CN113240059A - Handwritten Chinese character quality evaluation method based on deep learning - Google Patents
Handwritten Chinese character quality evaluation method based on deep learning Download PDFInfo
- Publication number
- CN113240059A CN113240059A CN202110559715.0A CN202110559715A CN113240059A CN 113240059 A CN113240059 A CN 113240059A CN 202110559715 A CN202110559715 A CN 202110559715A CN 113240059 A CN113240059 A CN 113240059A
- Authority
- CN
- China
- Prior art keywords
- chinese characters
- handwritten
- characters
- database
- paper
- 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 37
- 238000013135 deep learning Methods 0.000 title claims abstract description 18
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 12
- 238000002372 labelling Methods 0.000 claims abstract description 5
- 230000006978 adaptation Effects 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
Images
Classifications
-
- 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
-
- 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
- G06F40/106—Display of layout of documents; Previewing
-
- 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
Abstract
The invention discloses a handwritten Chinese character quality evaluation method based on deep learning, which comprises the following steps: s1, placing the paper before the collection assembly, collecting the size of the paper and the number of Chinese characters, judging whether the size of the paper can be collected normally, if so, collecting the characteristics of the handwritten Chinese characters, and if not, adjusting the position of the paper and then collecting the paper; s2, judging whether all Chinese characters can be completely collected, if so, finishing the collection process, and if not, collecting the characteristics of a single Chinese character in a short distance; s3, identifying all the handwritten Chinese characters, labeling the Chinese characters which can not be identified, and providing a plurality of Chinese characters with high similarity to the Chinese characters for selection.
Description
Technical Field
The invention relates to the technical field of Chinese character evaluation, in particular to a handwritten Chinese character quality evaluation method based on deep learning.
Background
The modern Chinese characters refer to regular Chinese character font after regular script, including traditional Chinese characters and simplified Chinese characters. Modern Chinese characters have evolved from oracle, golden, Zhuzhui, Xiaozhui, clerical script, cursive script, regular script, running script, etc. The Chinese characters are created and improved for the precedent invention of Chinese nationality, and are indispensable links for maintaining all dialect areas of Chinese nationality. The earliest Chinese characters in existence are the oracle bone script of Ying Shang and later golden script in 1300 years before about the Gongyuan, which are transformed into Zhuzhujin in the West week, then to the minor seal and clerical script in Qin dynasty, to the great slaver of Hanwei, and to the regular script of Hanwei. Regular script prevails in the south and north of Wei and jin and is passed through so far.
When calligraphy is practiced, the quality of Chinese characters written by the user needs to be evaluated, the Chinese characters can be improved better, calligraphy practice is facilitated, the existing method is large in workload, strong in subjectivity, uncertain in evaluation results and different from person to person, and good practice effect cannot be achieved, so that the handwritten Chinese character quality evaluation method based on deep learning is provided and used for solving the problems.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides a handwritten Chinese character quality evaluation method based on deep learning.
In order to achieve the purpose, the invention adopts the following technical scheme:
the handwritten Chinese character quality evaluation method based on deep learning comprises the following steps:
s1, placing the paper before the collection assembly, collecting the size of the paper and the number of Chinese characters, judging whether the size of the paper can be collected normally, if so, collecting the characteristics of the handwritten Chinese characters, and if not, adjusting the position of the paper and then collecting the paper;
s2, judging whether all Chinese characters can be completely collected, if so, finishing the collection process, and if not, collecting the characteristics of a single Chinese character in a short distance;
s3, recognizing all handwritten Chinese characters, labeling the Chinese characters which cannot be recognized, providing a plurality of Chinese characters with high similarity to the Chinese characters for selection, helping to recognize the Chinese characters through manual selection, recording the Chinese characters and storing the Chinese characters in a database, so that the recognition process can be normally performed next time;
s4, comparing the recognized Chinese characters with the Chinese characters in the database, judging whether the characters to be compared exist, if so, inputting the name of the compared characters, comparing the handwritten Chinese characters with the database, if not, comparing the handwritten Chinese characters with all the characters in the database, finding the character with the highest adaptation degree, judging whether the adapted character can be found, if so, comparing the handwritten Chinese characters with the database, if not, manually inputting the name of the character to be compared, and then comparing;
s5, evaluating the Chinese characters, scoring according to the distribution positions of the Chinese characters on the paper, scoring according to the recognition degrees of the Chinese characters, scoring according to the comparison results of the Chinese characters and the fonts in the database, averaging all the scores, and finishing the scoring process;
and S6, displaying the handwritten Chinese characters and the Chinese characters in the database in the display system at the same time.
