CN112818634A - Calligraphy work style migration system, method and terminal - Google Patents
Calligraphy work style migration system, method and terminal Download PDFInfo
- Publication number
- CN112818634A CN112818634A CN202110127483.1A CN202110127483A CN112818634A CN 112818634 A CN112818634 A CN 112818634A CN 202110127483 A CN202110127483 A CN 202110127483A CN 112818634 A CN112818634 A CN 112818634A
- Authority
- CN
- China
- Prior art keywords
- calligraphy
- style
- target
- character
- original
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000005012 migration Effects 0.000 title claims abstract description 39
- 238000013508 migration Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 37
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 37
- 239000013598 vector Substances 0.000 claims description 27
- 230000005484 gravity Effects 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000002194 synthesizing effect Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 150000001875 compounds Chemical class 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000695274 Processa Species 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/103—Formatting, i.e. changing of presentation of documents
- G06F40/106—Display of layout of documents; Previewing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/38—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/383—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/103—Formatting, i.e. changing of presentation of documents
- G06F40/109—Font handling; Temporal or kinetic typography
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Library & Information Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Controls And Circuits For Display Device (AREA)
Abstract
The invention discloses a calligraphy work style migration system and a mode, wherein the system comprises: the system comprises a database, a retrieval module, a single character synthesis module, a radical synthesis module and a chapter migration module; the method can realize the calligraphy style migration of all the original calligraphy characters in the original calligraphy works and simultaneously perform the page layout on the generated target calligraphy style works to finish the migration from the font style to the page layout by extracting the calligraphy style characteristic components of the original calligraphy characters from the three layers of the calligraphies, the knot bodies and the calligraphies, and performing whole character retrieval, radical retrieval and synthesis, simplified retrieval and synthesis on the original calligraphy characters. The method and the device can generate various calligraphy style fonts with clear font styles and meeting the requirements of users, can also typeset calligraphy works, and provide a feasible thought for the generation of personalized calligraphy works.
Description
Technical Field
The invention relates to the field of image processing, in particular to a calligraphy work style migration system, a calligraphy work style migration method and a calligraphy work style migration terminal.
Background
With the acceleration of the process of electronic informatization, electronic devices such as computers and mobile phones are greatly convenient for people to live and work, but because the types of the fonts built in the electronic devices are limited, the fonts used by people are thousands of uniform, the personalization is lost, and particularly when an electronic mail or chatting software is used for writing chats, the effect of seeing characters like the face is difficult to achieve, so that a method capable of generating the personalized style fonts is required to be found.
Disclosure of Invention
The calligraphy work style migration system, the method and the terminal can generate the personalized calligraphy word stock and synthesize the calligraphy works with the designated style.
In order to achieve the above object, the present invention provides a calligraphy work style migration system, comprising:
the database is used for classifying and storing the calligraphy characters in the calligraphy works;
the retrieval module is connected with the database and is used for analyzing the original calligraphy works and retrieving target calligraphy characters in a target calligraphy style corresponding to the original calligraphy characters in the original calligraphy works from the database;
the individual character synthesis module is connected with the retrieval module and is used for decomposing radicals of the original calligraphy characters, retrieving target radicals of a target calligraphy style corresponding to the original radicals of the original calligraphy characters from a database and synthesizing the target calligraphy characters;
the system comprises a single character synthesis module, a radical synthesis module and a database, wherein the single character synthesis module is connected with the single character synthesis module and is used for carrying out stroke decomposition on an original radical, retrieving target strokes of a target style corresponding to the original strokes of the original radical from the database and synthesizing the target radical;
and the seal migration module is connected with the retrieval module and is used for carrying out seal layout on the target calligraphy character.
Further, the database includes:
a calligraphy style database for storing calligraphy words of different calligraphy styles;
a radical database for storing radicals of calligraphic words of different calligraphic styles;
a stroke database for storing strokes of radicals of different calligraphy styles.
The invention also provides a calligraphy work style migration method, which comprises the following steps:
s1, the retrieval module sequentially retrieves the middle calligraphy characters corresponding to the original calligraphy characters in the original calligraphy works, and selects the middle calligraphy character closest to the target calligraphy style as the target calligraphy character to migrate the original calligraphy character;
if the retrieval module does not retrieve the middle calligraphy character corresponding to the original calligraphy character or the style of the middle calligraphy character does not accord with the target calligraphy style requirement, executing the step S2;
s2, the single character synthesis module searches the middle radicals corresponding to the original radicals of the original calligraphy characters and synthesizes the middle calligraphy characters; if the style of the middle calligraphy character meets the style requirement of the target calligraphy, taking the middle calligraphy character as the target calligraphy character to migrate the original calligraphy character;
if the single character synthesis module does not retrieve all radicals forming the original calligraphy character or the style of the synthesized intermediate calligraphy character does not meet the style requirement of the target calligraphy, executing step S3;
s3, the radical synthesizing module searches the middle strokes corresponding to the original strokes of the original radicals, synthesizes all the middle radicals forming the original calligraphy characters and then synthesizes the middle calligraphy characters again; if the style of the middle calligraphy character meets the style requirement of the target calligraphy, taking the middle calligraphy character as the target calligraphy character to migrate the original calligraphy character;
if the style of the synthesized intermediate calligraphy character does not meet the style requirement of the target calligraphy, the step S3 is repeatedly executed;
and S4, generating a target calligraphy style work after all the original calligraphy characters in the original calligraphy work are migrated, and performing page layout on the generated target calligraphy style work by the chapter migration module.
