CN102880874B - Character identifying method and Character recognizer - Google Patents
Character identifying method and Character recognizer Download PDFInfo
- Publication number
- CN102880874B CN102880874B CN201210374937.6A CN201210374937A CN102880874B CN 102880874 B CN102880874 B CN 102880874B CN 201210374937 A CN201210374937 A CN 201210374937A CN 102880874 B CN102880874 B CN 102880874B
- Authority
- CN
- China
- Prior art keywords
- character
- cryptographic hash
- picture
- word
- hash
- 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
Landscapes
- Character Discrimination (AREA)
Abstract
Character identifying method provided by the invention, comprises the steps: b, utilizes hash algorithm to calculate the cryptographic hash of character picture in picture to be identified; C, from the character set cryptographic hash training storehouse set up in advance, inquire about cryptographic hash, obtain the character that cryptographic hash is corresponding, wherein b comprises: b1, from picture to be identified, extract word character picture; B2, monochromatic filter process is carried out to described word character picture; B3, described hash algorithm is utilized to calculate the cryptographic hash of described word character picture; Wherein, character set cryptographic hash training storehouse comprises: character code and the cryptographic hash corresponding with described character code; C comprises: c1, from set up in advance character set cryptographic hash training storehouse inquire about described cryptographic hash, obtain the character code that described cryptographic hash is corresponding.C2, the described character code obtained is converted to corresponding character.Present invention also offers a kind of Character recognizer.Adopt character identifying method of the present invention or Character recognizer, the recognition speed to character can be improved.
Description
Technical field
The present invention relates to character recognition technologies, particularly relate to a kind of character identifying method and Character recognizer.
Background technology
Character recognition refers to the technology of the automatic identification character such as the equipment that utilizes computing machine, and wherein character comprises: Chinese character, English alphabet, numeral, punctuate etc., namely such as: comprising: 1,2,3, A, B, C, #, word, etc.Traditional character recognition technologies, such as OCR (OpticalCharacterRecognition, optical character identification), ubiquitous problem is: when processing the document of a large amount of fixed character collection, treatment effeciency is low, and this is mainly owing to carrying out complicated calculating to each character in traditional character recognition technology.
Summary of the invention
In view of this, the invention provides a kind of character identifying method and Character recognizer.The recognition speed to character can be improved.
The character identifying method that the embodiment of the present invention provides, comprises the steps:
B, hash algorithm is utilized to calculate the cryptographic hash of character picture in picture to be identified;
C, from set up in advance character set cryptographic hash training storehouse inquire about described cryptographic hash, obtain the character that described cryptographic hash is corresponding;
Wherein, described b comprises:
B1, from described picture to be identified, extract word character picture;
B2, monochromatic filter process is carried out to described word character picture;
B3, described hash algorithm is utilized to calculate the cryptographic hash of described word character picture;
Wherein, described character set cryptographic hash training storehouse comprises: character code and the cryptographic hash corresponding with described character code; Described c comprises:
C1, from set up in advance character set cryptographic hash training storehouse inquire about described cryptographic hash, obtain the character code that described cryptographic hash is corresponding;
C2, the described character code obtained is converted to corresponding character.
Further, also comprise before described b: a, set up character set cryptographic hash training storehouse;
Described a comprises:
A1, the word character picture extracted in character set;
A2, monochromatic filter process is carried out to the monocase picture of described extraction;
A3, hash algorithm is utilized to calculate the cryptographic hash of the word character picture of described extraction;
A4, the character code of the cryptographic hash of the monocase picture of described extraction and described word character picture is stored into character set cryptographic hash training storehouse accordingly.
Further, described hash algorithm adopts the 5th edition Message Digest 5 MD5.
Further, described character code adopts binary coding.
Further, in the described character set cryptographic hash training storehouse set up in advance, the form store character of table coding and the cryptographic hash corresponding with described character code is adopted.