Preferably, in step S5, the score is given according to the distribution position of the chinese characters on the paper, and the more reasonable the distribution of the chinese character positions according to the paper size, the higher the score is, and the closer the font size degree is, the higher the score is.
Preferably, in step S5, the score is given according to the degree of recognition of the chinese character, and the higher the degree of recognition of the chinese character is, the higher the score is.
Preferably, in the step S5, the score is given according to the comparison result between the chinese characters and the fonts in the database, and the higher the attaching degree is, the higher the score is.
Preferably, in step S6, the handwritten chinese characters and the chinese characters in the database are displayed in the display system at the same time, wherein the chinese characters in the database are displayed in red at the bottom, and the handwritten chinese characters are displayed in light blue at the top, which are displayed in an overlapping manner, so that the user can compare the handwritten chinese characters and the chinese characters.
Compared with the prior art, the invention has the beneficial effects that:
1. in the invention, the size of the paper and the number of Chinese characters are collected by placing the paper in front of the collection assembly, whether the size of the paper can be normally collected is judged, if so, the characteristics of the handwritten Chinese characters are collected, and if not, the position of the paper is adjusted and then the paper is collected;
2. in the invention, whether all Chinese characters can be completely collected is judged, if yes, the collection process is completed, and if not, the characteristics of a single Chinese character are collected in a short distance;
3. in the invention, by identifying all handwritten Chinese characters, Chinese characters which cannot be identified are labeled, a plurality of Chinese characters with high similarity to the Chinese characters are provided for selection, the Chinese characters are identified through manual selection, and the Chinese characters are recorded and stored in a database, so that the next identification process can be conveniently and normally carried out;
4. in the invention, the recognized Chinese characters are compared with the Chinese characters in the database to judge whether the characters to be compared exist, if yes, the name of the compared characters is input, the handwritten Chinese characters are compared with the database, if not, the handwritten Chinese characters are compared with all the characters of the Chinese characters in the database to find the character with the highest adaptation degree, whether the adapted character can be found is judged, if yes, the handwritten Chinese characters are compared with the database, if not, the name of the character to be compared is manually input, and then the character is compared;
5. in the invention, Chinese characters are evaluated, the Chinese characters are scored according to the distribution positions of the Chinese characters on paper, the Chinese characters are scored according to the identification degrees of the Chinese characters, the Chinese characters are scored according to the comparison result of the Chinese characters and the characters in a database, all scores are averaged, and the scoring process is completed;
6. in the invention, the handwritten Chinese characters and the Chinese characters in the database are displayed in the display system at the same time.
The invention has simple structure, can automatically identify all handwritten Chinese characters by collecting the characteristics of the handwritten Chinese characters and the size of paper, compares the handwritten Chinese characters with the characters which the user wants to compare according to the requirements of the user, can automatically identify and finish the comparison process if the handwritten Chinese characters do not exist, and gives corresponding scores, and is convenient to use.