Further, in step S1, the retrieving module selects an intermediate calligraphy word closest to the style of the target calligraphy from the intermediate calligraphy words as the target calligraphy word through style similarity calculation;
the style similarity calculation formula is as follows:
SIMC=a1Z(z'i,z'j)+a2T(t'i,t'j)+a3S(s'i,s'j)+a4G(wi,wj)
in the formula, alpha1+α2+α3+α4=1,
Z(z'i,z'j),K(t'i,t'j),S(s'i,s'j) Respectively representing whether the middle calligraphy character and the calligraphy character with the target calligraphy style have similarity in the chapter, knot and stroke characteristics, G (w)i,wj) Indicating whether two words belong to the same work, alpha1、α2、α3、α4Are all constant parameters, s'i,z'i,t'i,wiChapter feature vector, knot feature vector, stroke feature vector and belonging work, s'j,z'j,t'j,wjAnd respectively representing a chapter characteristic vector, a knot characteristic vector, a stroke characteristic vector and the work of the calligraphy character with the target calligraphy style.
Further, in the step S2, the individual character retrieving module retrieves an intermediate radical from the radical database, which corresponds to the original radical of the original calligraphy character and meets the requirement of the target calligraphy style, by using the target calligraphy character retrieved in the step S1 as the target calligraphy style reference character and combining with the matching degree calculation;
the matching degree calculation formula is as follows:
MR=λ1RC(ct,c)+λ2RP(pt,p)+λ3RW(rwt,Wf)+λ4RV(vt,v)+λ5RG(wt,wf)
in the formula (I), the compound is shown in the specification,
λ1+λ2+λ3+λ4+λ5=1,
RC(ct,c)=|ct-c|,
RW(rwt,Wf)=|rwt-Wf|,
RV(vt,v)=|vt-v|,
Wf,wffor the target calligraphy style reference word's average stroke width and resulting work, RC (c)tC) indicates whether the stroke complexity of the retrieved radical is the same as that of the radical of the target calligraphy style reference word, RP (p)tP) denotes whether the retrieved radical is in the same position as the radical of the target calligraphy style reference word, RW (RW)t,Wf) RV (v) indicating whether the stroke width of the retrieved radical is the same as the stroke width of the radical of the target calligraphy style reference wordtV) indicates whether the retrieved radical has the same aspect ratio, RG (w), as the radical of the target calligraphy style reference wordt,wf) Denotes whether the target radical and the target reference word are from the same work, lambda1、λ2、λ3、λ4、λ5Are all constant parameters, ct,pt,rwt,vt,wtAnd c, p, rw, v respectively represent the radical stroke complexity, the radical position, the radical stroke width and the radical aspect ratio of the target calligraphy style reference character.
Further, the method for synthesizing the intermediate calligraphy character by the intermediate radicals comprises the following steps:
the middle radicals are spatially arranged according to the positions of the original radicals of the original calligraphy characters in the calligraphy characters;
setting an initial distance between the middle radicals; so as to avoid the adhesion between the middle radicals;
adjusting the relative position of each middle radical in the middle calligraphy characters; to ensure that the center of gravity of the middle calligraphy character tends to the average center of gravity of the target calligraphy style;
adjusting the size of the synthesized intermediate calligraphy character; so as to ensure that the height, width and height-width ratio of the synthesized middle calligraphy character is consistent with the style of the target calligraphy;
when the relative position of each middle radical is adjusted, the following constraint conditions are met:
|G's.x-Gs.x|≤a1
|G's.y-Gs.y|≤a2
in formula (II), G'x,G'yRespectively showing the horizontal and vertical coordinates of the center of gravity of the middle calligraphy character after the first adjustment position of the middle part,respectively representing target calligraphy stylesAbscissa and ordinate of mean center of gravity of all calligraphy characters in work in which reference character is located, Gs.x,Gs.yHorizontal and vertical coordinates G 'each representing an initial center of gravity of a middle head of a synthesized middle calligraphy character's.x,G's.yRespectively, the horizontal and vertical coordinates a of the center of gravity after the first adjustment of the middle part1,a2Is a constant number, σf.x,σf.yRespectively representing the standard deviation, delta, of the horizontal and vertical coordinates of the center of gravityx,δyAre all threshold constant parameters.
Further, in step S3, the middle part of the radical synthesizing module searches, through metric calculation and in combination with the target calligraphy style reference word, the middle strokes corresponding to the original strokes of the original radical and meeting the target calligraphy style requirements from the stroke database;
the formula of the metric calculation is:
in the formula (I), the compound is shown in the specification,
ω1+ω2+ω3+ω4+ω5=1,
B(b,b')=|b-b'|,
Q(q,q')=|q-q'|,
SV(sv,sv')=|sv-sv'|,
SP(sp,sp')=√(spx-sp'x)2+(spy-sp'y)2,
B(b,b'),Q(q,q'),SV(sv,sv'),SP(sp,sp'),respectively representing the similarity between the middle stroke and the target calligraphy style reference character in the head-tail width, the twisting degree, the height-width ratio, the position and the width of the stroke, omega1、ω2、ω3、ω4、ω5B, q, sv and sp respectively represent the stroke head width, the stroke twisting degree, the stroke height-width ratio and the stroke position of the stroke of the target calligraphy style reference character, b ', q', sv ', sp' and sw respectively represent the stroke head width, the stroke twisting degree, the stroke height-width ratio, the stroke position and the stroke width of the middle stroke,represents the average stroke width of all the calligraphic words in the calligraphic work where the target calligraphy style reference word is located.
Further, in step S3, when the middle radicals of the original calligraphy characters are synthesized from the middle strokes, the middle strokes are synthesized according to the positions in the radicals of the standard book body by using the standard book body with the similar style as the target calligraphy as a reference.
Further, in step S4, when the chapter migration module performs page layout on the target calligraphy style works, analyzing the row-column distance of the calligraphy works similar to the target calligraphy style, so as to adjust the row-column distance of the target calligraphy style works for reference;
when the chapter migration module adjusts the row-column distance of the target calligraphy style works, the following constraint conditions are met:
|Gnx-Href|≤ε
wherein ε is a threshold constant, HrefIs the center of gravity abscissa, G, of the first word of each line of the target calligraphy style worknxThe barycentric coordinate after adjusting for the rest characters of the row;
|D'nrow-D'row|≤σr
|D'ncol-D'col|≤σc
D'nrow,D'ncolrespectively the adjusted line and column distances, D'row,D'colRespectively the row-column spacing, σ, of the reference calligraphic workrAnd σcRespectively, the standard deviation of the target calligraphy style row-column distance.