Correspondingly, the embodiment of the present invention additionally provides a kind of Character recognizer, comprising:
First processing unit, for the cryptographic hash utilizing hash algorithm to calculate character picture in picture to be identified;
Second processing unit, for inquiring about the cryptographic hash that described first processing unit calculates from the character set cryptographic hash training storehouse set up in advance, obtains the character that described cryptographic hash is corresponding;
Wherein, the first processing unit comprises:
Extract subelement, for extracting word character picture from described picture to be identified;
First filter process subelement, carries out monochromatic filter process for the word character picture extracted described extraction subelement;
First computation subunit, for the cryptographic hash utilizing hash algorithm to calculate the word character picture after described first filter process subelement process;
Wherein, described character set cryptographic hash training storehouse comprises: character code and the cryptographic hash corresponding with described character code; Described second processing unit, comprising:
Inquiry subelement, for inquiring about described cryptographic hash from the character set cryptographic hash training storehouse set up in advance, obtains the character code that described cryptographic hash is corresponding;
Conversion subelement, is converted to corresponding character for the character code inquired by described inquiry subelement.
Further, also comprise: set up unit, for setting up character set cryptographic hash training storehouse, described unit of setting up comprises:
Extract subelement, for extracting the word character picture in character set;
Second filter process subelement, carries out monochromatic filter process for the monocase picture extracted described extraction subelement;
Second computation subunit, for the cryptographic hash utilizing hash algorithm to calculate the word character picture after described second filter process subelement process;
Storing sub-units, is stored into character set cryptographic hash training storehouse accordingly for the cryptographic hash of the monocase picture by described extraction and the character code of described word character picture.
Further, described hash algorithm adopts the 5th edition Message Digest 5 MD5.
Further, described character code adopts binary coding.
Further, in the described character set cryptographic hash training storehouse set up in advance, the form store character of table coding and the cryptographic hash corresponding with described character code is adopted.
Beneficial effect of the present invention:
The embodiment of the present invention is when identifying the character in picture, by calculating the cryptographic hash of the character picture in picture, then in the character set cryptographic hash training storehouse set up in advance, this cryptographic hash is inquired about, namely can obtain the character that the cryptographic hash of this character picture is corresponding, namely achieve the identification of character.Due to this kind of form, by the character set cryptographic hash training storehouse set up in advance, cryptographic hash is directly utilized to compare, character is identified, because this simplify character recognition process, the recognition speed to character can be improved, being applicable to process the huge picture information of batch.
Accompanying drawing explanation
Below in conjunction with drawings and Examples, the invention will be further described:
Fig. 1 is the schematic flow sheet of the first embodiment of character identifying method of the present invention.
Fig. 2 is the schematic flow sheet of the embodiment of step S11 in Fig. 1.
Fig. 3 is the schematic flow sheet of the embodiment of step S12 in Fig. 1.
Fig. 4 is the schematic flow sheet of the second embodiment of character identifying method of the present invention.
Fig. 5 is the schematic flow sheet of the embodiment of step S10 in Fig. 4.
Fig. 6 is the structural representation of the first embodiment of Character recognizer of the present invention.
Fig. 7 is the structural representation of the embodiment of the first processing unit in Fig. 6.
Fig. 8 is the structural representation of the embodiment of the second processing unit in Fig. 6.
Fig. 9 is the structural representation of the second embodiment of Character recognizer of the present invention.
Figure 10 is the structural representation of the embodiment setting up unit in Fig. 9.
Embodiment
Please refer to Fig. 1, is the schematic flow sheet of the first embodiment of character identifying method of the present invention.The method flow process is mainly used for identifying the character in fixed character collection picture.It comprises the steps:
Step S11, hash algorithm is utilized to calculate the cryptographic hash of character picture in picture to be identified.
Wherein, hash algorithm and hash, a kind of by the function of the message compression of random length to the eap-message digest of a certain regular length, more famous hash algorithm comprises: MD2(MessageDigestAlgorithm2, second edition Message Digest 5), MD4(the 4th edition Message Digest 5), MD5(the 5th edition Message Digest 5) and SHA(SecureHashAlgorithm, Secure Hash Algorithm).Wherein MD5 be a kind of comparative maturity, the hash algorithm that is widely used, therefore select MD5 to calculate the cryptographic hash of character picture in picture to be identified in the present embodiment.
Introduce a kind of embodiment of step S11 below.As shown in Figure 2, step S11 comprises the steps: in one embodiment
Step S111, from described picture to be identified, extract word character picture.
Step S112, monochromatic filter process is carried out to described word character picture.
