CN109032383B - Input method based on handwriting recognition - Google Patents
Input method based on handwriting recognition Download PDFInfo
- Publication number
- CN109032383B CN109032383B CN201811068249.0A CN201811068249A CN109032383B CN 109032383 B CN109032383 B CN 109032383B CN 201811068249 A CN201811068249 A CN 201811068249A CN 109032383 B CN109032383 B CN 109032383B
- Authority
- CN
- China
- Prior art keywords
- standard
- character
- library
- equal
- probability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
- G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
- G06F3/0233—Character input methods
Abstract
The invention relates to the technical field of handwriting input, and particularly discloses an input method based on handwriting recognition, which comprises the steps of providing a character standard library, wherein each standard internal code establishes a corresponding relation with each standard character in the character standard library according to a specific probability; converting characters input by handwriting into handwriting internal codes; comparing the handwritten internal code with all standard internal codes, and then marking at least one standard internal code which is similar to or identical with the handwritten internal code as an internal code to be selected; listing standard characters corresponding to the internal code to be selected according to the probability from high to low for selection, and marking the selected standard characters as target characters; and adjusting the probability by combining the handwritten internal code and the target character. The invention provides an input method based on handwriting recognition, and the recognition accuracy can be gradually improved along with the increase of the use frequency of a user.
Description
Technical Field
The invention relates to the technical field of handwriting input, in particular to an input method based on handwriting recognition.
Background
Handwriting Recognition (Hand Writing Recognition) refers to a process of converting ordered tracks generated during Writing on a handwriting device into character internal codes in an informationization manner, is actually a mapping process from a coordinate sequence of a handwriting track to the character internal codes, and is one of the most natural and convenient means for man-machine interaction.
Handwriting recognition of characters is an important research field of pattern recognition, and has been widely researched and paid attention in recent decades, and with the emergence of deep learning technology, handwriting recognition of characters based on deep learning has been developed in a breakthrough manner in terms of method and performance in recent years.
At present, the following two technologies are mainly used for realizing handwriting recognition:
1) the deep learning method comprises the following steps: the handwriting recognition method based on deep learning end-to-end (namely, inputting a picture to a mathematical model and outputting a recognition result by the mathematical model) needs to firstly write a large amount of character data by handwriting and label each character, and then the deep neural network used by the deep learning method can be trained to enable the model to learn the mode of each character. After a large amount of data is trained for a long enough time, the model can recognize the handwritten characters more accurately.
2) The single character characteristic method comprises the following steps: the method based on the character features includes extracting stroke order, strokes, shapes, outlines and components, converting the features into numbers, expressing Chinese characters with the numbers and establishing Chinese character database. Extracting data from the characters generated during handwriting according to a method for establishing a template library and a database, and matching the data with the data in the database to find out similar Chinese characters.
The above two methods have the following disadvantages: the handwriting habit of the user cannot be adaptively updated, so that the recognition accuracy cannot be improved along with the increase of the use frequency of the user.
Disclosure of Invention
An object of the present invention is to provide an input method based on handwriting recognition, in which the recognition accuracy can be gradually improved as the frequency of use of users increases.
To achieve the above object, the present invention provides an input method based on handwriting recognition, comprising:
providing a character standard library, wherein each standard internal code establishes a corresponding relation with each standard character in the character standard library according to a specific probability;
converting characters input by handwriting into handwriting internal codes;
comparing the handwritten internal code with all standard internal codes, and then marking at least one standard internal code which is similar to or identical with the handwritten internal code as an internal code to be selected;
listing standard characters corresponding to the internal code to be selected from high to low according to probability for selection, and marking the selected standard characters as target characters;
and adjusting the probability by combining the handwritten internal code and the target character.
As a preferred embodiment, the step of providing the character standard library specifically includes:
and acquiring a character standard library based on a single character characteristic method.
In a preferred embodiment, the standard character is at least one of greek letters, arabic numerals, english letters, chinese characters, chinese character strokes, chinese character components, and punctuation marks.
As a preferred embodiment, the step of providing a character standard library, where each standard inner code establishes a corresponding relationship with each standard character in the character standard library with a specific probability specifically includes:
providing a standard library of characters (C, P) having N standard characters, each standard inner code C i All with probability p in Establishing a corresponding relation with each standard character;
wherein:
i: i is a positive integer, and i is not more than N;
n: n is a positive integer and is less than or equal to N;
c: a standard inner code;
p: the probability that the standard inner code corresponds to the standard character;
c i : a standard inner code corresponding to a feature vector extracted from the ith standard character of the character standard library (C, P);
p in ：c i probability of the nth standard character from the character standard library (C, P).
As a preferred implementation, the step of adjusting the probability by combining the handwritten inner code and the target character includes:
the handwritten inner code and the probability relation (c) between the handwritten inner code and the standard character N+1 ,p (N+1)n ) Adding into the character standard library (C, P) to form a non-standard character library
Establishing a standard (C, P) and non-standard character library
Match error function therebetween
Wherein:
λ：0≤λ≤1；
c k : the standard inner code corresponding to the feature vector extracted from the kth standard character of the character standard library (C, P) is in C k The method comprises the following steps: k is more than or equal to 1 and less than or equal to N;
p jk ：c j probability of the kth standard character from the character standard library (C, P), at P jk The method comprises the following steps: j is more than or equal to 1 and less than or equal to N, and k is more than or equal to 1 and less than or equal to N;
obtaining
When the minimum value c is obtained k And p jk Taking the value of (A);
according to c k And p jk And solving for C and P, thereby obtaining an adjusted character standard library (C, P).
The invention has the beneficial effects that: the input method based on handwriting recognition is characterized in that a recognition result of handwriting input is added into an original character standard library to adjust the character standard library, and along with repeated adjustment for multiple times, the recommended character sequence can be continuously adjusted according to the handwriting habit of a user, so that the recognition accuracy can be gradually improved along with the increase of the use frequency of the user, and the recognition accuracy is more and more in line with the use habit of the user.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a block diagram of an input method based on handwriting recognition according to an embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be understood that when an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. When a component is referred to as being "disposed on" another component, it can be directly on the other component or intervening components may also be present.
Furthermore, the terms "long", "short", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of describing the present invention, but do not indicate or imply that the referred devices or elements must have the specific orientations, be configured to operate in the specific orientations, and thus are not to be construed as limitations of the present invention.
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
As shown in fig. 1, the present embodiment provides an input method based on handwriting recognition, which includes steps S10 to S50.
S10: providing a character standard library, wherein each standard inner code establishes a corresponding relation with each standard character in the character standard library according to a specific probability.
Preferably, the character standard library can be obtained based on a single character feature method, and the character standard library can also be directly imported from other databases. The initial library of character criteria is only a prototype, providing the basis for the subsequent adjustment steps. Further, the standard character is at least one of Greek letters, Arabic numerals, English letters, Chinese characters, Chinese character strokes, Chinese character components and punctuation marks.
In particular, a character standard library often contains a large number of standard characters. If the symbol handwritten by the user is identical to the standard character in the character standard library, the handwritten internal code of the handwritten character is consistent with the standard internal code.
When a user writes a symbol (such as 'one'), the symbol is firstly converted into a standard inner code, however, a plurality of standard characters all contain the standard inner code of the symbol (for example, the standard inner codes corresponding to the standard characters such as 'one', 'two' or 'three' all contain the standard inner code of 'one'), so that the standard inner code of the symbol has different probability relations with different standard characters (it can be understood that the standard code of 'one' is completely the same as the standard code of 'one', so that the probability between the standard code of 'one' and the standard character 'one' is definitely higher, the probability between the standard code of 'one' and the standard character 'two' is the second, the probability between the standard code of 'one' and the standard character 'three' is lower, and the system recommends an identification result, such as a recommendation result of 1, to the user according to the high-low relation of the probability, Firstly, performing primary filtration; 2. II, performing secondary filtration; 3. three, then for user selection).
Further, S10 may specifically be: providing a standard library of characters (C, P) having N characters, each standard inner code C i All with a probability p in And establishing corresponding relation with each standard character.
Wherein:
i: i is a positive integer, and i is less than or equal to N;
n: n is a positive integer and is less than or equal to N;
c: a standard inner code;
p: the probability that the standard inner code corresponds to the standard character;
c i : extracting a standard inner code corresponding to the characteristic vector from the ith standard character;
p in ：c i from the first n Probability of individual standard characters. Initially, the character criteria library (C, P) is not adjusted, so for C i In addition to p ii 1, the remainder of p i1 、p i2 、p i3 ……p i(i-1) 、p i(i+1) ……p in Are all 0.
S20: and converting the characters input by handwriting into a handwritten internal code.
Specifically, the characters input by handwriting often include a plurality of feature vectors, and the handwriting internal code of the characters input by handwriting is obtained by converting the feature vectors into codes.
Further, since the original standard library of characters already contains N standard characters, the handwritten inner code is labeled c N+1 Accordingly, c N+1 The probability between the standard character and each standard character in the standard character library is p N+1 . It will be appreciated that, initially, c is due to the fact that c is before the selection is made N+1 There is temporarily no association with any standard character in the standard character library, so all p's are now present N+1 Are all 0.
S30: and comparing the handwritten internal code with all standard internal codes, and then marking at least one standard internal code which is similar to or identical with the handwritten internal code as an internal code to be selected.
In particular, since it is almost impossible for the handwritten font to completely coincide with the standard character, the handwritten inner code may be different from the standard inner code. However, since the same feature vector exists for both the handwritten font and the standard character (e.g., the handwritten "person" and the standard character "person" are both first left-falling and then right-falling). Therefore, a standard inner code similar to or identical to the handwritten inner code exists (for example, the handwritten input 'person' is obtained, and the standard inner codes of 'person', 'enter' and 'eight', etc. are marked as the inner code to be selected because the obtained feature vector is 'left falling first and then right falling').
S40: and listing the standard characters corresponding to the selected inner code from high to low according to the probability for selection, and marking the selected standard characters as target characters.
Specifically, for example, "person" is input by handwriting, and the standard inner codes with feature vectors of "first left-falling and then right-falling" are all marked as the inner codes to be selected, so that "person", "in" and "eight" are all preferentially recommended as the standard characters with higher probability. The standard characters such as "" one "" and "" big "" are the next.
Further, when the target character is selected, c is compared with c N+1 Associated p (N+1)n E.g. the Yth standard character is selected as the target character, then for c N+1 To say, in addition top (N+1)Y 1, the remainder of p i1 、p i2 、p i3 ……p (N+1)(Y-1) 、p (N+1)(Y+1) ……p (N+1)n Are all 0. Of course, p is adjusted continuously according to the handwritten code (N+1)n Also, the value of (A) may change continuously, for example, p appears i1 ＝0.1、p i2 ＝0、p i3 ＝0.7……p (N+1)(Y-1) ＝0.8、p (N+1)Y ＝1、p (N+1)(Y+1) ＝0……p (N+1)n 0.4, etc.
S50: and adjusting the probability by combining the handwritten internal code and the target character.
Specifically, S50 includes:
s501: the handwritten inner code and the probability relation (c) between the handwritten inner code and the standard character N+1 ,p (N+1)n ) Adding into the character standard library (C, P) to form a non-standard character library
S502: establishing a standard library (C, P) and a non-standard library of characters
Match error function between
Wherein:
λ：0≤λ≤1；
c k : the standard inner code corresponding to the feature vector extracted from the kth standard character of the character standard library (C, P) is in C k The method comprises the following steps: k is more than or equal to 1 and less than or equal to N;
p jk ：c j probability of the kth standard character from the character standard library (C, P), at P jk The method comprises the following steps: j is more than or equal to 1 and less than or equal to N, and k is more than or equal to 1 and less than or equal to N;
S503: obtaining
When the minimum value c is obtained k And p jk Taking the value of (A); in particular, it can be appreciated that the match error function
Is about c k And p jk Function of (a) present c k And p jk So as to match the error function
Taking the minimum value. The significance of the minimum is that it is desirable to add new data (c) N+1 ,p (N+1)n ) Thereafter, the new data is minimized (c) N+1 ,p (N+1)n ) The influence on the original character standard libraries (C, P) is beneficial to improving the stability of the whole system and effectively preventing the accidental handwriting result from having excessive influence on the recognition system.
S504: according to c k And p jk And solving for C and P, thereby obtaining an adjusted character standard library (C, P).
Specifically, S10-S50 are repeatedly executed, and with repeated adjustment for many times, the input method based on handwriting recognition provided by the present embodiment continuously adjusts the recommended character sequence according to the handwriting habit of the user, and the recognition accuracy can be gradually improved as the use frequency of the user increases, and is more and more in line with the use habit of the user.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (3)
1. An input method based on handwriting recognition, comprising:
providing a character standard library, wherein each standard internal code establishes a corresponding relation with each standard character in the character standard library according to a specific probability;
converting characters input by handwriting into handwriting internal codes;
comparing the handwritten internal code with all standard internal codes, and then marking at least one standard internal code which is similar to or identical with the handwritten internal code as an internal code to be selected;
listing standard characters corresponding to the internal code to be selected from high to low according to probability for selection, and marking the selected standard characters as target characters;
adjusting the probability by combining the handwritten internal code and the target character;
the step of providing a character standard library, wherein each standard internal code establishes a corresponding relationship with each standard character in the character standard library with a specific probability specifically comprises the following steps:
providing a standard library of characters (C, P) having N standard characters, each standard inner code C i All with a probability p in Establishing a corresponding relation with each standard character;
wherein:
i: i is a positive integer, and i is not more than N;
n: n is a positive integer and is less than or equal to N;
c: a standard inner code;
p: the probability that the standard inner code corresponds to the standard character;
c i : a standard inner code corresponding to a feature vector extracted from the ith standard character of the character standard library (C, P);
p in ：c i probability of the nth standard character from the character standard library (C, P);
the step of adjusting the probability by combining the handwritten inner code and the target character comprises the following steps:
the handwritten inner code and the probability relation (c) between the handwritten inner code and the standard character N+1 ,p (N+1)n ) Adding into the character standard library (C, P) to form a non-standard character library
Establishing a standard (C, P) and non-standard character library
Match error function therebetween
Wherein:
λ：0≤λ≤1；
c k : the standard inner code corresponding to the feature vector extracted from the kth standard character of the character standard library (C, P) is in C k The method comprises the following steps: k is more than or equal to 1 and less than or equal to N;
p jk ：c j probability of the kth standard character from the character standard library (C, P), at P jk The method comprises the following steps: j is more than or equal to 1 and less than or equal to N, and k is more than or equal to 1 and less than or equal to N;
obtaining
When the minimum value c is obtained k And p jk Taking the value of (A);
according to c k And p jk And solving for C and P, thereby obtaining an adjusted character standard library (C, P).
2. The handwriting recognition-based input method according to claim 1, wherein said step of providing a standard library of characters is specifically:
and acquiring a character standard library based on a single character characteristic method.
3. The handwriting recognition-based input method according to claim 1 or 2, wherein the standard character is at least one of greek letters, arabic numerals, english letters, chinese characters, chinese character strokes, chinese character radicals and punctuation marks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811068249.0A CN109032383B (en) | 2018-09-13 | 2018-09-13 | Input method based on handwriting recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811068249.0A CN109032383B (en) | 2018-09-13 | 2018-09-13 | Input method based on handwriting recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109032383A CN109032383A (en) | 2018-12-18 |
| CN109032383B true CN109032383B (en) | 2022-09-16 |
Family
ID=64622085
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811068249.0A Active CN109032383B (en) | 2018-09-13 | 2018-09-13 | Input method based on handwriting recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109032383B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111310547B (en) * | 2019-12-04 | 2023-05-30 | 武汉汉德瑞庭科技有限公司 | Method for extracting and authenticating stroke order characteristics in online handwriting authentication |
| CN111222503B (en) * | 2020-04-16 | 2020-09-08 | 湖南师范大学 | Handwritten spectrum character recognition method of guqin abbreviated character spectrum fingering method |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5392367A (en) * | 1991-03-28 | 1995-02-21 | Hsu; Wen H. | Automatic planar point pattern matching device and the matching method thereof |
| CN1260524A (en) * | 1999-01-13 | 2000-07-19 | 国际商业机器公司 | Method for automatically cutting and identiying hand written Chinese characters and system for using said method |
| US6094506A (en) * | 1995-10-25 | 2000-07-25 | Microsoft Corporation | Automatic generation of probability tables for handwriting recognition systems |
| CN101135946A (en) * | 2006-08-30 | 2008-03-05 | 黄金富 | Chinese characters hand-written input method |
| CN101149804A (en) * | 2006-09-19 | 2008-03-26 | 北京三星通信技术研究有限公司 | Self-adaptive hand-written discrimination system and method |
| CN101477425A (en) * | 2009-01-08 | 2009-07-08 | 广东国笔科技股份有限公司 | Method and system for recognizing hand-written character input |
| CN102446055A (en) * | 2010-09-30 | 2012-05-09 | 阿尔派株式会社 | Character handwriting input method and device |
| CN105809102A (en) * | 2015-01-16 | 2016-07-27 | 新普科技股份有限公司 | Handwriting Track Recognition Method and System |
| CN106570457A (en) * | 2016-10-14 | 2017-04-19 | 上海新同惠自动化系统有限公司 | Chinese and Japanese character identification method |
| CN107169517A (en) * | 2017-05-11 | 2017-09-15 | 广东小天才科技有限公司 | Judge method, terminal device and the computer-readable recording medium of repetition stroke |
| CN107833500A (en) * | 2017-11-17 | 2018-03-23 | 广东小天才科技有限公司 | A kind of learning method based on hand-written terminal and hand-written terminal |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6836759B1 (en) * | 2000-08-22 | 2004-12-28 | Microsoft Corporation | Method and system of handling the selection of alternates for recognized words |
| US20050192802A1 (en) * | 2004-02-11 | 2005-09-01 | Alex Robinson | Handwriting and voice input with automatic correction |
-
2018
- 2018-09-13 CN CN201811068249.0A patent/CN109032383B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5392367A (en) * | 1991-03-28 | 1995-02-21 | Hsu; Wen H. | Automatic planar point pattern matching device and the matching method thereof |
| US6094506A (en) * | 1995-10-25 | 2000-07-25 | Microsoft Corporation | Automatic generation of probability tables for handwriting recognition systems |
| CN1260524A (en) * | 1999-01-13 | 2000-07-19 | 国际商业机器公司 | Method for automatically cutting and identiying hand written Chinese characters and system for using said method |
| CN101135946A (en) * | 2006-08-30 | 2008-03-05 | 黄金富 | Chinese characters hand-written input method |
| CN101149804A (en) * | 2006-09-19 | 2008-03-26 | 北京三星通信技术研究有限公司 | Self-adaptive hand-written discrimination system and method |
| CN101477425A (en) * | 2009-01-08 | 2009-07-08 | 广东国笔科技股份有限公司 | Method and system for recognizing hand-written character input |
| CN102446055A (en) * | 2010-09-30 | 2012-05-09 | 阿尔派株式会社 | Character handwriting input method and device |
| CN105809102A (en) * | 2015-01-16 | 2016-07-27 | 新普科技股份有限公司 | Handwriting Track Recognition Method and System |
| CN106570457A (en) * | 2016-10-14 | 2017-04-19 | 上海新同惠自动化系统有限公司 | Chinese and Japanese character identification method |
| CN107169517A (en) * | 2017-05-11 | 2017-09-15 | 广东小天才科技有限公司 | Judge method, terminal device and the computer-readable recording medium of repetition stroke |
| CN107833500A (en) * | 2017-11-17 | 2018-03-23 | 广东小天才科技有限公司 | A kind of learning method based on hand-written terminal and hand-written terminal |
Non-Patent Citations (3)
| Title |
|---|
| Bounding the Probability of Error for High Precision Optical Character Recognition;Gary B. Huang等;《Journal of Machine Learning Research》;20120228;363-387 * |
| 基于模板匹配的手写字符识别技术实现;王一丹等;《甘肃联合大学学报》;20120910;第26卷(第5期);53-58 * |
| 基于统计模型的DTW 签名认证系统;鄢晨丹等;《技术研究》;20150710(第07期);64-70 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109032383A (en) | 2018-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110909548B (en) | Chinese named entity recognition method, device and computer readable storage medium | |
| CN102449640B (en) | Recognizing handwritten words | |
| CN110009027B (en) | Image comparison method and device, storage medium and electronic device | |
| US8559723B2 (en) | Letter model and character bigram based language model for handwriting recognition | |
| CN110334179B (en) | Question-answer processing method, device, computer equipment and storage medium | |
| CN107004140A (en) | Text recognition method and computer program product | |
| CN107491729B (en) | Handwritten digit recognition method based on cosine similarity activated convolutional neural network | |
| EP3825920A1 (en) | Positional embeddings for document processing | |
| US6035063A (en) | Online character recognition system with improved standard strokes processing efficiency | |
| CN109032383B (en) | Input method based on handwriting recognition | |
| US8543382B2 (en) | Method and system for diacritizing arabic language text | |
| CN114730241B (en) | Gesture and stroke recognition in touch user interface input | |
| CN110610180A (en) | Method, device and equipment for generating recognition set of wrongly-recognized words and storage medium | |
| US6731803B1 (en) | Points based handwriting recognition system | |
| CN113673228A (en) | Text error correction method, text error correction device, computer storage medium and computer program product | |
| US8665221B2 (en) | Method and apparatus for inputting Chinese characters | |
| JP2007213433A (en) | Character retrieving apparatus | |
| CN110414622A (en) | Classifier training method and device based on semi-supervised learning | |
| US20140022180A1 (en) | Method for Inputting and Searching Chinese Characters with Easy-Strokes | |
| CN111506726A (en) | Short text clustering method and device based on part-of-speech coding and computer equipment | |
| CN102360265B (en) | The method and device of word selection is treated in determination in a kind of handwriting input | |
| US7295707B2 (en) | Method for aligning gesture features of image | |
| US20150063698A1 (en) | Assisted OCR | |
| CN112800752B (en) | Error correction method, apparatus, device and storage medium | |
| CN112241738B (en) | Mathematical formula character recognition method and device and electronic equipment |
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 |