：
B (I)=(m, α)
Harmonious loss of the image under each template is calculated respectively, chooses the corresponding template of minimum harmony degree.
The present invention proposes a kind of adaptive template matching strategy based on polymerization, division.Wherein,
Splitting rule is：T templates are split into X templates, and V templates are split into Y templates, X, and Y templates are without division；
Polymeric rule is：T matrix polymerizations be V templates, X matrix polymerizations be Y templates, V, Y templates without polymerization.
The adaptive template matching algorithm flow is as follows：
Template to be matched is initialized for T templates；
WHILE templates to be matched change,
Original image I is matched with template to be matched,
Mi average Ave and variance Varp in statistical mask region,
Peak concentration point is calculated according to Ave+3*VarpAve+3*Varp；
IF maximum points number is more than 2,
THEN
According to splitting rule classification model, template to be matched is updated；
ELSE
IF is not present in peak concentration point and template area less than average o'clock more than 70,
THEN
According to polymeric rule polymerizing template, template to be matched is updated.
ELSE matchings terminate.
Finally, herein match colors strategy on the basis of the aesthetic property of color consider text can be readability.Textcolor with
Image background color is too similar, influences whether text reading.View-based access control model contrast principle, on the basis of harmonization, using contrast
Color strategy is matched colors for text.
For the template (template X and Y) based on complementary colours facies principle, it is first determined the fan of text filed place template area
Area, secondly calculates the angular bisector of correspondence sector, and it is textcolor to take color corresponding to the contrastive colours of angular bisector.
For the template (template V and T) based on close form and aspect principle, the angular bisector of direct calculation template sector region,
Color corresponding to the contrastive colours of its angular bisector is textcolor.
Step 03, output includes the original image of the character image after color matching.
The characteristic of the method for the present invention and innovation are, it is proposed that a kind of full automatic photo watermark Automatic Typesetting and color matching
Method, when providing the user for text point, the design recommendation of color matching, instructs user to complete picture and text and takes with reference to design principle
With design.Visual salient region and design science relative theory based on image, optimize to character block and are laid out.Based on this, it is sharp
With harmonious in colourization principle, lift picture aesthetic quality and matched colors using visual contrast principle for word.
So far, combined preferred embodiment shown in the drawings describes technical scheme, still, this area
Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these embodiments.Without departing from this
On the premise of the principle of invention, those skilled in the art can make equivalent change or replacement to correlation technique feature, these
Technical scheme after changing or replacing it is fallen within protection scope of the present invention.
Claims (10)
1. a kind of image layout color matching method, it is characterised in that comprise the following steps：
Step 01, optimal text point region is chosen for character image in original image, the optimal text point region is
The relatively low band of position of visual saliency；
Step 02, it is word color matching in the character image；
Step 03, output includes the original image of the character image after color matching.
2. image layout color matching method according to claim 1, it is characterised in that when choosing optimal text point region,
The calculation formula of visual saliency is：
HOFsal(p)=max (1, Hsal(p)+Obj(p)+Face(p))
Wherein, HOFsal(p) it is visual saliency, p is a pixel in any original image, Hsal(p) it is each pixel
Layering notable figure, Obj(p) it is the object area sign picture of each pixel, Face(p) represented for the facial zone of each pixel
Figure, Hsal(p)、ObjAnd F (p)ace(p) span is between 0~1.
3. image layout color matching method according to claim 2, it is characterised in that choosing the optimal text for character image
During word location region, the formula for calculating text point is：
<mrow>
<msup>
<mi>R</mi>
<mo>*</mo>
</msup>
<mo>=</mo>
<mi>arg</mi>
<munder>
<mrow>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
</mrow>
<mi>R</mi>
</munder>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mi>x</mi>
<mo>,</mo>
<mi>j</mi>
<mo>=</mo>
<mi>y</mi>
</mrow>
<mrow>
<mi>x</mi>
<mo>+</mo>
<mi>w</mi>
<mo>,</mo>
<mi>y</mi>
<mo>+</mo>
<mi>h</mi>
</mrow>
</munderover>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>-</mo>
<msub>
<mi>HOF</mi>
<mrow>
<mi>s</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
</msub>
<mo>(</mo>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mo>)</mo>
<mo>+</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>s</mi>
<mi>p</mi>
</mrow>
</msub>
<mo>(</mo>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
Wherein, R*For optimal text point region, (x, y) is that the rectangular area upper left corner pinpoints coordinate, and w is rectangle width of frame, and h is
Rectangle frame height, Tsp(i, j) is the space importance in (i, j) place pixel.
4. image layout color matching method according to claim 2, it is characterised in that the form and aspect histogram of the visual saliency
Calculation formula is：
Wherein, M*For form and aspect histogram, the form and aspect histogram M*For original image in chrominance space all pixels point p it is full
With degree and the distribution of the significance weighted of the value of visual saliency, H represents color component,Represent two matrix correspondence position elements
Product.