The invention also provides a calligraphy work style migration terminal, which comprises: the system comprises a processor, a memory and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the calligraphy work style migration method.
The invention has the following advantages:
the method extracts the characteristics of the original calligraphy characters from the three levels of the calligraphic characters, the knot bodies and the calligraphic, and can realize calligraphy style migration of all the original calligraphy characters in the original calligraphy works and simultaneously perform page layout on the generated target calligraphy style works after whole character retrieval, radical retrieval and synthesis, simplified retrieval and synthesis of the original calligraphy characters, thereby completing the migration from the font style to the page layout. The method can generate various calligraphy styles with clear font styles and meeting the requirements of users, can also typeset calligraphy works, and provides a feasible thought for the generation of personalized calligraphy works.
Drawings
FIG. 1 is a schematic diagram of a calligraphic work style migration system;
FIG. 2 is a schematic diagram of a method for style migration of a calligraphic work;
FIG. 3 is a schematic diagram illustrating the adjustment of the head position of the middle calligraphic part;
FIG. 4 is a schematic diagram of the process of synthesizing intermediate radicals.
Detailed Description
The invention is further described in detail below with reference to the following figures and specific examples. Advantages and features of the present invention will become apparent from the following description and from the claims. It is to be noted that the drawings are in a very simplified form and are all used in a non-precise ratio for the purpose of facilitating and distinctly aiding in the description of the embodiments of the present invention.
As shown in FIG. 1, the present invention provides a calligraphy work style migration system, comprising:
the database is used for classifying and storing the calligraphy characters in the calligraphy works;
the retrieval module is connected with the database and is used for analyzing the original calligraphy works and retrieving target calligraphy characters in a target calligraphy style corresponding to the original calligraphy characters in the original calligraphy works from the database;
the individual character synthesis module is connected with the retrieval module and is used for decomposing radicals of the original calligraphy characters, retrieving target radicals of a target calligraphy style corresponding to the original radicals of the original calligraphy characters from a database and synthesizing the target calligraphy characters;
the system comprises a single character synthesis module, a radical synthesis module and a database, wherein the single character synthesis module is connected with the single character synthesis module and is used for carrying out stroke decomposition on an original radical, retrieving target strokes of a target style corresponding to the original strokes of the original radical from the database and synthesizing the target radical;
and the seal migration module is connected with the retrieval module and is used for carrying out seal layout on the target calligraphy character.
The database, comprising:
a calligraphy style database for storing calligraphy words of different calligraphy styles;
a radical database for storing radicals of calligraphic words of different calligraphic styles;
a stroke database for storing strokes of radicals of different calligraphy styles.
As shown in fig. 2, the present invention further provides a method for transferring styles of calligraphy works, comprising the following steps:
s1, the retrieval module sequentially retrieves middle calligraphy characters corresponding to the original calligraphy characters in the original calligraphy works, and selects the middle calligraphy character closest to the style of the target calligraphy as the target calligraphy character to migrate the original calligraphy character through style similarity;
if the retrieval module does not retrieve the middle calligraphy character corresponding to the original calligraphy character or the style of the middle calligraphy character does not meet the target calligraphy style requirement, the step S2 is executed.
Specifically, the user may autonomously select the original calligraphy work and the target calligraphy style to be migrated. The retrieval module carries out page analysis on the original calligraphy work, so that a plurality of original calligraphy characters contained in the original calligraphy work are extracted. Meanwhile, the retrieval module retrieves the calligraphy works with the calligraphy style requirements from the database according to the target calligraphy style requirements, extracts all calligraphy characters in the calligraphy works, analyzes the calligraphy feature components, calculates the average value of the calligraphy feature components of all calligraphy characters in the calligraphy works, and uses the average value as the feature vector of the target calligraphy style. Specifically, the calligraphy feature component includes: chapter method, knot body and pen method.
And the retrieval module sequentially retrieves whether intermediate calligraphy characters corresponding to the original calligraphy characters exist from the calligraphy style database according to the character information of each original calligraphy character. If the retrieval is successful, namely at least one middle calligraphy character corresponding to the calligraphy character in the original calligraphy work exists in the calligraphy style database, the retrieval module performs style similarity calculation through calligraphy style characteristic components of the calligraphy characters passing through three levels of the chapter, the knot and the stroke, and selects the middle calligraphy character closest to the target calligraphy style from the retrieved middle calligraphy characters as the target calligraphy character to migrate the original calligraphy character.
The features z of the chapter include the spacing D of the wordscPromoting the circulation of qi H; the knot body characteristics t comprise a height-width ratio V, an area ratio S, a gravity center G and an average pen width W; the stroke characteristics s include the head-to-tail width Z of the strokeσStroke twist degree b, stroke weight change q, stroke aspect ratio sv and stroke position ratio sp. And normalizing the calligraphy style feature components to express feature vectors of calligraphy characters. The expression of the characteristic vector of the calligraphy character is as follows: c ═ t ', s ', w, where w represents the work to which the calligraphy word belongs, z ═ D 'c,H'), t'=(V',S',G',W'),s'=(Z'σB ', q', sv ', sp') are the feature vector of the seal method, the feature vector of the knot body and the feature vector of the pen method after normalization processing respectively.
Is provided with Ci=(z'i,t'i,s'i,wi),Cj=(z'j,t'j,s'j,wj) Is a feature vector of two different calligraphy words, the style similarity between the two different calligraphy words is calculated by the following formula:
SIMC=a1Z(z'i,z'j)+a2T(t'i,t'j)+a3S(s'i,s'j)+a4G(wi,wj)
in the formula, alpha1+α2+α3+α4=1
Z(z'i,z'j),K(t'i,t'j),S(s'i,s'j) Respectively representing whether the middle calligraphy character and the calligraphy character with the target calligraphy style have similarity in the chapter, knot and stroke characteristics, G (w)i,wj) Indicating whether two words belong to the same work, alpha1、α2、α3、α4Are all constant parameters, s'i,z'i,t'i,wiChapter feature vector, knot feature vector, stroke feature vector and belonging work, s'j,z'j,t'j,wjAnd respectively representing a chapter characteristic vector, a knot characteristic vector, a stroke characteristic vector and the work of the calligraphy character with the target calligraphy style.
