CN112861520A - Chinese character structure optimization method and system based on computer - Google Patents
Chinese character structure optimization method and system based on computer Download PDFInfo
- Publication number
- CN112861520A CN112861520A CN202110250992.3A CN202110250992A CN112861520A CN 112861520 A CN112861520 A CN 112861520A CN 202110250992 A CN202110250992 A CN 202110250992A CN 112861520 A CN112861520 A CN 112861520A
- Authority
- CN
- China
- Prior art keywords
- word
- character
- sample
- library
- words
- 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 83
- 238000005457 optimization Methods 0.000 title claims abstract description 29
- 230000008569 process Effects 0.000 claims abstract description 18
- 230000006870 function Effects 0.000 claims description 33
- 230000001915 proofreading effect Effects 0.000 claims description 9
- 238000007670 refining Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 230000000670 limiting effect Effects 0.000 abstract description 6
- 238000010801 machine learning Methods 0.000 abstract description 6
- 238000010606 normalization Methods 0.000 description 7
- 238000013461 design Methods 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 5
- 238000013508 migration Methods 0.000 description 4
- 230000005012 migration Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/268—Morphological analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/24—Character recognition characterised by the processing or recognition method
- G06V30/242—Division of the character sequences into groups prior to recognition; Selection of dictionaries
- G06V30/244—Division of the character sequences into groups prior to recognition; Selection of dictionaries using graphical properties, e.g. alphabet type or font
- G06V30/245—Font recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Controls And Circuits For Display Device (AREA)
Abstract
The invention provides a Chinese character structure optimization method and a system based on a computer, wherein the method comprises the following steps: acquiring sample characters in a preset sample library; thinning the sample character to obtain a thinned character of the sample character; thinning all words in a preset word stock according to the sample words and the thinned words thereof to obtain the thinned words of the whole word stock; and performing stroke filling on the refined characters in the whole character library to obtain the commercial character library. The words generated by the method through the steps of thinning, combining, limiting and filling have good normative, meet the normative standard of the commercial words, and the amplified words are local and more accurate and can be used as a commercial word stock. The method converts a large part of the optimization process into parameter limitation in machine learning, greatly reduces the workload of manual optimization and font designers, and solves the problem that a large amount of labor and time are consumed in the existing method.
Description
Technical Field
The invention belongs to the technical field of word stock design, and particularly relates to a Chinese character structure optimization method and system based on a computer.
Background
In the traditional method for designing the word stock, a font designer designs all Chinese characters, and the Chinese characters become fonts which can be used on electronic equipment through image processing and other modes. With the progress of the artificial intelligence era, the artificial intelligence method also gradually enters the field of word stock design.
For example, existing methods for applying artificial intelligence to word stock design include: the user writes a certain number of Chinese characters, and the character style of the user characteristic is generated according to the writing style of the user. The design idea is based on the idea of 'style migration', namely the relation between the characters written by the user and the corresponding characters in the standard fonts (such as Song style and black body) is found in a machine learning mode, and the relation is used for other characters, so that the fonts with the personal styles of the user can be generated.
However, the method of style migration has many limitations, and the most important problem is that the generated font has no "normative" statement, and although style inheritance is obvious, the font is far from serving as a standard and commercially available word stock. Because of this, the method of "style migration" is often applied to generating handwritten fonts, rather than commercial fonts with sufficient normalization.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a Chinese character structure optimization method and a Chinese character structure optimization system based on a computer, which improve the standardization of generated characters and meet the standard of commercial characters.
In a first aspect, a computer-based Chinese character structure optimization method includes the following steps:
acquiring sample characters in a preset sample library;
thinning the sample character to obtain a thinned character of the sample character;
thinning all words in a preset word stock according to the sample words and the thinned words thereof to obtain the thinned words of the whole word stock;
and performing stroke filling on the refined characters in the whole character library to obtain the commercial character library.
Preferably, the typeface of the sample library is designed by a font designer; the sample character is in a picture format.
Preferably, the refining the sample word to obtain a refined word of the sample word specifically includes:
and thinning the sample character, and extracting a single-pixel framework of the sample character to obtain the thinned character of the sample character.