5. image layout color matching method according to claim 4, it is characterised in that after normalized, the form and aspect Nogata
The formula of figure is：
<mrow>
<mi>M</mi>
<mo>=</mo>
<mrow>
<mo>{</mo>
<mrow>
<msub>
<mi>M</mi>
<mi>i</mi>
</msub>
<mo>/</mo>
<msub>
<mi>M</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<msubsup>
<mi>M</mi>
<mi>i</mi>
<mo>*</mo>
</msubsup>
<mrow>
<msubsup>
<mi>&Sigma;M</mi>
<mi>i</mi>
<mo>*</mo>
</msubsup>
</mrow>
</mfrac>
</mrow>
<mo>}</mo>
</mrow>
</mrow>
Wherein, M is the form and aspect histogram after normalized.
6. image layout color matching method according to claim 2, it is characterised in that the step 02 is specially：
Select with the immediate template of original image color substep, the original image is carried out at harmonization according to the template
Reason, and matched colors accordingly for the word.
7. image layout color matching method according to claim 6, it is characterised in that in modeling plate, calculates the original graph
As the harmonious loss under each template, formula is：
<mrow>
<mi>F</mi>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>,</mo>
<mi>T</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mi>&Sigma;</mi>
<mrow>
<mi>p</mi>
<mo>&Element;</mo>
<mi>I</mi>
</mrow>
</munder>
<mo>/</mo>
<mo>/</mo>
<mi>H</mi>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>s</mi>
<mi>p</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>)</mo>
</mrow>
<mo>/</mo>
<mo>/</mo>
<mo>*</mo>
<msub>
<mi>HOF</mi>
<mrow>
<mi>s</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>)</mo>
</mrow>
</mrow>
Wherein, F represents harmonious loss, and I represents original image, and T represents template, and H represents color component, and // // represents that certain is of the same colour
Be transferred to the interval minimum colour wheel of given template away from；Tsp(p) pixel p space importance is represented.
8. image layout color matching method according to claim 6, it is characterised in that in modeling plate, calculates the original graph
As the harmonious loss under each template, formula is：
<mrow>
<msub>
<mi>F</mi>
<mi>m</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>,</mo>
<msub>
<mi>T</mi>
<mi>m</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>arg</mi>
<munder>
<mi>min</mi>
<mi>&alpha;</mi>
</munder>
<mi>F</mi>
<mo>(</mo>
<mrow>
<mi>I</mi>
<mo>,</mo>
<msub>
<mi>T</mi>
<mi>m</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>&alpha;</mi>
<mo>)</mo>
</mrow>
</mrow>
<mo>)</mo>
</mrow>
Wherein, F represents harmonious loss, and I represents original image, TmTemplate is represented, m ∈ { X, Y, V, T }, α represents any rotation of template
Gyration.
9. according to the described image typesetting color matching method of claim 7 or 8, it is characterised in that the selection strategy of the template is：
B (I)=(m, α)
<mrow>
<mo>(</mo>
<mi>m</mi>
<mo>,</mo>
<mi>&alpha;</mi>
<mo>)</mo>
<mo>=</mo>
<mi>arg</mi>
<munder>
<mi>min</mi>
<mi>m</mi>
</munder>
<msub>
<mi>F</mi>
<mi>m</mi>
</msub>
<mo>(</mo>
<mi>I</mi>
<mo>)</mo>
</mrow>
10. image layout color matching method according to claim 9, it is characterised in that the selection strategy of the template includes poly-
Close and divide, be specially：
Division：T templates are split into X templates, and V templates are split into Y templates, and X, Y template are without division；
Polymerization：T matrix polymerizations are V templates, and X matrix polymerizations are Y templates, and V, Y template are without polymerization.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710165372.3A CN107103635A (en) | 2017-03-20 | 2017-03-20 | Image layout color matching method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710165372.3A CN107103635A (en) | 2017-03-20 | 2017-03-20 | Image layout color matching method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107103635A true CN107103635A (en) | 2017-08-29 |
Family
ID=59675356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710165372.3A Pending CN107103635A (en) | 2017-03-20 | 2017-03-20 | Image layout color matching method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107103635A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009712A (en) * | 2019-03-01 | 2019-07-12 | 华为技术有限公司 | A kind of picture and text composition method and its relevant apparatus |
| CN110264545A (en) * | 2019-06-19 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Picture Generation Method, device, electronic equipment and storage medium |
| CN110706310A (en) * | 2019-08-23 | 2020-01-17 | 华为技术有限公司 | Image-text fusion method and device and electronic equipment |
| CN114491092A (en) * | 2022-01-26 | 2022-05-13 | 深圳市前海手绘科技文化有限公司 | Method and system for recommending materials according to document content and color matching |