The smaller the style similarity value, the more similar the style of the intermediate calligraphy character is to the target calligraphy style. And selecting the middle calligraphy character with the minimum stroke similarity value in each middle calligraphy character as a target calligraphy character to perform style migration on the original calligraphy character. If the style of the middle calligraphy character does not meet the style requirement of the target calligraphy character, the step S2 is executed.
If the retrieval module does not retrieve the intermediate calligraphy word corresponding to the original calligraphy word, step S2 is executed.
S2, the single character synthesis module searches the middle radicals corresponding to the original radicals of the original calligraphy characters through matching degree calculation and synthesizes the middle calligraphy characters; if the style of the middle calligraphy character meets the style requirement of the target calligraphy, taking the middle calligraphy character as the target calligraphy character to migrate the original calligraphy character;
if the individual character synthesis module does not retrieve all the radicals constituting the original calligraphy character, or the style of the synthesized intermediate calligraphy character does not meet the style requirement of the target calligraphy, step S3 is executed.
Specifically, the single-character synthesis module performs radical decomposition search on the original calligraphy characters which are not successfully searched in the step S1 or the original calligraphy characters which have been searched for the middle calligraphy character but have style similarity with the target calligraphy style which does not meet the requirement. And simultaneously, the intermediate calligraphy word closest to the target calligraphy style retrieved in step S1 is taken as the target calligraphy style reference word, and the target calligraphy style reference word is subjected to radical decomposition.
The single character synthesis module searches whether all radicals forming the original calligraphy character exist in a radical database, if so, the matching degree is calculated according to five characteristics of multi-radical stroke complexity c, radical position p, radical stroke width rw, radical height-width ratio v and the work w to which the radicals belong, and the radical characteristic of the target calligraphy style reference character is used as a reference to select the radical closest to the target calligraphy style from the searched radicals as the middle radical of the original calligraphy character.
The matching degree calculation formula is as follows:
MR=λ1RC(ct,c)+λ2RP(pt,p)+λ3RW(rwt,Wf)+λ4RV(vt,v)+λ5RG(wt,wf)
in the formula (I), the compound is shown in the specification,
λ1+λ2+λ3+λ4+λ5=1,
RC(ct,c)=|ct-c|,
RW(rwt,Wf)=|rwt-Wf|,
RV(vt,v)=|vt-v|,
Wf,wffor the target calligraphy style reference word's average stroke width and resulting work, RC (c)tC) indicates whether the stroke complexity of the retrieved radical is the same as that of the radical of the target calligraphy style reference word, RP (p)tP) denotes whether the retrieved radical is in the same position as the radical of the target calligraphy style reference word, RW (RW)t,Wf) RV (v) indicating whether the stroke width of the retrieved radical is the same as the stroke width of the radical of the target calligraphy style reference wordtV) indicates whether the retrieved radical has the same aspect ratio, RG (w), as the radical of the target calligraphy style reference wordt,wf) Denotes whether the target radical and the target reference word are from the same work, lambda1、λ2、λ3、λ4、λ5Are all constant parameters, ct,pt,rwt,vt,wtAnd c, p, rw, v respectively represent the radical stroke complexity, the radical position, the radical stroke width and the radical aspect ratio of the target calligraphy style reference character.
After all the middle radicals forming the original calligraphy character are searched, the original radicals of the original calligraphy character are synthesized according to the position of the original radicals in the calligraphy character. Because the size of the middle radical is not consistent with that of the original radical, the situation that the middle radicals are partially overlapped may occur in the synthesis processAs shown in fig. 3a, therefore, in the preliminary synthesis process, an initial distance dr needs to be added between the middle radicals, as shown in fig. 3 b. Although the initial distance is increased to avoid the radical conglutination, the style of the whole character is different from the target style, and the position of the middle radical and the size of the synthesized middle calligraphy character need to be further adjusted, as shown in fig. 3 c. Let the initial center of gravity of the middle radical of the middle calligraphy character be (G)s.x,Gs.y) The center of gravity after the position adjustment of the first intermediate portion is (G's.x,G's.y) The center of gravity of the middle calligraphy character after the position of the middle part is adjusted first is (G'x,G'y) The average gravity center of all calligraphy characters in the work of the target calligraphy style reference character isStandard deviation of center of gravity of (σ)f.x,σf.y) Taking the middle radical with the minimum coordinate value of the top left vertex of the minimum bounding box in all the middle radicals as a reference radical, adjusting the positions of other middle radicals, and checking whether the gravity center of the whole middle calligraphy character after adjustment tends to the average gravity center of the target style by using the following formula:
if the center of gravity position of the middle calligraphy character is not in the acceptable range, continuing to adjust, wherein the horizontal and vertical coordinate adjustment step length is 1 pixel unit until the center of gravity position of the middle calligraphy character falls into the ideal range; in order to ensure that the relative positions of the middle radicals in the calligraphy characters in the adjusting process are not excessively deviated, the following formula is used for constraint:
|G's.x-Gs.x|≤a1
|G's.y-Gs.y|≤a2
in the formula, a1,a2Is a constant.
After the position of the middle radical is adjusted, the height, width and aspect ratio of the synthesized middle calligraphy character are often inconsistent with the style of the target calligraphy, the middle calligraphy character needs to be zoomed, and the size of the middle calligraphy character is adjusted according to the following formula:
wherein, W's,H's,V'sRespectively the adjusted width, height and height-width ratio of the synthesized middle calligraphy character, respectively the average width, height and aspect ratio of all the characters in the work of the target calligraphy style reference character, and the standard deviation of the average width, height and aspect ratio is
And performing style similarity calculation on the synthesized intermediate calligraphy character through a retrieval module, and if the synthesized intermediate calligraphy character is similar to the style conforming to the style of the target calligraphy, taking the intermediate calligraphy character as the target calligraphy character to migrate the original calligraphy character. If the style of the synthesized intermediate calligraphy character does not meet the style requirement of the target calligraphy, or the single character synthesis module does not retrieve all radicals forming the original calligraphy character, step S3 is executed.