Drawings
FIG. 1 is a general working flow chart of the method for evaluating the quality of handwritten Chinese characters based on deep learning according to the present invention;
FIG. 2 is a block diagram of the structure of the method for evaluating the quality of handwritten Chinese characters based on deep learning according to the present invention;
FIG. 3 is a Chinese character acquisition flow chart of the handwritten Chinese character quality evaluation method based on deep learning according to the present invention;
FIG. 4 is a flow chart of Chinese character recognition of the method for evaluating the quality of handwritten Chinese characters based on deep learning according to the present invention;
FIG. 5 is a Chinese character comparison flow chart of the handwritten Chinese character quality evaluation method based on deep learning according to the present invention;
FIG. 6 is a Chinese character scoring flow chart of the handwritten Chinese character quality evaluation method based on deep learning according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
Example one
Referring to fig. 1-6, the handwritten Chinese character quality evaluation method based on deep learning comprises the following steps:
s1, placing the paper before the collection assembly, collecting the size of the paper and the number of Chinese characters, judging whether the size of the paper can be collected normally, if so, collecting the characteristics of the handwritten Chinese characters, and if not, adjusting the position of the paper and then collecting the paper;
s2, judging whether all Chinese characters can be completely collected, if so, finishing the collection process, and if not, collecting the characteristics of a single Chinese character in a short distance;
s3, recognizing all handwritten Chinese characters, labeling the Chinese characters which cannot be recognized, providing a plurality of Chinese characters with high similarity to the Chinese characters for selection, helping to recognize the Chinese characters through manual selection, recording the Chinese characters and storing the Chinese characters in a database, so that the recognition process can be normally performed next time;
s4, comparing the recognized Chinese characters with the Chinese characters in the database, judging whether the characters to be compared exist, if so, inputting the name of the compared characters, comparing the handwritten Chinese characters with the database, if not, comparing the handwritten Chinese characters with all the characters in the database, finding the character with the highest adaptation degree, judging whether the adapted character can be found, if so, comparing the handwritten Chinese characters with the database, if not, manually inputting the name of the character to be compared, and then comparing;
s5, evaluating the Chinese characters, scoring according to the distribution positions of the Chinese characters on the paper, scoring according to the recognition degrees of the Chinese characters, scoring according to the comparison results of the Chinese characters and the fonts in the database, averaging all the scores, and finishing the scoring process;
and S6, displaying the handwritten Chinese characters and the Chinese characters in the database in the display system at the same time.
Example two
The embodiment is improved on the basis of the first embodiment: the step S5 is performed according to the distribution position of the chinese characters on the paper, the more reasonable the distribution of the chinese character positions according to the size of the paper, the higher the score, the closer the font size degree and the higher the score, the step S5 is performed according to the recognition degree of the chinese characters, the higher the recognition degree of the chinese characters and the higher the score, the step S5 is performed according to the comparison result of the chinese characters and the fonts in the database, the higher the fit degree and the higher the score, the step S6 simultaneously displays the handwritten chinese characters and the chinese characters in the database in the display system, wherein the chinese characters in the database are red at the bottom, the handwritten chinese characters are light blue at the top, and the two are displayed in an overlapping manner, which is convenient for the user to compare.
The working principle is as follows: placing paper before the acquisition assembly, acquiring the size of the paper and the number of Chinese characters, judging whether the size of the paper can be normally acquired, if so, acquiring the characteristics of the handwritten Chinese characters, and if not, adjusting the position of the paper and then acquiring; judging whether all Chinese characters can be completely collected, if so, finishing the collection process, and if not, collecting the characteristics of a single Chinese character in a short distance; identifying all handwritten Chinese characters, labeling the Chinese characters which cannot be identified, providing a plurality of Chinese characters with high similarity to the Chinese characters for selection, helping to identify the Chinese characters through manual selection, recording the Chinese characters and storing the Chinese characters in a database, and facilitating the normal operation of the next identification process; comparing the recognized Chinese characters with Chinese characters in a database, judging whether characters to be compared exist or not, if