Preferably, after obtaining the refined words of the whole word stock, the method further comprises:
when isolated pixel points appear in the obtained refined character or gaps exist among the pixel points, deleting the isolated pixel points by using a topology invariant correction method or connecting the gaps;
when the obtained refined character part meets the expectation, setting a first standard function corresponding to the target function in the refining process to enable the obtained refined character to meet the expectation;
when the obtained refined word is completely not expected or can not be identified, adding the word corresponding to the refined word into the sample library to be used as the sample word of the sample library.
Preferably, the thinned word part is expected to comprise a single point after the stroke of the 'point' is thinned, and the trend is not; or there is a large radian connection in the refined word.
Preferably, the stroke filling of the refined words in the whole word stock specifically includes:
setting a second standard function, and filling strokes of the refined words in the whole word stock by using the second standard function and the target function;
the second normative function is used for setting a thickness coefficient, edge smoothness and layout style.
Preferably, the thickness coefficient is proportional to the number of pixels of the generated word and inversely proportional to the number of pixels of the refined word, and is influenced by the number of crossed strokes and the number of connected graph branches in the refined word.
Preferably, the stroke filling of the refined word with edge smoothness step comprises:
traversing a preset standard library, and taking out a pixel map in the neighborhood of a pixel point of each stroke or each element positioned at the edge in the standard library according to a preset domain range to obtain a proofreading library;
and taking out a neighborhood pixel map of the pixel points positioned at the edge in the generated word obtained after the stroke filling according to the domain range, if the neighborhood pixel map is not in the proofreading library, judging that the generated word is not smooth locally, and carrying out manual detection.
Preferably, after determining that the generated word is not locally smooth, the method further comprises:
adding a neighborhood inner pixel map of the generated word with local non-smoothness into the proofreading library;
alternatively, a map of intra-neighborhood pixels located in the calibration library is added to the second canonical function.
In a second aspect, a computer-based chinese character structure optimization system includes a processor, an input device, an output device, and a memory, where the processor, the input device, the output device, and the memory are connected to each other, where the memory is used to store a computer program, and the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method of the first aspect.
According to the technical scheme, the Chinese character structure optimization method and system based on the computer have the advantages that the words generated through the steps of thinning, combining, limiting and filling have good normalization, the normalization standard of commercial words is met, the local parts of the generated words are more accurate after the words are amplified, and the words can be used as a commercial word stock. The method converts a large part of the optimization process into parameter limitation in machine learning, greatly reduces the workload of manual optimization and font designers, and solves the problem that a large amount of labor and time are consumed in the existing method. The method and the system have good expandability, are matched with the establishment of the database, and can be applied to solving a plurality of other similar problems.
Drawings
In order to more clearly illustrate the detailed description of the invention or the technical solutions in the prior art, the drawings that are needed in the detailed description of the invention or the prior art will be briefly described below. Throughout the drawings, like elements or portions are generally identified by like reference numerals. In the drawings, elements or portions are not necessarily drawn to scale.
Fig. 1 is a schematic flow chart of a method for optimizing a chinese character structure according to an embodiment of the present method.
Fig. 2 is a schematic diagram of a "flipped" refinement word provided in the first embodiment of the present method.
Fig. 3 is a specific flowchart of a method for optimizing a chinese character structure according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a part of the refinement word that is not in accordance with the expectation provided by the first embodiment of the method.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and therefore are only examples, and the protection scope of the present invention is not limited thereby. It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which the invention pertains.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
As used in this specification and the appended claims, the term "if" may be interpreted contextually as "when", "upon" or "in response to a determination" or "in response to a detection". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
The first embodiment is as follows:
a method for optimizing a chinese character structure based on a computer, referring to fig. 1, comprising the steps of:
acquiring sample characters in a preset sample library;
thinning the sample character to obtain a thinned character of the sample character;
thinning all words in a preset word stock according to the sample words and the thinned words thereof to obtain the thinned words of the whole word stock;
and performing stroke filling on the refined characters in the whole character library to obtain the commercial character library.