S3, the radical synthesis module searches the middle strokes corresponding to the original strokes of the original radicals through measurement calculation, synthesizes all the middle radicals forming the original calligraphy characters and synthesizes the middle calligraphy characters again; if the style of the middle calligraphy character meets the style requirement of the target calligraphy, taking the middle calligraphy character as the target calligraphy character to migrate the original calligraphy character;
if the style of the synthesized intermediate calligraphy character does not meet the target calligraphy style requirement, the step S3 is repeatedly executed.
Specifically, the radical synthesis module performs stroke decomposition on the original radicals forming the original calligraphy character to obtain an original stroke, and decomposes a reference radical of the target calligraphy style reference character into a reference stroke. The radical synthesis module retrieves strokes corresponding to original strokes forming the original radicals in a stroke database, measures and calculates five characteristic values of stroke head-tail width b, stroke twisting degree q, stroke height-width ratio sv, stroke position sp and stroke width sw, and selects strokes closest to the target calligraphy style from the retrieved strokes as middle strokes of the original radicals by taking the stroke characteristics of the target calligraphy style reference characters as references.
And setting the intermediate stroke feature vector as S ═ b, q, sv and sp, setting the target stroke feature vector as S ═ b ', q', sv ', sp' sw, and setting the average stroke width of the work where the target calligraphy style reference word is located asThe metric calculation formula is as follows:
in the formula, ω1+ω2+ω3+ω4+ω5=1;
B(b,b')=|b-b'|;
Q(q,q')=|q-q'|;
SV(sv,sv')=|sv-sv'|;
SP(sp,sp')=√(spx-sp'x)2+(spy-sp'y)2;
B(b,b'),Q(q,q'),SV(sv,sv'),SP(sp,sp'),Respectively calculating the similarity between the middle stroke and the target calligraphy style reference character in the head-tail width, twisting degree, height-width ratio, position and width of stroke, omega1、ω2、ω3、ω4、ω5B, q, sv and sp respectively represent the stroke head width, the stroke twisting degree, the stroke height-width ratio and the stroke position of the stroke of the target calligraphy style reference character, b ', q', sv ', sp' and sw respectively represent the stroke head width, the stroke twisting degree, the stroke height-width ratio, the stroke position and the stroke width of the middle stroke,represents the average stroke width of all the calligraphic words in the calligraphic work where the target calligraphy style reference word is located.
After the intermediate stroke retrieval is completed, a standard book body with a style similar to that of the target calligraphy is used as a reference, and the positions of all intermediate strokes are laid out by referring to the positions of all strokes forming the radicals in the standard book body, so that the intermediate radicals are synthesized. As shown in fig. 4, taking the example of synthesizing the "lone" character, the "lone" character is first obtained from the standard book block file, and then the positions of the standard strokes forming the "lone" character in the standard book block are obtained and represented by the stroke minimum bounding box. And zooming the middle strokes into the corresponding standard stroke sizes, and synthesizing the middle strokes according to the positions in the radicals of the standard book body to obtain the complete middle radicals.
After the middle radicals of the original calligraphy characters are synthesized, the middle radicals are synthesized into the middle calligraphy characters by adopting a radical synthesis module, and style similarity calculation is carried out on the middle calligraphy characters by adopting a retrieval module. And if the style of the synthesized intermediate calligraphy character is similar to the style of the target calligraphy character, taking the intermediate calligraphy character as the target calligraphy character to migrate the original calligraphy character. And if the style of the synthesized intermediate calligraphy character does not meet the style requirement of the target calligraphy, repeating the steps until the intermediate calligraphy character meeting the style requirement of the target calligraphy is synthesized.
And S4, generating a target calligraphy style work after all the original calligraphy characters in the original calligraphy work are migrated, and performing page layout on the generated target calligraphy style work by the chapter migration module.
And the chapter migration module analyzes the row-column distance of the calligraphy works similar to the target calligraphy style in the database, adjusts the row-column distance of the generated target calligraphy style works by taking the row-column distance as reference, and ensures that each row of characters of the target calligraphy style works have the effect of promoting qi.
Setting the barycentric abscissa H of the first character of each line of the generated target calligraphy style worksrefThe base line of the line, the barycentric coordinate of the adjusted rest characters of the line is GnxThe adjusted row-column pitch is (D'nrow,D'ncol) Reference row-column pitch is (D'row,D'col) To ensure that each line of words is always substantially collinear during each adjustment, the following equation is used for the constraint:
|Gnx-Href|≤ε
where ε is a threshold constant.
In order to satisfy that the row-column distance after each adjustment always approaches the target style, the adjustment is carried out by using the following formula:
|D'nrow-D'row|≤σr
|D'ncol-D'col|≤σc
σrand σcRespectively, the standard deviation of the target calligraphy style row-column distance.
The method is used for evaluating the generated target calligraphy style works, and if the target calligraphy style works are not satisfactory, weights of three calligraphy style characteristic components of a chapter method, a knot body and a stroke method can be appointed to synthesize the target calligraphy style works of the user psychography again so as to increase the personalized design function of the calligraphy works style.
The invention also provides a calligraphy work style migration terminal, which comprises: the system comprises a processor, a memory and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the calligraphy work style migration method.
While the present invention has been described in detail with reference to the preferred embodiments thereof, it should be understood that the above description should not be taken as limiting the invention. Various modifications and alterations to this invention will become apparent to those skilled in the art upon reading the foregoing description. Accordingly, the scope of the invention should be determined from the following claims.