Step S113, described hash algorithm is utilized to calculate the cryptographic hash of described word character picture.
Wherein, step S111 picture to be identified may comprise the character that many needs identify, these characters extract by step S111 one by one from picture to be identified, obtain word character picture.After step S111 obtains word character picture, because the color of character in practice can change, but the form of font is constant, therefore by step S112, word character picture can be carried out monochromatic filter process, word character picture is made to become monochrome, such as: black, carry out follow-up process again, the object that namely step S112 realizes is by " JPG ", Font processing in the picture of the form such as " PNG " and " GIF " is monochromatic, be monochromatic by the color treatments of picture by step S112, the follow-up impact that can not be subject to color when carrying out character recognition can be made, improve accuracy and the speed of character recognition.After above-mentioned process, then utilize hash algorithm by step S113, such as: MD5, calculate the cryptographic hash of word character picture.
Step S12, from set up in advance character set cryptographic hash training storehouse inquire about described cryptographic hash, obtain the character that described cryptographic hash is corresponding.
Wherein, character set cryptographic hash training storehouse comprises: character code and the cryptographic hash corresponding with described character code.This character set cryptographic hash training storehouse can be set up before step S11, when needs identify character, directly called this character set cryptographic hash training storehouse.Further, this character set cryptographic hash training storehouse can dynamically upgrade.In above-mentioned, character code can select " UNICODE ", " GBK ", " GB2312 ", " UTF-8 ", " UTF-16 " etc., a string character code is a corresponding character uniquely, therefore obtains the character code that cryptographic hash is corresponding, is also just equivalent to have identified character.
Character set cryptographic hash training storehouse can adopt form store character coding and the cryptographic hash of table, and their corresponding relation, in actual applications, can by " Song typeface ", " regular script ", " Arial ", cryptographic hash (such as hash algorithm adopts MD5) asked by the character picture of the fixed word bodily form formula that ordinary symbols such as " black matrixes " is concentrated, stored in character set cryptographic hash training storehouse, character code can adopt UTF-8 coding to show, wherein the UTF-8 of character " you " is encoded to " 0xE4BDA0 ", the UTF-8 of character " 1 " is encoded to " 0x31 ", the UTF-8 of character " 9 " is encoded to " 0x39 ", the UTF-8 of character " good " is encoded to " 0xE5A5BD ", such as adopt the form of table one, it is the cryptographic hash that the character picture of " you " " 1 " " 9 " " good " in " black matrix " character set is calculated, cryptographic hash corresponding to each character code is see table one.
Table one
| Cryptographic hash | Character code |
| 0C01C48C3D7BB3592AC58BFF8018D3BF | 0xE4BDA0 |
| C4CA4238A0B923820DCC509A6F75849B | 0x31 |
| 45C48CCE2E2D7FBDEA1AFC51C7C6AD26 | 0x39 |
| 21D4300CFCD0E37CC53CA917A1E2970C | 0xE5A5BD |
| ...... | ...... |
In addition, because the character picture of same character (as character " 1 ") in multiple character set (as the character set such as " Song typeface ", " regular script ", " Arial ", " black matrix ") is different, therefore the difference of character picture causes the cryptographic hash drawn also to be different.When identification character time, using cryptographic hash as keyword query.
Introduce a kind of embodiment of step S12 below.As shown in Figure 3, step S12 comprises the steps: in one embodiment
Step S121, from set up in advance character set cryptographic hash training storehouse inquire about described cryptographic hash, obtain the character code that described cryptographic hash is corresponding.
Step S122, the described character code obtained is converted to corresponding character.
Such as: in character set cryptographic hash training storehouse, there is following corresponding relation: the cryptographic hash (adopting MD5 to obtain) that the character code of " 9 " is corresponding is " 45C48CCE2E2D7FBDEA1AFC51C7C6AD26 ".If in certain character recognition, the cryptographic hash obtained when the mode by step S11 is " 45C48CCE2E2D7FBDEA1AFC51C7C6AD26 ", so by step S121 namely can know this cryptographic hash corresponding be the character code of " 9 ", therefore namely step S122 can obtain character " 9 " by the character code of " 9 ", namely identifies character " 9 ".
The present embodiment is when identifying the character in picture, by calculating the cryptographic hash of the character picture in picture, then establish in advance character set cryptographic hash training storehouse in inquire about this cryptographic hash, namely can obtain the character that the cryptographic hash of this character picture is corresponding, namely achieve the identification of character.Due to this kind of form, by the character set cryptographic hash training storehouse set up in advance, cryptographic hash is directly utilized to compare, character is identified, because this simplify character recognition process, the recognition speed to character can be improved, being applicable to process the huge picture information of batch.In addition, the present embodiment, in the identifying of character, also carries out the operation of monochromatic filter process, thus makes the identification of character not by the impact of color, therefore can improve speed and the accuracy of character recognition further.Relative to the recognition speed of existing OCR technology 10 characters per second, the character identifying method of the application can reach identification per second 1000 characters.
Please refer to Fig. 4, is the schematic flow sheet of the second embodiment of character identifying method of the present invention.The key distinction of this embodiment and Fig. 1 embodiment is also to comprise: step S10, set up character set cryptographic hash and train storehouse.Wherein, as shown in Figure 5, step S10 comprises at its a kind of embodiment:
Step S101, the word character picture extracted in character set.
Step S102, monochromatic filter process is carried out to the monocase picture of described extraction.
Step S103, hash algorithm is utilized to calculate the cryptographic hash of the word character picture of described extraction;
Step S104, the character code of the cryptographic hash of the monocase picture of described extraction and described word character picture is stored into character set cryptographic hash training storehouse accordingly.
Character set wherein in step S101 can be ttf(trueTypeFont) character set, ttf character set is widely used character set in a kind of computing machine.The method of above-mentioned steps S101-step S103 and step S111-step 113 is substantially similar, its key distinction point is: the part owing to being the process setting up character set cryptographic hash training storehouse in step S101, therefore the character picture extracted from character set is known, and step S111 is owing to being the part to the process that picture to be identified identifies, its word character picture extracted is unknown.
Below the equipment that can perform said method flow process is introduced.
Please refer to Fig. 6, is the structural representation of the first embodiment of Character recognizer of the present invention.This Character recognizer 1 is mainly used for identifying the character in fixed character collection picture.It comprises:
First processing unit 11, for the cryptographic hash utilizing hash algorithm to calculate character picture in picture to be identified.
Wherein, hash algorithm and hash, a kind of by the function of the message compression of random length to the eap-message digest of a certain regular length, more famous hash algorithm comprises: MD2(MessageDigestAlgorithm2, second edition Message Digest 5), MD4(the 4th edition Message Digest 5), MD5(the 5th edition Message Digest 5) and SHA(SecureHashAlgorithm, Secure Hash Algorithm).Wherein MD5 be a kind of comparative maturity, the hash algorithm that is widely used, therefore select MD5 to calculate the cryptographic hash of character picture in picture to be identified in the present embodiment.
Introduce the embodiment of a kind of concrete structure of the first processing unit 11 below.As shown in Figure 7, the first processing unit 11 comprises as lower component at its a kind of embodiment:
Extract subelement 111, for extracting word character picture from picture to be identified.
First filter process subelement 112, for carrying out monochromatic filter process to the word character picture extracting subelement 111 extraction.
First computation subunit 113, for the cryptographic hash utilizing hash algorithm to calculate the word character picture after the first filter process subelement 112 process.
Wherein, picture to be identified may comprise the character that many needs identify, extracts subelement 111 and is extracted from picture to be identified one by one by these characters, obtain word character picture.After extraction subelement 111 obtains word character picture, because the color of character in practice can change, but the form of font is constant, therefore by the first filter process subelement 112, word character picture can be carried out monochromatic filter process, word character picture is made to become monochrome, such as: black, follow-up process is being carried out.The object that first filter process subelement 112 is equivalent to realize is for monochromatic by the Font processing in the picture of the form such as " JPG ", " PNG " and " GIF ", be monochromatic by the color treatments of picture by the first filter process subelement 112, the follow-up impact that can not be subject to color when carrying out character recognition can being made, improve accuracy and the speed of character recognition.After above-mentioned process, then utilize hash algorithm by the first computation subunit 113, such as: MD5, calculate the cryptographic hash of word character picture.