| CN117669493A (en) * | 2023-12-08 | 2024-03-08 | 安徽省医学情报研究所 | Intelligent image-text typesetting method and system based on significance detection |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101253759A (en) * | 2005-05-30 | 2008-08-27 | 富士胶片株式会社 | Album creating apparatus, album creating method and computer readable medium |
-
2017
- 2017-03-20 CN CN201710165372.3A patent/CN107103635A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101253759A (en) * | 2005-05-30 | 2008-08-27 | 富士胶片株式会社 | Album creating apparatus, album creating method and computer readable medium |
Non-Patent Citations (1)
| Title |
|---|
| 孟一平 等: "照片水印自动排版与配色", 《中国图象图形学报》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009712A (en) * | 2019-03-01 | 2019-07-12 | 华为技术有限公司 | A kind of picture and text composition method and its relevant apparatus |
| WO2020177584A1 (en) * | 2019-03-01 | 2020-09-10 | 华为技术有限公司 | Graphic typesetting method and related device |
| EP3910598A4 (en) * | 2019-03-01 | 2022-03-30 | Huawei Technologies Co., Ltd. | Graphic typesetting method and related device |
| US11790584B2 (en) | 2019-03-01 | 2023-10-17 | Huawei Technologies Co., Ltd. | Image and text typesetting method and related apparatus thereof |
| CN110264545A (en) * | 2019-06-19 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Picture Generation Method, device, electronic equipment and storage medium |
| WO2020253766A1 (en) * | 2019-06-19 | 2020-12-24 | 北京字节跳动网络技术有限公司 | Picture generation method and apparatus, electronic device, and storage medium |
| CN110706310A (en) * | 2019-08-23 | 2020-01-17 | 华为技术有限公司 | Image-text fusion method and device and electronic equipment |
| WO2021036715A1 (en) * | 2019-08-23 | 2021-03-04 | 华为技术有限公司 | Image-text fusion method and apparatus, and electronic device |
| CN110706310B (en) * | 2019-08-23 | 2021-10-22 | 华为技术有限公司 | Image-text fusion method and device and electronic equipment |
| CN114491092A (en) * | 2022-01-26 | 2022-05-13 | 深圳市前海手绘科技文化有限公司 | Method and system for recommending materials according to document content and color matching |
| CN114491092B (en) * | 2022-01-26 | 2023-02-10 | 深圳市前海手绘科技文化有限公司 | Method and system for recommending materials according to document contents and color matching |
| CN117669493A (en) * | 2023-12-08 | 2024-03-08 | 安徽省医学情报研究所 | Intelligent image-text typesetting method and system based on significance detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107103635A (en) | Image layout color matching method | |
| CN103971126B (en) | A kind of traffic sign recognition method and device | |
| US9691145B2 (en) | Methods and systems for automated selection of regions of an image for secondary finishing and generation of mask image of same | |
| CN102567731B (en) | Extraction method for region of interest | |
| CN102024156B (en) | Method for positioning lip region in color face image | |
| CN102663806B (en) | Artistic-vision-based cartoon stylized rendering method of image | |
| CN104794693B (en) | A kind of portrait optimization method of face key area automatic detection masking-out | |
| CN103093447B (en) | Cutting and splicing method of concentration of pictures of computer | |
| CN108010034A (en) | Commodity image dividing method and device | |
| CN106991462A (en) | Three-dimensional code generating method | |
| CN104063888B (en) | A kind of wave spectrum artistic style method for drafting based on feeling of unreality | |
| CN107945244A (en) | A kind of simple picture generation method based on human face photo | |
| CN107031033A (en) | It is a kind of can 3D printing hollow out Quick Response Code model generating method and system | |
| CN105023253A (en) | Visual underlying feature-based image enhancement method | |
| CN106530317B (en) | A kind of scoring of simple picture computer and auxiliary painting methods | |
| CN108427828A (en) | A kind of device of automatic assessment planar design placement quality and optimization | |
| CN103946868A (en) | Processing method and system for medical images | |
| CN111860369A (en) | Fraud identification method and device and storage medium | |
| CN107481206A (en) | MIcrosope image background equalization Processing Algorithm | |
| CN103793549B (en) | Adopt the computer-aided crewel embroidery production method of fuzzy clustering and random walk | |
| CN107133562A (en) | A kind of gesture identification method based on extreme learning machine | |
| CN103778430A (en) | Rapid face detection method based on combination between skin color segmentation and AdaBoost | |
| CN104809721B (en) | A kind of caricature dividing method and device | |
| CN106846399A (en) | A kind of method and device of the vision center of gravity for obtaining image | |
| CN109726633A (en) | A kind of face critical point detection method based on look-up table activation primitive |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170829 |
|
| WD01 | Invention patent application deemed withdrawn after publication |