Claims (10)
1. A calligraphy work style migration system, comprising:
the database is used for classifying and storing the calligraphy characters in the calligraphy works;
the retrieval module is connected with the database and is used for analyzing the original calligraphy works and retrieving target calligraphy characters in a target calligraphy style corresponding to the original calligraphy characters in the original calligraphy works from the database;
the single character synthesis module is connected with the retrieval module and is used for decomposing radicals of the original calligraphy characters, retrieving target radicals of a target calligraphy style corresponding to the original radicals of the original calligraphy characters from a database and synthesizing the target calligraphy characters;
the system comprises a single character synthesis module, a radical synthesis module and a database, wherein the single character synthesis module is connected with the single character synthesis module and is used for carrying out stroke decomposition on an original radical, retrieving target strokes of a target style corresponding to the original strokes of the original radical from the database and synthesizing the target radical;
and the seal migration module is connected with the retrieval module and is used for carrying out seal layout on the target calligraphy character.
2. The calligraphy work style migration system of claim 1 wherein said database comprises:
a calligraphy style database for storing calligraphy words of different calligraphy styles;
a radical database for storing radicals of calligraphic words of different calligraphic styles;
a stroke database for storing strokes of radicals of different calligraphy styles.
3. A method for performing a style migration of a calligraphy work using the style migration system of any one of claims 1-2, comprising the steps of:
s1, the retrieval module sequentially retrieves the middle calligraphy characters corresponding to the original calligraphy characters in the original calligraphy works, and selects the middle calligraphy character closest to the target calligraphy style as the target calligraphy character to migrate the original calligraphy character;
if the retrieval module does not retrieve the middle calligraphy character corresponding to the original calligraphy character or the style of the middle calligraphy character does not meet the target calligraphy style requirement, executing the step S2;
s2, the single character synthesis module searches the middle radicals corresponding to the original radicals of the original calligraphy characters and synthesizes the middle calligraphy characters; if the style of the middle calligraphy character meets the style requirement of the target calligraphy, taking the middle calligraphy character as the target calligraphy character to migrate the original calligraphy character;
if the single character synthesis module does not retrieve all radicals forming the original calligraphy character or the style of the synthesized intermediate calligraphy character does not accord with the style requirement of the target calligraphy, executing step S3;
s3, the radical synthesizing module searches the middle strokes corresponding to the original strokes of the original radicals, synthesizes all the middle radicals forming the original calligraphy characters and then synthesizes the middle calligraphy characters again; if the style of the middle calligraphy character meets the style requirement of the target calligraphy, taking the middle calligraphy character as the target calligraphy character to migrate the original calligraphy character;
if the style of the synthesized intermediate calligraphy character does not meet the style requirement of the target calligraphy, the step S3 is repeatedly executed;
and S4, generating a target calligraphy style work after all the original calligraphy characters in the original calligraphy work are migrated, and performing page layout on the generated target calligraphy style work by the chapter migration module.
4. The method for migrating calligraphic work styles according to claim 3 wherein in step S1, the search module selects the intermediate calligraphy word closest to the target calligraphic style from the intermediate calligraphy words through style similarity calculation as the target calligraphy word;
the style similarity calculation formula is as follows:
SIMC=a1Z(z′i,z′j)+a2T(t′i,t′j)+a3S(s′i,s′j)+a4G(wi,wj)
in the formula, alpha1+α2+α3+α4=1,
Z(z′i,z′j),K(t′i,t′j),S(s′i,s′j) Respectively representing whether the middle calligraphy character and the calligraphy character with the target calligraphy style have similarity in the chapter, knot and stroke characteristics, G (w)i,wj) Indicating whether two words belong to the same work, alpha1、α2、α3、α4Are all constant parameters,s′i,z′i,t′i,wiChapter feature vector, knot feature vector, stroke feature vector and belonging work, s'j,z′j,t′j,wjAnd respectively representing a chapter characteristic vector, a knot characteristic vector, a stroke characteristic vector and the work of the calligraphy character with the target calligraphy style.
5. The method for migrating calligraphic works style according to claim 3 wherein in step S2, the individual character retrieval module uses the target calligraphy character retrieved in step S1 as the target calligraphy style reference character and combines the matching degree calculation to retrieve the middle radical corresponding to the original radical of the original calligraphy character and meeting the target calligraphy style requirement from the radical database;
the matching degree calculation formula is as follows:
MR=λ1RC(ct,c)+λ2RP(pt,p)+λ3RW(rwt,Wf)+λ4RV(vt,v)+λ5RG(wt,wf)
in the formula (I), the compound is shown in the specification,
λ1+λ2+λ3+λ4+λ5=1,
RC(ct,c)=|ct-c|,
RW(rwt,Wf)=|rwt-Wf|,
RV(vt,v)=|vt-v|,
Wf,wfaverage stroke width and self of reference characters for target calligraphy styleWork, RC (c)tC) indicates whether the stroke complexity of the retrieved radical is the same as that of the radical of the target calligraphy style reference word, RP (p)tP) denotes whether the retrieved radical is in the same position as the radical of the target calligraphy style reference word, RW (RW)t,Wf) RV (v) indicating whether the stroke width of the retrieved radical is the same as the stroke width of the radical of the target calligraphy style reference wordtV) indicates whether the retrieved radical has the same aspect ratio, RG (w), as the radical of the target calligraphy style reference wordt,wf) Denotes whether the target radical and the target reference word are from the same work, lambda1、λ2、λ3、λ4、λ5Are all constant parameters, ct,pt,rwt,vt,wtAnd c, p, rw and v respectively represent the radical stroke complexity, the radical position, the radical stroke width and the radical aspect ratio of the target calligraphy style reference character.