Specifically, the method adopts an artificial intelligence technology, and provides a method for generating a standard and commercially available commercial word stock based on the sample words provided by a font designer. The number of the sample words can be set according to specific requirements, for example, 50 sample words are preset in the sample library.
The method is mainly optimized according to the thinned character, and the thinned character can embody a single-pixel framework of a font structure, and is shown in figure 2. Because the existing style migration method can greatly change the skeleton of the font, in order to improve the normalization of the generated character, the method is optimized according to the refined character, and can prevent the skeleton of the generated character from being greatly changed.
The method performs a 'fleshing' process from a refined word to a complete word (i.e., generating a word) after the refined word of the entire word library is generated through artificial intelligence. In the generation process of the refined words and the 'fleshing' process, a series of normative functions can be added, so that the normative of the generated words cannot be damaged in each step of the process.
The words generated by the method through the steps of thinning, combining, limiting and filling have good normative, meet the normative standard of the commercial words, and the amplified words are local and more accurate and can be used as a commercial word stock. The method converts a large part of the optimization process into parameter limitation in machine learning, greatly reduces the workload of manual optimization and font designers, and solves the problem that a large amount of labor and time are consumed in the existing method. The method has good expandability, is matched with the establishment of the database, and can be applied to solving a plurality of other similar problems.
Preferably, the typeface of the sample library is designed by a font designer; the sample character is in a picture format.
Referring to fig. 3, the refining the sample word to obtain a refined word of the sample word specifically includes:
and thinning the sample character, and extracting a single-pixel framework of the sample character to obtain the thinned character of the sample character.
Specifically, the font designer needs to maintain good normativity when designing the typeface. The method is not suitable for the handwriting font generation scene with poor normalization. The refined words are skeletons extracted from sample words in a picture format.
In addition, the refined words of the whole word stock obtained by the method have the following problems:
1. the generated refined word is completely unexpected or cannot be recognized, and it is impossible to directly recognize what word is generated.
2. The resulting refined word parts are not as expected, for example, local structures that should not occur in chinese characters occur, or confusion with similar characters occurs, or a combination problem occurs with the skeleton structure, see fig. 4. The "peak" word generated normally is shown on the left side of FIG. 4, and the "peak" word with the combination of the skeletal structures is shown on the right side.
3. The generated refined word basically accords with expectation, but has detail problems, such as isolated pixel points which do not influence the whole word structure, or the condition of vacancy in the middle of the pixel points and the like.
In view of the above three problems, the method proposes three solutions.
1. In the first case, the word corresponding to the refined word is added to the sample library as a sample word of the sample library, that is, a word that a specified font designer needs to design. Because there are cases where some of the composite structures that may appear in the chinese characters are missed when designing the sample library. Therefore, the method is actually used for detecting defects and repairing leaks. When the situation that the character can not be identified occurs, the method adopts a mode of artificial judgment, namely, the artificial judgment is adopted to judge whether the character needs to be added as a sample character.
2. And aiming at the second situation, setting a first standard function corresponding to the target function in the thinning process to enable the obtained thinned characters to be expected, namely setting the first standard function matched with the target function, so that the final generated word can eliminate the local structure which is not expected. For example, a single point is likely to appear in a refined word of a "point" stroke, but actually the "point" stroke is directional, but the situation that the recognition is unknown occurs because the pixel distribution near the gravity center is too uniform, a first canonical function should be added, the number of pixel points of each connected branch in the refined word should be set to be greater than or equal to k, and k is a specific value set by a user or obtained according to experiments. For example, when a large radian connection occurs in a refined word, for most fonts, the stroke connection position should have a very obvious angle, but in the process of refining the word or generating the refined word, the radian connection may occur at the stroke connection position, and at this time, a first canonical function is added to indicate that the pixel points at the stroke intersection should be close to linear distribution.
3. And for the third situation, deleting isolated pixel points or connecting vacant places by using a topology invariant correction method. For example, for the refined word 'good', the refined word should have two connected branches, if the number of the connected branches for generating the word is not 2, the number of the connected branches is added into the requirement of the first canonical function, so that the conditions of stroke adhesion and the like can not occur when the 'good' is generated next time.
The order of implementation of the above three solutions should be 3 → 2 → 1, and when neither of the 3 rd nor 2 nd methods can solve the problem of the generated refined word, the manual processing is performed and added into the sample word stock. In addition, in the implementation process of the methods 3 and 2, if an irregular legend possibly appearing in the refined word is found, a corresponding normative function can be set for automatic processing.
Preferably, the stroke filling of the refined words in the whole word stock specifically includes:
setting a second standard function, and filling strokes of the refined words in the whole word stock by using the second standard function and the target function;
the second normative function is used for setting a thickness coefficient, edge smoothness and layout style.
Specifically, the complete character obtained by stroke filling in the method is in a picture format, the refined character becomes the complete character after the 'fleshiness' process, and a commercial character library is formed by all the complete characters. As the 'fleshing' process adopts a pure machine learning method, all the optimization processes need to set a second standard function so as to achieve the common limiting effect on the generated words by the second standard function and the target function. The second canonical function is set by the following dimensions:
1. and (4) thickness limitation. In order to make the thickness degree of the words in the commercial word stock similar, a thickness coefficient needs to be set, and the thickness coefficient is proportional to the pixel number of the generated words and inversely proportional to the pixel number of the refined words and is influenced by the stroke intersection number and the graph connection branch number. The method can set different thickness coefficient ranges aiming at different sample banks, namely, the minimum value and the maximum value of the thickness coefficient of the whole body sample are calculated, and the thickness coefficient is limited not to exceed the range in the 'fleshing' process. In the whole word generation process, the 'fleshiness' can be completed without limitation, then the thickness coefficient of the generated word is calculated, the next link is started when the thickness coefficient meets the condition, and the thickness limit is added to the objective function of the word when the thickness coefficient does not meet the condition.
2. The edge smoothness limits. The smoothness of the edge can be seen as the degree of local thickness variation, which is global and not local, so the method requires finding data describing the degree of local boundary variation. The method comprises the following specific steps: and traversing the standard stroke library and the primitive library, and taking out pixel maps of all the pixel points positioned at the edge in 5-by-5 neighborhoods to serve as a proofreading library. When the complete word is generated, whether a pixel map of pixel points positioned at the edge in a 5 × 5 neighborhood exists in a proofreading library or not is checked, if yes, the local part of the word is judged not to be smooth, and at the moment, manual detection is carried out. After the edge smoothness judgment is completed, adding a neighborhood inner pixel map of the generated word with local non-smoothness into the proofreading library; alternatively, a map of intra-neighborhood pixels located in the calibration library is added to the second canonical function. Thus, after a certain number of manual tests, the collation library can be determined.
3. The unification of the font and the layout style is realized. In order to make the generated word beautiful in real use, the margin of the generated word needs to be left to a proper degree. To achieve this goal, the method combines two factors, the font center of gravity and the maximum square containing the word, to achieve word centering and resizing for a particular font, so that the specification and the appearance are maintained even after a plurality of words are combined.
Example two:
a Chinese character structure optimization system based on a computer comprises a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are connected with each other, the memory is used for storing a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the method.
The word generated by the system through the steps of thinning, combining, limiting and filling has good normalization, accords with the normalization standard of the commercial word, is locally more accurate after being amplified, and can be used as a commercial word stock. The system converts a large part of the optimization process into parameter limitation in machine learning, greatly reduces the workload of manual optimization and font designers, and solves the problem that a large amount of labor and time are consumed in the existing method. The system has good expandability, is matched with the establishment of a database, and can be applied to solving a plurality of other similar problems.
It should be understood that in the embodiments of the present invention, the Processor may be a Central Processing Unit (CPU), and the Processor may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The input device may include a touch pad, a fingerprint sensor (for collecting fingerprint information of a user and direction information of the fingerprint), a microphone, etc., and the output device may include a display (LCD, etc.), a speaker, etc.
The memory may include both read-only memory and random access memory, and provides instructions and data to the processor. The portion of memory may also include non-volatile random access memory. For example, the memory may also store device type information.
For the sake of brief description, the system provided by the embodiment of the present invention may refer to the corresponding content in the foregoing embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the present invention, and they should be construed as being included in the following claims and description.
Claims (10)
1. A Chinese character structure optimization method based on a computer is characterized by comprising the following steps:
acquiring sample characters in a preset sample library;
thinning the sample character to obtain a thinned character of the sample character;
thinning all words in a preset word stock according to the sample words and the thinned words thereof to obtain the thinned words of the whole word stock;
and performing stroke filling on the refined characters in the whole character library to obtain the commercial character library.
2. The computer-based Chinese character configuration optimization method of claim 1,
the typeface of the sample library is designed by a font designer; the sample character is in a picture format.
3. The computer-based Chinese character configuration optimization method of claim 1, wherein the step of refining the typeface to obtain the refined typeface specifically comprises:
and thinning the sample character, and extracting a single-pixel framework of the sample character to obtain the thinned character of the sample character.
4. The computer-based Chinese character configuration optimization method of claim 2, further comprising, after said obtaining refined words of the entire word library:
when isolated pixel points appear in the obtained refined character or gaps exist among the pixel points, deleting the isolated pixel points by using a topology invariant correction method or connecting the gaps;
when the obtained refined character part meets the expectation, setting a first standard function corresponding to the target function in the refining process to enable the obtained refined character to meet the expectation;
when the obtained refined word is completely not expected or can not be identified, adding the word corresponding to the refined word into the sample library to be used as the sample word of the sample library.
5. The computer-based Chinese character configuration optimization method of claim 4,
the thinned word part is expected to comprise a single point after the 'point' stroke is thinned, and the trend is not existed; or there is a large radian connection in the refined word.
6. The computer-based Chinese character structure optimization method of claim 4, wherein the stroke filling of the refined words of the entire word stock specifically comprises:
setting a second standard function, and filling strokes of the refined words in the whole word stock by using the second standard function and the target function;
the second normative function is used for setting a thickness coefficient, edge smoothness and layout style.
7. The computer-based Chinese character configuration optimization method of claim 6,
the thickness coefficient is in direct proportion to the number of pixels of the generated word and in inverse proportion to the number of pixels of the thinned word, and is influenced by the number of crossed strokes and the number of graph connected branches in the thinned word.
8. The computer-based Chinese character configuration optimization method of claim 6,
the step of stroke filling the refined word with edge smoothness includes:
traversing a preset standard library, and taking out a pixel map in the neighborhood of a pixel point of each stroke or each element positioned at the edge in the standard library according to a preset domain range to obtain a proofreading library;
and taking out a neighborhood pixel map of the pixel points positioned at the edge in the generated word obtained after the stroke filling according to the domain range, if the neighborhood pixel map is not in the proofreading library, judging that the generated word is not smooth locally, and carrying out manual detection.
9. The computer-based Chinese character configuration optimization method of claim 8, further comprising, after determining that the generated word is not locally smooth:
adding a neighborhood inner pixel map of the generated word with local non-smoothness into the proofreading library;
alternatively, a map of intra-neighborhood pixels located in the calibration library is added to the second canonical function.
10. A Chinese character structure optimization system based on a computer is characterized in that,
comprising a processor, an input device, an output device and a memory, said processor, input device, output device and memory being interconnected, wherein said memory is adapted to store a computer program comprising program instructions, said processor being configured to invoke said program instructions to perform the method according to any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110250992.3A CN112861520A (en) | 2021-03-08 | 2021-03-08 | Chinese character structure optimization method and system based on computer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110250992.3A CN112861520A (en) | 2021-03-08 | 2021-03-08 | Chinese character structure optimization method and system based on computer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112861520A true CN112861520A (en) | 2021-05-28 |
Family
ID=75994611
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110250992.3A Pending CN112861520A (en) | 2021-03-08 | 2021-03-08 | Chinese character structure optimization method and system based on computer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112861520A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104156721A (en) * | 2014-07-31 | 2014-11-19 | 南京师范大学 | Off-line Chinese character stroke extraction method based on template matching |
| CN106384112A (en) * | 2016-09-08 | 2017-02-08 | 西安电子科技大学 | Rapid image text detection method based on multi-channel and multi-dimensional cascade filter |
| CN109635883A (en) * | 2018-11-19 | 2019-04-16 | 北京大学 | The Chinese word library generation method of the structural information guidance of network is stacked based on depth |
| CN109753953A (en) * | 2017-11-03 | 2019-05-14 | 腾讯科技（深圳）有限公司 | The method, apparatus of localization of text, electronic equipment and storage medium in image |
| CN111507064A (en) * | 2020-04-15 | 2020-08-07 | 伍曙光 | Word stock design method based on computer |
| CN112434763A (en) * | 2020-11-24 | 2021-03-02 | 伍曙光 | Chinese character skeleton generating method based on computer |
-
2021
- 2021-03-08 CN CN202110250992.3A patent/CN112861520A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104156721A (en) * | 2014-07-31 | 2014-11-19 | 南京师范大学 | Off-line Chinese character stroke extraction method based on template matching |
| CN106384112A (en) * | 2016-09-08 | 2017-02-08 | 西安电子科技大学 | Rapid image text detection method based on multi-channel and multi-dimensional cascade filter |
| CN109753953A (en) * | 2017-11-03 | 2019-05-14 | 腾讯科技（深圳）有限公司 | The method, apparatus of localization of text, electronic equipment and storage medium in image |
| CN109635883A (en) * | 2018-11-19 | 2019-04-16 | 北京大学 | The Chinese word library generation method of the structural information guidance of network is stacked based on depth |
| CN111507064A (en) * | 2020-04-15 | 2020-08-07 | 伍曙光 | Word stock design method based on computer |
| CN112434763A (en) * | 2020-11-24 | 2021-03-02 | 伍曙光 | Chinese character skeleton generating method based on computer |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111814794A (en) | Text detection method and device, electronic equipment and storage medium | |
| CN113486828B (en) | Image processing method, device, equipment and storage medium | |
| CN111368638A (en) | Spreadsheet creation method and device, computer equipment and storage medium | |
| CN112990175B (en) | Method, device, computer equipment and storage medium for recognizing handwritten Chinese characters | |
| EP3846122B1 (en) | Method and apparatus for generating background-free image, device, and medium | |
| CN112926421B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN112001399B (en) | Image scene classification method and device based on local feature saliency | |
| CN111415373A (en) | Target tracking and segmenting method, system and medium based on twin convolutional network | |
| CN114038004A (en) | Certificate information extraction method, device, equipment and storage medium | |
| CN112868021A (en) | Letter detection device, method and system | |
| CN113205047A (en) | Drug name identification method and device, computer equipment and storage medium | |
| CN114972947B (en) | Depth scene text detection method and device based on fuzzy semantic modeling | |
| CN114241432A (en) | License plate recognition method, equipment and storage medium | |
| CN112115884A (en) | Form recognition method and system | |
| CN114723677A (en) | Image defect detection method, image defect detection device, image defect detection equipment and storage medium | |
| CN111160358B (en) | Image binarization method, device, equipment and medium | |
| US12106474B2 (en) | Method of determining a distribution of stem cells in a cell image, electronic device, and storage medium | |
| CN115311666A (en) | Image-text recognition method and device, computer equipment and storage medium | |
| CN111241974A (en) | Bill information acquisition method and device, computer equipment and storage medium | |
| CN114332894A (en) | Image text detection method and device | |
| CN116802683A (en) | Image processing method and system | |
| CN112861520A (en) | Chinese character structure optimization method and system based on computer | |
| CN117593420A (en) | Plane drawing labeling method, device, medium and equipment based on image processing | |
| CN111862159A (en) | Improved target tracking and segmentation method, system and medium for twin convolutional network | |
| CN113033593A (en) | Text detection training method and device based on deep learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210528 |