so, inputting the names of the characters to be compared, comparing the handwritten Chinese characters with the database, if not, comparing the handwritten Chinese characters with all the characters in the database, finding the character with the highest adaptation degree, judging whether the adapted character can be found or not, if so, comparing the handwritten Chinese characters with the database, if not, manually inputting the name of the character to be compared, and then comparing, for example, a user writes a 'Wang', a device compares the 'Wang' of the handwritten Chinese characters with the 'Wang' of the Song style, so that the Song style can be input automatically, and if not, the system automatically recognizes the character with the highest adaptation degree and compares the character; evaluating the Chinese characters, scoring according to the distribution positions of the Chinese characters on the paper, scoring according to the recognition degrees of the Chinese characters, scoring according to the comparison results of the Chinese characters and the fonts in the database, averaging all the scores, and finishing the scoring process; and simultaneously displaying the handwritten Chinese characters and the Chinese characters in the database in a display system.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (5)
1. The handwritten Chinese character quality evaluation method based on deep learning is characterized by comprising the following steps:
s1, placing the paper before the collection assembly, collecting the size of the paper and the number of Chinese characters, judging whether the size of the paper can be collected normally, if so, collecting the characteristics of the handwritten Chinese characters, and if not, adjusting the position of the paper and then collecting the paper;
s2, judging whether all Chinese characters can be completely collected, if so, finishing the collection process, and if not, collecting the characteristics of a single Chinese character in a short distance;
s3, recognizing all handwritten Chinese characters, labeling the Chinese characters which cannot be recognized, providing a plurality of Chinese characters with high similarity to the Chinese characters for selection, helping to recognize the Chinese characters through manual selection, recording the Chinese characters and storing the Chinese characters in a database, so that the recognition process can be normally performed next time;
s4, comparing the recognized Chinese characters with the Chinese characters in the database, judging whether the characters to be compared exist, if so, inputting the name of the compared characters, comparing the handwritten Chinese characters with the database, if not, comparing the handwritten Chinese characters with all the characters in the database, finding the character with the highest adaptation degree, judging whether the adapted character can be found, if so, comparing the handwritten Chinese characters with the database, if not, manually inputting the name of the character to be compared, and then comparing;
s5, evaluating the Chinese characters, scoring according to the distribution positions of the Chinese characters on the paper, scoring according to the recognition degrees of the Chinese characters, scoring according to the comparison results of the Chinese characters and the fonts in the database, averaging all the scores, and finishing the scoring process;
and S6, displaying the handwritten Chinese characters and the Chinese characters in the database in the display system at the same time.
2. The method for evaluating the quality of handwritten Chinese characters based on deep learning of claim 1, wherein in said step S5, the positions of Chinese characters are scored according to their distribution positions on the paper, the more reasonable the distribution of Chinese character positions according to the size of the paper, the higher the score, the closer the font size degree, and the higher the score.
3. The method for evaluating the quality of handwritten Chinese characters based on deep learning of claim 1, wherein in step S5, the degree of recognition of Chinese characters is graded according to their degree, and the higher the degree of recognition of Chinese characters is, the higher the grade is.
4. The method for evaluating the quality of handwritten Chinese characters based on deep learning of claim 1, wherein in step S5, the score is given according to the comparison result between the Chinese characters and the fonts in the database, and the higher the degree of fit, the higher the score.
5. The method of claim 1, wherein in step S6, the handwritten chinese characters are displayed in the display system simultaneously with the chinese characters in the database, wherein the chinese characters in the database are red at the bottom and the handwritten chinese characters are light blue at the top, and the two are displayed in an overlapping manner for easy comparison by the user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110559715.0A CN113240059A (en) | 2021-05-21 | 2021-05-21 | Handwritten Chinese character quality evaluation method based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110559715.0A CN113240059A (en) | 2021-05-21 | 2021-05-21 | Handwritten Chinese character quality evaluation method based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113240059A true CN113240059A (en) | 2021-08-10 |
Family
ID=77138201