6. The method for migrating styles of calligraphy works according to claim 3 wherein the method for synthesizing intermediate calligraphic words from intermediate radicals comprises the steps of:
the middle radicals are spatially arranged according to the positions of the original radicals of the original calligraphy characters in the calligraphy characters;
setting an initial distance between the middle radicals; so as to avoid the adhesion between the middle radicals;
adjusting the relative position of each middle radical in the middle calligraphy characters; to ensure that the center of gravity of the middle calligraphy character tends to the average center of gravity of the target calligraphy style;
adjusting the size of the synthesized intermediate calligraphy character; ensuring that the height, width, height-width ratio of the synthesized middle calligraphy character are consistent with the style of the target calligraphy;
when the relative position of each middle radical is adjusted, the following constraint conditions are met:
|G′s.x-Gs.x|≤a1
|G′s.y-Gs.y|≤a2
in formula (II), G'x,G′yRespectively showing the horizontal and vertical coordinates of the center of gravity of the middle calligraphy character after the first adjustment position of the middle part,abscissa and ordinate, G, respectively representing the mean center of gravity of all calligraphy characters in the work in which the target calligraphy style reference character is locateds.x,Gs.yRespectively represents the horizontal and vertical coordinates G 'of the initial barycenter of the middle radical of the synthesized middle calligraphy character's.x,G′s.yRespectively, the horizontal and vertical coordinates a of the center of gravity after the first adjustment of the middle part1,a2Is a constant, σf.x,σf.yRespectively representing the standard deviation, delta, of the horizontal and vertical coordinates of the center of gravityx,δyAre all threshold constant parameters.
7. The method for migrating calligraphic work style of claim 3 wherein in step S3, the radical synthesis module retrieves from the stroke database the intermediate strokes corresponding to the original strokes of the original radical and meeting the requirements of the target calligraphic style by metric calculation in combination with the target calligraphic style reference word;
the formula of the metric calculation is:
in the formula (I), the compound is shown in the specification,
ω1+ω2+ω3+ω4+ω5=1,
B(b,b')=|b-b'|,
Q(q,q')=|q-q'|,
SV(sv,sv')=|sv-sv'|,
SP(sp,sp')=√(spx-sp′x)2+(spy-sp′y)2,
B(b,b'),Q(q,q'),SV(sv,sv'),SP(sp,sp'),whether the strokes of the middle stroke and the target calligraphy style reference character have similarity in stroke head-tail width, stroke wiggling degree, stroke height-width ratio, stroke position and stroke width, omega1、ω2、ω3、ω4、ω5B, q, sv and sp respectively represent the stroke head width, the stroke twisting degree, the stroke height-width ratio and the stroke position of the stroke of the target calligraphy style reference character, b ', q', sv ', sp' and sw respectively represent the stroke head width, the stroke twisting degree, the stroke height-width ratio, the stroke position and the stroke width of the middle stroke,and the average stroke width of all the calligraphy characters in the calligraphy works where the target calligraphy style reference characters are located is shown.
8. The method for migrating styles of calligraphy works according to claim 3, wherein in step S3, when synthesizing the middle radicals of original calligraphy characters from middle strokes, the middle strokes are synthesized according to the positions in the standard body radicals by using the standard body similar to the target calligraphy style as a reference.
9. The method for migrating calligraphic work styles according to claim 3, wherein in step S4, when the calligraphic migration module performs page layout on the target calligraphic style work, the column-row spacing of the calligraphic work similar to the target calligraphic style is analyzed, so as to adjust the column-row spacing of the target calligraphic style work according to the column-row spacing;
when the chapter migration module adjusts the row-column distance of the target calligraphy style works, the following constraint conditions are met:
|Gnx-Href|≤ε
wherein ε is a threshold constant, HrefIs the center of gravity abscissa, G, of the first word of each line of the target calligraphy style worknxThe barycentric coordinate after adjusting for the rest characters of the row;
|D′nrow-D′row|≤σr
|D′ncol-D′col|≤σc
D′nrow,D′ncolrespectively the adjusted line and column distances, D'row,D′colRespectively the row-column spacing, σ, of the reference calligraphic workrAnd σcRespectively, the standard deviation of the target calligraphy style row-column distance.
10. Calligraphy work style migration terminal, its characterized in that contains: a processor, a memory, and a computer program stored on the memory and executable on the processor, the processor executing the computer program to implement the calligraphic work style migration method of claims 3-9 above.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110127483.1A CN112818634B (en) | 2021-01-29 | 2021-01-29 | Handwriting style migration system, method and terminal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110127483.1A CN112818634B (en) | 2021-01-29 | 2021-01-29 | Handwriting style migration system, method and terminal |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112818634A true CN112818634A (en) | 2021-05-18 |
CN112818634B CN112818634B (en) | 2024-04-05 |
Family
ID=75860277
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110127483.1A Active CN112818634B (en) | 2021-01-29 | 2021-01-29 | Handwriting style migration system, method and terminal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112818634B (en) |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001296854A (en) * | 2000-04-13 | 2001-10-26 | Fujitsu Ltd | Character generation device and recording medium |
CN101236577A (en) * | 2008-02-29 | 2008-08-06 | 浙江大学 | Computer aided calligraphy tablet design method |
CN101599180A (en) * | 2009-03-05 | 2009-12-09 | 浙江大学 | Computer calligraphy based on the handwriting style imitation generates method automatically |
CN102637078A (en) * | 2012-02-27 | 2012-08-15 | 厦门大学 | Method for generating structurally optimized Chinese character patterns |