Second processing unit 12, for inquiring about the cryptographic hash that the first processing unit 11 calculates from the character set cryptographic hash training storehouse set up in advance, obtains the character that this cryptographic hash is corresponding.
Wherein, character set cryptographic hash training storehouse comprises: character code and the cryptographic hash corresponding with described character code.This character set cryptographic hash training storehouse can be set up in advance, when needs identify character, directly calls this character set cryptographic hash training storehouse.Further, this character set cryptographic hash training storehouse can dynamically upgrade.In above-mentioned, character code can select binary coding, and a string binary coding is a corresponding character uniquely, therefore obtains the binary coding that cryptographic hash is corresponding, is also just equivalent to have identified character.Character set cryptographic hash training storehouse can adopt form store character coding and the cryptographic hash of table, and their corresponding relation, shown in such as, table one in above-mentioned.
Introduce a kind of concrete structure of the second processing unit 12 below.As shown in Figure 8, the second processing unit 12 comprises:
Inquiry subelement 121, for inquiring about described cryptographic hash from the character set cryptographic hash training storehouse set up in advance, obtains the character code that described cryptographic hash is corresponding.
Conversion subelement 122, is converted to corresponding character for inquiry subelement 121 is inquired about the character code obtained.
Such as: in character set cryptographic hash training storehouse, there is following corresponding relation: the cryptographic hash (adopting MD5 to obtain) that the character code of " 9 " is corresponding is " 45C48CCE2E2D7FBDEA1AFC51C7C6AD26 ".If in certain character recognition, when the cryptographic hash obtained by character recognition is " 45C48CCE2E2D7FBDEA1AFC51C7C6AD26 ", so by inquiry subelement 121 namely can know this cryptographic hash corresponding be the character code of " 9 ", therefore change subelement 122 and namely can obtain character " 9 " by the character code of " 9 ", namely identify character " 9 ".
The present embodiment is when identifying the character in picture, by calculating the cryptographic hash of the character picture in picture, then establish in advance character set cryptographic hash training storehouse in inquire about this cryptographic hash, namely can obtain the character that the cryptographic hash of this character picture is corresponding, namely achieve the identification of character.Due to this kind of form, by the character set cryptographic hash training storehouse set up in advance, cryptographic hash is directly utilized to compare, character is identified, because this simplify character recognition process, the recognition speed to character can be improved, being applicable to process the huge picture information of batch.In addition, the present embodiment, in the identifying of character, also carries out the operation of monochromatic filter process, thus makes the identification of character not by the impact of color, therefore can improve speed and the accuracy of character recognition further.Relative to the recognition speed of existing OCR technology 10 characters per second, the character identifying method of the application can reach identification per second 1000 characters.
Please refer to Fig. 9, is the structural representation of the second embodiment of Character recognizer of the present invention.The structure key distinction of this embodiment and Fig. 6 embodiment is also to comprise: set up unit 10, for setting up character set cryptographic hash training storehouse.Wherein, as shown in Figure 10, a kind of structure setting up unit 10 comprises:
Extract subelement 101, for extracting the word character picture in character set.
Second filter process subelement 102, for carrying out monochromatic filter process to the monocase picture extracting subelement 101 extraction.
Second computation subunit 103, for the cryptographic hash utilizing hash algorithm to calculate the word character picture after the second filter process subelement 102 process;
Storing sub-units 104, the character code for this word character picture cryptographic hash and the second computation subunit 103 of extracting the monocase picture that subelement 101 extracts calculated is stored into character set cryptographic hash training storehouse accordingly.
Wherein character set can be ttf(trueTypeFont) character set, ttf character set is widely used character set in a kind of computing machine.An above-mentioned part owing to being the process setting up character set cryptographic hash training storehouse, the character picture therefore extracted from character set is known.
What finally illustrate is, above embodiment is only in order to illustrate technical scheme of the present invention and unrestricted, although with reference to preferred embodiment to invention has been detailed description, those of ordinary skill in the art is to be understood that, can modify to technical scheme of the present invention or equivalent replacement, and not departing from aim and the scope of technical solution of the present invention, it all should be encompassed in the middle of right of the present invention.
Claims (6)
1. a character identifying method, is characterized in that: comprise the steps:
A, set up character set cryptographic hash training storehouse;
B, hash algorithm is utilized to calculate the cryptographic hash of character picture in picture to be identified;
C, from set up in advance character set cryptographic hash training storehouse inquire about described cryptographic hash, obtain the character that described cryptographic hash is corresponding;
Wherein, described a comprises:
A1, the word character picture extracted in character set;
A2, monochromatic filter process is carried out to the word character picture of described extraction;
A3, hash algorithm is utilized to calculate the cryptographic hash of the word character picture of described extraction;
A4, the character code of the cryptographic hash of the word character picture of described extraction and described word character picture is stored into character set cryptographic hash training storehouse accordingly;
Described b comprises:
B1, from described picture to be identified, extract word character picture;
B2, monochromatic filter process is carried out to described word character picture;
B3, described hash algorithm is utilized to calculate the cryptographic hash of described word character picture;
Wherein, described character set cryptographic hash training storehouse comprises: character code and the cryptographic hash corresponding with described character code;
Described c comprises:
C1, from set up in advance character set cryptographic hash training storehouse inquire about described cryptographic hash, obtain the character code that described cryptographic hash is corresponding;
C2, the described character code obtained is converted to corresponding character;
Described hash algorithm adopts the 5th edition Message Digest 5 MD5.
2. character identifying method as claimed in claim 1, is characterized in that: described character code adopts binary coding.
3. character identifying method as claimed in claim 1, is characterized in that: in the described character set cryptographic hash training storehouse set up in advance, adopt the form store character of table coding and the cryptographic hash corresponding with described character code.
4. a Character recognizer, is characterized in that: comprising:
First processing unit, for the cryptographic hash utilizing hash algorithm to calculate character picture in picture to be identified;
Second processing unit, for inquiring about the cryptographic hash that described first processing unit calculates from the character set cryptographic hash training storehouse set up in advance, obtains the character that described cryptographic hash is corresponding;
Wherein, the first processing unit comprises:
Extract subelement, for extracting word character picture from described picture to be identified;
First filter process subelement, carries out monochromatic filter process for the word character picture extracted described extraction subelement;
First computation subunit, for the cryptographic hash utilizing hash algorithm to calculate the word character picture after described first filter process subelement process;
Wherein, described character set cryptographic hash training storehouse comprises: character code and the cryptographic hash corresponding with described character code; Described second processing unit, comprising:
Inquiry subelement, for inquiring about described cryptographic hash from the character set cryptographic hash training storehouse set up in advance, obtains the character code that described cryptographic hash is corresponding;
Conversion subelement, is converted to corresponding character for the character code inquired by described inquiry subelement;
Also comprise: set up unit, for setting up character set cryptographic hash training storehouse, described unit of setting up comprises:
Extract subelement, for extracting the word character picture in character set;
Second filter process subelement, carries out monochromatic filter process for the word character picture extracted described extraction subelement;
Second computation subunit, for the cryptographic hash utilizing hash algorithm to calculate the word character picture after described second filter process subelement process;
Storing sub-units, is stored into character set cryptographic hash training storehouse accordingly for the cryptographic hash of the word character picture by described extraction and the character code of described word character picture;
Described hash algorithm adopts the 5th edition Message Digest 5 MD5.
5. Character recognizer as claimed in claim 4, is characterized in that: described character code adopts binary coding.
6. Character recognizer as claimed in claim 5, is characterized in that: in the described character set cryptographic hash training storehouse set up in advance, adopt the form store character of table coding and the cryptographic hash corresponding with described character code.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210374937.6A CN102880874B (en) | 2012-09-29 | 2012-09-29 | Character identifying method and Character recognizer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210374937.6A CN102880874B (en) | 2012-09-29 | 2012-09-29 | Character identifying method and Character recognizer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102880874A CN102880874A (en) | 2013-01-16 |
| CN102880874B true CN102880874B (en) | 2016-04-13 |
Family
ID=47482191
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210374937.6A Active CN102880874B (en) | 2012-09-29 | 2012-09-29 | Character identifying method and Character recognizer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102880874B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107045612A (en) * | 2017-01-19 | 2017-08-15 | 安徽数洋科技有限公司 | A kind of encryption method of web data |
| CN109766893A (en) * | 2019-01-09 | 2019-05-17 | 北京数衍科技有限公司 | Picture character recognition methods suitable for receipt of doing shopping |
| CN111091128B (en) * | 2019-12-18 | 2023-09-22 | 北京数衍科技有限公司 | Character picture classification method and device and electronic equipment |
| CN113516044A (en) * | 2021-05-18 | 2021-10-19 | 山东新一代信息产业技术研究院有限公司 | Paper contract credit enhancement method and system based on OCR and Hash algorithm |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2166248C (en) * | 1995-12-28 | 2000-01-04 | Abdel Naser Al-Karmi | Optical character recognition of handwritten or cursive text |
| CN101364267A (en) * | 2007-08-09 | 2009-02-11 | 富士通株式会社 | Character recognition apparatus, character recognition method, and computer product |
| CN102364496A (en) * | 2011-11-24 | 2012-02-29 | 无锡慧眼电子科技有限公司 | Method and system for identifying automobile license plates automatically based on image analysis |
-
2012
- 2012-09-29 CN CN201210374937.6A patent/CN102880874B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2166248C (en) * | 1995-12-28 | 2000-01-04 | Abdel Naser Al-Karmi | Optical character recognition of handwritten or cursive text |
| CN101364267A (en) * | 2007-08-09 | 2009-02-11 | 富士通株式会社 | Character recognition apparatus, character recognition method, and computer product |
| CN102364496A (en) * | 2011-11-24 | 2012-02-29 | 无锡慧眼电子科技有限公司 | Method and system for identifying automobile license plates automatically based on image analysis |
Non-Patent Citations (1)
| Title |
|---|
| 基于图像匹配的汉字识别系统研究与实现;武桐;《中国硕士论文全文数据库》;20100726;第4-12，22-33页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102880874A (en) | 2013-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104732228B (en) | A kind of detection of PDF document mess code, the method for correction | |
| CN102880874B (en) | Character identifying method and Character recognizer | |
| CN111444705A (en) | Error correction method, device, equipment and readable storage medium | |
| CN104750666B (en) | A kind of recognition methods of text character codes mode and system | |
| CN103123618A (en) | Text similarity obtaining method and device | |
| CN110895651A (en) | Address standardization processing method, device, equipment and computer readable storage medium | |
| CN101996160A (en) | Method and system for processing script data | |
| CN103514238A (en) | Sensitive word recognition processing method based on classification searching | |
| CN106227808B (en) | A kind of method and method for judging rubbish mail removing mail interference information | |
| CN107220333B (en) | character search method based on Sunday algorithm | |
| JP2014197844A (en) | Encoder for encoding text on matrix code symbol and decoder for decoding matrix code symbol | |
| CN103425931A (en) | Abnormal web script detection method and system | |
| CN101013420A (en) | Method for identifying coding form of Chinese text | |
| CN104360988A (en) | Method and device for identifying coding mode of Chinese characters | |
| WO2018041036A1 (en) | Keyword searching method, apparatus and terminal | |
| CN106777061B (en) | Information hiding system and method based on webpage text and image and extraction method | |
| CN103136166B (en) | Method and device for font determination | |
| CN104933030A (en) | Uygur language spelling examination method and device | |
| CN108536693A (en) | A kind of filtering sensitive words method, apparatus, electronic equipment, storage medium | |
| WO2018226221A1 (en) | Context-dependent shared dictionaries | |
| CN102253983A (en) | Method and system for identifying Chinese high-risk words | |
| CN108108267B (en) | Data recovery method and device | |
| US9619670B1 (en) | Detecting user credentials from inputted data | |
| CN101741392A (en) | Huffman decoding method for fast resolving code length | |
| CN110472202B (en) | Unicode-based information embedding and extracting method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180726 Address after: 230088 room 405-5, R & D center of China (Hefei) International Intelligent Speech Industrial Park, 3333, hi tech Road, Hefei, Anhui. Patentee after: Anhui Tai Yue Xiang Sheng Software Co., Ltd. Address before: 401121 3, 1 floor, office building, south wing of mercury science and technology building, 5 new Mount Huangshan Road, North New District, Chongqing. Patentee before: Chongqing Xinmei Agricultural Information Technology Co.,Ltd. |