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110559715.0A Pending CN113240059A (en) | 2021-05-21 | 2021-05-21 | Handwritten Chinese character quality evaluation method based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113240059A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101295371A (en) * | 2008-04-18 | 2008-10-29 | 浙江大学 | Computer estimation method of Chinese character writing morphology beauty |
US20180005058A1 (en) * | 2016-06-29 | 2018-01-04 | Konica Minolta Laboratory U.S.A., Inc. | Path score calculating method for intelligent character recognition |
CN109214471A (en) * | 2018-10-10 | 2019-01-15 | 北京米蓝科技有限公司 | Evaluate the method and system of the written word in copybook of practising handwriting |
WO2019101338A1 (en) * | 2017-11-24 | 2019-05-31 | Ecole Polytechnique Federale De Lausanne (Epfl) | Method of handwritten character recognition confirmation |
CN110796131A (en) * | 2019-10-31 | 2020-02-14 | 黑龙江文茁教育科技有限公司 | Chinese character writing evaluation system |
CN112183046A (en) * | 2020-09-25 | 2021-01-05 | 北京有竹居网络技术有限公司 | Method, device and equipment for evaluating quality of handwriting and storage medium |
CN112434699A (en) * | 2020-11-25 | 2021-03-02 | 杭州六品文化创意有限公司 | Automatic extraction and intelligent scoring system for handwritten Chinese characters or components and strokes |
-
2021
- 2021-05-21 CN CN202110559715.0A patent/CN113240059A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101295371A (en) * | 2008-04-18 | 2008-10-29 | 浙江大学 | Computer estimation method of Chinese character writing morphology beauty |
US20180005058A1 (en) * | 2016-06-29 | 2018-01-04 | Konica Minolta Laboratory U.S.A., Inc. | Path score calculating method for intelligent character recognition |
WO2019101338A1 (en) * | 2017-11-24 | 2019-05-31 | Ecole Polytechnique Federale De Lausanne (Epfl) | Method of handwritten character recognition confirmation |
CN109214471A (en) * | 2018-10-10 | 2019-01-15 | 北京米蓝科技有限公司 | Evaluate the method and system of the written word in copybook of practising handwriting |
CN110796131A (en) * | 2019-10-31 | 2020-02-14 | 黑龙江文茁教育科技有限公司 | Chinese character writing evaluation system |
CN112183046A (en) * | 2020-09-25 | 2021-01-05 | 北京有竹居网络技术有限公司 | Method, device and equipment for evaluating quality of handwriting and storage medium |
CN112434699A (en) * | 2020-11-25 | 2021-03-02 | 杭州六品文化创意有限公司 | Automatic extraction and intelligent scoring system for handwritten Chinese characters or components and strokes |
Non-Patent Citations (3)
Title |
---|
RONGJU SUN 等: "Aesthetic Visual Quality Evaluation of Chinese Handwritings", 《2015 24TH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE(IJCAI)》, 25 July 2015 (2015-07-25), pages 2510 - 2516 * |
WEIHUA AN: "Quality Evaluation and Key Defects Correction for Handwritten Chinese Characters", 《2019 6TH INTERNATIONAL CONFERENCE ON SYSTEMS AND INFORMATICS(ICSAI)》, 27 February 2020 (2020-02-27), pages 752 - 756 * |
王求真 等: "汉字书写质量的模糊分析方法", 《计算机工程与应用》, vol. 49, no. 21, 1 November 2013 (2013-11-01), pages 180 - 185 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104376315B (en) | A kind of detection method and application based on Computer Image Processing and pattern-recognition | |
CN108090429B (en) | Vehicle type recognition method for graded front face bayonet | |
CN103324609B (en) | Text proofreading apparatus and text proofreading method | |
CN103810506B (en) | A kind of hand-written Chinese character strokes recognition methods | |
CN101038686B (en) | Method for recognizing machine-readable travel certificate | |
CN101937286B (en) | Light pen track identification system and method | |
CN105045780B (en) | A kind of recognition methods of invoice brief note semantic information and device | |
CN105426890B (en) | A kind of graphical verification code recognition methods of character distortion adhesion | |
CN110796131A (en) | Chinese character writing evaluation system | |
CN101630362B (en) | Chinese character writing quality evaluation method based on confidence level | |
CN106228157B (en) | Coloured image word paragraph segmentation and recognition methods based on image recognition technology | |
CN101777060A (en) | Automatic evaluation method and system of webpage visual quality | |
CN104794485B (en) | A kind of method and device for identifying written word | |
CN104598885A (en) | Method for detecting and locating text sign in street view image | |
CN101470802B (en) | Object detection apparatus and method thereof | |
CN111814751A (en) | Vehicle attribute analysis method and system based on deep learning target detection and image recognition | |
CN108235115B (en) | Method and terminal for positioning human voice area in song video | |
CN105719328A (en) | Normalizing method and system of hand drawing geometric figures | |
CN115240210A (en) | System and method for auxiliary exercise of handwritten Chinese characters | |
CN104715157A (en) | Cognition impairment evaluating system and method based on clock drawing test | |
Ramaiah et al. | Accent detection in handwriting based on writing styles | |
CN111860487A (en) | Inscription marking detection and recognition system based on deep neural network | |
CN113240059A (en) | Handwritten Chinese character quality evaluation method based on deep learning | |
CN108241862A (en) | A kind of inspection item information automation extraction method of medical inspection list | |
CN109508714B (en) | Low-cost multi-channel real-time digital instrument panel visual identification method and system |
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 |
Application publication date: 20210810 |
|
RJ01 | Rejection of invention patent application after publication |