CN103136769A (en) * | 2011-12-02 | 2013-06-05 | 北京三星通信技术研究有限公司 | Method and device of generation of writing style font of user |
CN106611172A (en) * | 2015-10-23 | 2017-05-03 | 北京大学 | Style learning-based Chinese character synthesis method |
CN109408776A (en) * | 2018-10-09 | 2019-03-01 | 西华大学 | A kind of calligraphy font automatic generating calculation based on production confrontation network |
CN110570481A (en) * | 2019-07-31 | 2019-12-13 | 中国地质大学(武汉) | calligraphy word stock automatic repairing method and system based on style migration |
CN110570346A (en) * | 2019-08-19 | 2019-12-13 | 西安理工大学 | Method for performing style migration on calligraphy based on cyclic generation countermeasure network |
CN111062290A (en) * | 2019-12-10 | 2020-04-24 | 西北大学 | Method and device for constructing Chinese calligraphy style conversion model based on generation confrontation network |
CN111475667A (en) * | 2020-04-08 | 2020-07-31 | 西北农林科技大学 | Mechanical arm regular script style calligraphy writing method based on handwritten Chinese characters |
CN112183027A (en) * | 2020-08-31 | 2021-01-05 | 同济大学 | Artificial intelligence based artwork generation system and method |
-
2021
- 2021-01-29 CN CN202110127483.1A patent/CN112818634B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001296854A (en) * | 2000-04-13 | 2001-10-26 | Fujitsu Ltd | Character generation device and recording medium |
CN101236577A (en) * | 2008-02-29 | 2008-08-06 | 浙江大学 | Computer aided calligraphy tablet design method |
CN101599180A (en) * | 2009-03-05 | 2009-12-09 | 浙江大学 | Computer calligraphy based on the handwriting style imitation generates method automatically |
CN103136769A (en) * | 2011-12-02 | 2013-06-05 | 北京三星通信技术研究有限公司 | Method and device of generation of writing style font of user |
CN102637078A (en) * | 2012-02-27 | 2012-08-15 | 厦门大学 | Method for generating structurally optimized Chinese character patterns |
CN106611172A (en) * | 2015-10-23 | 2017-05-03 | 北京大学 | Style learning-based Chinese character synthesis method |
CN109408776A (en) * | 2018-10-09 | 2019-03-01 | 西华大学 | A kind of calligraphy font automatic generating calculation based on production confrontation network |
CN110570481A (en) * | 2019-07-31 | 2019-12-13 | 中国地质大学(武汉) | calligraphy word stock automatic repairing method and system based on style migration |
CN110570346A (en) * | 2019-08-19 | 2019-12-13 | 西安理工大学 | Method for performing style migration on calligraphy based on cyclic generation countermeasure network |
CN111062290A (en) * | 2019-12-10 | 2020-04-24 | 西北大学 | Method and device for constructing Chinese calligraphy style conversion model based on generation confrontation network |
CN111475667A (en) * | 2020-04-08 | 2020-07-31 | 西北农林科技大学 | Mechanical arm regular script style calligraphy writing method based on handwritten Chinese characters |
CN112183027A (en) * | 2020-08-31 | 2021-01-05 | 同济大学 | Artificial intelligence based artwork generation system and method |
Non-Patent Citations (6)
Title |
---|
JIANYONG CAO ET AL: "A Method of Calligraphy Style Transfer", SCIENTIFIC JOURNAL OF TECHNOLOGY, vol. 3, no. 3, 31 March 2021 (2021-03-31), pages 103 - 113 * |
JIANYONG CAO: "Hierarchical‐based Calligraphy Style Transfer", WORLD SCIENTIFIC RESEARCH JOURNAL, vol. 7, no. 5, 31 May 2021 (2021-05-31), pages 430 - 439 * |
MENGXI QIN ET AL: "Restore the Incomplete Calligraphy Based on Style Transfer", 2019 CHINESE CONTROL CONFERENCE, 11 October 2019 (2019-10-11), pages 8812 - 8817 * |
SONGHUA XU ET AL: "Automatic generation of artistic chinese calligraphy", IEEE INTELLIGENT SYSTEMS, vol. 20, no. 3, 13 June 2005 (2005-06-13), pages 32 - 39, XP011134771 * |
俞凯: "计算机书法若干关键技术研究", 中国博士学位论文全文数据库信息科技辑, no. 8, 15 August 2011 (2011-08-15), pages 138 - 71 * |
张钦禄: "基于生成对抗神经网络的字体风格迁移算法", 中国优秀硕士学位论文全文数据库信息科技辑, no. 6, 15 June 2020 (2020-06-15), pages 138 - 706 * |
Also Published As
Publication number | Publication date |
---|---|
CN112818634B (en) | 2024-04-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yu et al. | Toward realistic face photo–sketch synthesis via composition-aided GANs | |
US10127199B2 (en) | Automatic measure of visual similarity between fonts | |
KR102078473B1 (en) | Moving Body Tracking Device, Moving Body Tracking Method, and Moving Body Tracking Program | |
CN102113012B (en) | Finger shape estimating device, the presumption method of finger shape and program | |
CN110738101A (en) | Behavior recognition method and device and computer readable storage medium | |
KR20200070409A (en) | Human hairstyle creation method based on multiple feature search and transformation | |
CN104200240B (en) | A kind of Sketch Searching method based on content-adaptive Hash coding | |
CN107924452B (en) | Combined shape regression for face alignment in images | |
CN111931710B (en) | Online handwritten character recognition method and device, electronic equipment and storage medium | |
CN110678878B (en) | Apparent feature description attribute identification method and device | |
KR101794399B1 (en) | Method and system for complex and multiplex emotion recognition of user face | |
JP2020177615A (en) | Method of generating 3d facial model for avatar and related device | |
CN111598087A (en) | Irregular character recognition method and device, computer equipment and storage medium | |
CN112200147A (en) | Face recognition method and device, computer equipment and storage medium | |
CN110378329B (en) | Intelligent auxiliary calligraphy practicing system based on dot matrix pen | |
JP2016045884A (en) | Pattern recognition device and pattern recognition method | |
CN109710924B (en) | Text model training method, text recognition method, device, equipment and medium | |
JP4570995B2 (en) | MATCHING METHOD, MATCHING DEVICE, AND PROGRAM | |
CN112818634A (en) | Calligraphy work style migration system, method and terminal | |
CN113469091A (en) | Face recognition method, training method, electronic device and storage medium | |
CN109800702B (en) | Quick comparison method for finger vein identification and computer readable storage medium | |
CN111915676A (en) | Image generation method and device, computer equipment and storage medium | |
KR20210041856A (en) | Method and apparatus for generating learning data required to learn animation characters based on deep learning | |
US20230306054A1 (en) | Image selection apparatus, image selection method, and non-transitory computer-readable medium | |
US20230245342A1 (en) | Image selection apparatus, image selection method, and non-transitory computer-readable medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |