CN102915437A - Text information identification method and system - Google Patents
Text information identification method and system Download PDFInfo
- Publication number
- CN102915437A CN102915437A CN2011102199124A CN201110219912A CN102915437A CN 102915437 A CN102915437 A CN 102915437A CN 2011102199124 A CN2011102199124 A CN 2011102199124A CN 201110219912 A CN201110219912 A CN 201110219912A CN 102915437 A CN102915437 A CN 102915437A
- Authority
- CN
- China
- Prior art keywords
- character
- image
- text message
- cloud server
- server
- 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
Images
Abstract
The invention relates to a text information identification method and a system. The method comprises the following steps: a client acquires an image containing text information, and sends the image to a cloud server; the cloud server receives the image, processes the image, and extracts characters of the text information in the image; the cloud server processes the characters and acquires the characteristics of the characters; according to the characteristics of the characters, the cloud server queries a characteristic library set on the cloud server and performs characteristic matching with the characters in the characteristic library, so as to identify the characters, and further identify the text information; and the cloud server sends the identified text information to the client. In the invention, the client uploads the image to the cloud server, so that the identification process is performed on the cloud server. As the cloud server has strong computing power and expansion capability, the performance of the cloud server can meet the requirements of the characteristic library, the characteristic library and the identification capability are not limited by a user computer; the text information can be accurately identified; and the text information identification method and system is simple and efficient, and the identification rate is greatly improved.
Description
[technical field]
The present invention relates to a kind of information processing technology, relate in particular to a kind of text message recognition methods and system.
[background technology]
At present, the text message on paper document or the picture can not directly use, and needs in use manual input just can.For substituting manually input, usually adopt OCR (Optical Character Recognition optical character identification) technology that text message is identified.
But, traditional OCR technology, the user need to install a huge client software, and the computer hardware that requires to identify possesses enough handling properties in use.What the OCR technology was mainly faced is paper material, and the identification scene need to be considered a lot of problems, so discrimination can be subject to the restriction of complicated factor.The core technology index of discrimination is feature database.Because user computer hardware and processor performance do not possess enough requirements usually, recognition capability and feature database all are subject to the restriction of subscriber computer performance, greatly reduce the OCR technology to the discrimination of text message, can accurately not identify text message.
Simultaneously, after to text message identification, also need to carry out error correction.Because the ability of error correction depends on the quantity of information of feature database, feature database is subject to the restriction of the machine performance, thereby has greatly limited the ability of error correction, so that discrimination further reduces.
[summary of the invention]
In view of this, be necessary the text message recognition methods that provides a kind of discrimination high.
In addition, provide a kind of discrimination high text message recognition system.
A kind of text message recognition methods comprises the steps:
Client is obtained the image that comprises text message, and described image is sent to Cloud Server;
Described Cloud Server receives described image, and described image is processed, and extracts the character of described image Chinese version information;
Described Cloud Server is processed described character, obtains the feature of character;
Described Cloud Server is according to the feature of described character, and inquiry being arranged on feature database on the described Cloud Server, carries out characteristic matching with character in the feature database, character is identified, and then the identification text message;
Described Cloud Server is sent to client with the text message of identification.
A kind of text message recognition system comprises client and Cloud Server,
Described client is used for obtaining the image that comprises text message, and described image is sent to described Cloud Server;
Described Cloud Server comprises:
Transmitting/receiving server is used for receiving described image;
Image processing server is used for described image is processed, and extracts the character of described image Chinese version information;
The character processing server is used for described character is processed, and obtains character feature;
The feature database server, according to the feature of described character, inquiry being arranged on feature database on the feature database server, carries out characteristic matching with character in the feature database, character is identified, and then the identification text message; The feature database server is transferred to transmitting/receiving server with the text message of identification, and transmitting/receiving server is sent to described client with the text message of identification.
Above-mentioned text message recognition methods and system, client with image uploading to Cloud Server, identifying and Cloud Server all carry out at Cloud Server, Cloud Server has powerful computing power and extended capability, performance can satisfy the requirement of feature database, so that feature database and recognition capability are not subjected to the restriction of subscriber computer, thereby can identify text message accurately, simple, efficient, discrimination improves greatly.The user only need get final product by the client upload image, and Cloud Server just can be simultaneously for mass users provides service, and is greatly convenient for users.
[description of drawings]
Fig. 1 is the process flow diagram of an embodiment Chinese version information identifying method;
Fig. 2 is that Cloud Server is processed image among the embodiment, extracts the method flow diagram of the character of image Chinese version information;
Fig. 3 is the structural representation of an embodiment Chinese version information identification system;
Fig. 4 is the structural representation of image processing server among the embodiment.
[embodiment]
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail.
Fig. 1 is the process flow diagram of an embodiment Chinese version information identifying method.The method comprises:
S10: client is obtained the image that comprises text message, and image is sent to Cloud Server.
The object that the method is identified is the image with text message, and the text message in the image is identified.The image with text message that client is obtained perhaps is direct image for by papery or other media documents with text message scanned acquisition, also can be snapshot picture of screen printing content etc.In preferred embodiment, the image with text message that client is obtained is the snapshot picture that Instant Messenger (IM) software screen printing content obtains, text message in the sectional drawing image is identified, text message can directly be used, need not the text message in the sectional drawing image is manually inputted.The mode that client is uploaded by browser arrives Cloud Server with image uploading.
S20: Cloud Server receives image, and image is processed, and extracts the character of image Chinese version information.
Text message is comprised of a plurality of characters, and the identification text message need to extract each character of text message.Cloud Server can be cloud computing platform, also can be computational grid or a plurality of server that comprises a plurality of computing nodes.Cloud Server has powerful extended capability, huge computing power and mass memory ability, can receive simultaneously the image that a large amount of clients transmit, and provides service for mass users simultaneously.
Fig. 2 is that Cloud Server is processed image among the embodiment, extracts the method flow diagram of the character of image Chinese version information.Among this embodiment, Cloud Server receives image, and image is processed, and the step that extracts each character of image Chinese version information specifically comprises:
S21: image is carried out binary conversion treatment to set brightness value as standard, image is become black white image.
Usually, image is colored, has multiple color, the character color of text message mostly is the darker color of brightness value, and for the benefit of each character with the text message in the image extracts, and image need to be carried out binary conversion treatment, image is become black white image, character color is become black.Detailed process is: Cloud Server with the colour brightness value in the image greater than the white that is converted to of setting brightness value, otherwise be converted to black.Setting brightness value can adjust as required.
But because under some situations, it is black that there is background in image, text message is the situation of white, the i.e. situation of black matrix wrongly written or mispronounced character.For avoiding this situation to affect the identification of text message, further, this step comprises that also Cloud Server judges the image background look, be black, text message for the image transitions of white is that background is that white, text message are the step of the image of black with background, the image transitions that is about to the black matrix wrongly written or mispronounced character is the image of white gravoply, with black engraved characters.
S22: black white image contiguous pixels zone is scanned, obtained character zone.
In whole black white image, be not All Ranges all be character, may have the zone of non-character, this just need to remove the zone of non-character, only obtain character zone.
Among this embodiment, black white image contiguous pixels zone is scanned, the step of obtaining character zone is specially: the continuity of scanning black-white image black pixel, remove non-character zone according to the continuity of black pixel, and obtain character zone.
Because the pixel of character has certain continuity, and larger continuous blocks and less continuous blocks are not characters, thereby can remove non-character zone according to the continuity of black pixel, obtain character zone.Simultaneously, according to the feature of character own, such as the distribution density of pixel, rules degree, size etc., also can further remove non-character zone.
S23: to break ranks operation of character zone, character is extracted.
Consider that character zone mostly is arranging according to ranks of rule, therefore according to the ranks feature of rule, the character zone ranks that break are operated, single character is split off, thereby each character is extracted.
Among this embodiment, to break ranks operation of character zone, the step that character is extracted is specially: at first character zone is carried out every trade and cut apart, again to the ranks column split of whenever advancing, separate single character, each character is extracted.
In addition, required form for guaranteeing that picture format is identified by Cloud Server, Cloud Server receives image, image is processed, the step that extracts each character of image Chinese version information also comprises: the form of detected image, if the form of image is not for requiring form, then the format conversion with image is the step that requires form.In preferred embodiment, requiring form is the BMP form.
S30: Cloud Server is processed character, obtains the feature of character.
Among this embodiment, the size that is characterized as character of character and the quantity of character pixels point.Because there is the difference of font size in a plurality of characters, there is again the difference of runic and light face type in the character of identical font size.For ease of identification character, reduce workload, Cloud Server need to be processed each character, concrete disposal route is: character is carried out refinement, extract the skeleton of each character, obtain the pixel of character, the skeleton that extracts character namely is to represent this character with minimum pixel; Each character is all zoomed to the setting size, obtain the size of character.
S40: Cloud Server is according to the feature of character, and inquiry is arranged on the feature database on the Cloud Server, carries out characteristic matching with character in the feature database, character is identified, and then the identification text message.
Feature database establishes in advance, and is arranged on the Cloud Server.Comprise characters all in the character set in the feature database, also comprised the multiple variation of each character.The for example variation of font: the Song typeface, regular script etc.; Also has the variation of vector: such as italic etc.Also has conversion of font size etc.Because feature database is arranged on the Cloud Server, Cloud Server has powerful extended capability, huge computing power and mass memory ability, performance can satisfy the requirement of feature database, feature database can store and mate the required data of identification, thereby guarantees that each character can both identify accurately.
Among this embodiment, Cloud Server is according to the feature of character, inquiry being arranged on feature database on the Cloud Server, carries out characteristic matching with character in the feature database, and character is identified, and then the step of identification text message is specially: Cloud Server is according to size and the pixel of each character, search size and the pixel of character in the feature database, mate, determine the coded message that pixel is corresponding, identify each character, thereby identify text message.In preferred embodiment, Cloud Server carries out characteristic matching to character on a plurality of servers, and the identification text message improves recognition efficiency and discrimination.
S50: Cloud Server is sent to client with the text message of identification.
Text message is after identification, and Cloud Server is sent to client with text message, for the user directly, need not manual input.
Because when text message is identified, some character on the image can cause identification to make mistakes owing to the reason such as fuzzy when identification, is further accurate identification text message, guarantee discrimination, the method also comprises the step that Cloud Server carries out error correction to the text message of identifying.Be specially: Cloud Server mates the phrase in the text message with the phrase of storing in the feature database, carry out error correction, is sent to client after the error correction.
Feature database is arranged on the Cloud Server, can store the phrase of magnanimity.When error correction, utilize the ways customary of phrase, mate with the magnanimity phrase of storing in the feature database, can judge whether phrase is correct in the text message, correct the mistake in the text message, improve the accuracy rate of identification.For example, comprise " sun " this word in the text message, this point in " too " word is identified as " greatly " word owing to smudgy when identification.During error correction, find that " sun " is only correct phrase, but not " large sun " is corrected as " too " word with " greatly " word, thereby " large sun " corrected and come to be " sun ".
In addition, also provide a kind of text message recognition system.
Shown in Figure 3 is the structural representation of an embodiment Chinese version information identification system.Text information identification system comprises: client 100 and Cloud Server 200.
The object that this system identifies is the image with text message, and the text message in the image is identified.The image with text message that client 100 is obtained perhaps is direct image for by papery or other media documents with text message scanned acquisition, also can be snapshot picture of screen printing content etc.In preferred embodiment, the image with text message that client 100 is obtained is the snapshot picture that Instant Messenger (IM) software screen printing content obtains, text message in the sectional drawing image is identified, text message can directly be used, need not the text message in the sectional drawing image is manually inputted.The mode that client 100 is uploaded by browser arrives Cloud Server 200 with image uploading.
Cloud Server 200 can be cloud computing platform, also can be computational grid or a plurality of server that comprises a plurality of computing nodes.Cloud Server 200 has powerful extended capability, huge computing power and mass memory ability, can receive simultaneously the image that a large amount of clients transmit, and provides service for mass users simultaneously.
Among this embodiment, Cloud Server 200 comprises: transmitting/receiving server 210, image processing server 220, character processing server 230, feature database server 240 and error correction server 250.
Transmitting/receiving server 210 is used for reception and has the image of text message, and meets at image processing server 220.Among this embodiment, transmitting/receiving server 210 is HTTP (HTML (Hypertext Markup Language)) server.Simultaneously, transmitting/receiving server 210 is the form of detected image also, if the form of image is not for requiring form, then with the format conversion of image for requiring form.In preferred embodiment, requiring form is the BMP form.Transmitting/receiving server 210 can receive the image that a plurality of clients 100 send simultaneously.
220 pairs of images of image processing server are processed, and extract the character of image Chinese version information.
Text message is comprised of a plurality of characters, and the identification text message need to extract each character of text message to be identified.
Fig. 4 is the structural representation of image processing server among the embodiment.Among this embodiment, image processing server 220 comprises binarization block 221, character zone acquisition module 222 and character extraction module 223.
Usually, image is colored, has multiple color, the character color of text message mostly is the darker color of brightness value, and for the benefit of each character with the text message in the image extracts, and image need to be carried out binary conversion treatment, image is become black white image, character color is become black.221 pairs of images of binarization block carry out binary conversion treatment, with the be converted to white of the colour brightness value in the image greater than the setting brightness value, otherwise are converted to black.Setting brightness value can adjust as required.
But because under some situations, it is black that there is background in image, character color is the situation of white, the i.e. situation of black matrix wrongly written or mispronounced character.For avoiding this situation to affect the identification of text message, further, binarization block 221 is also judged the image background look, be black, text message for the image transitions of white is that background is that white, text message are the image of black with background, the image transitions that is about to the black matrix wrongly written or mispronounced character is the image of white gravoply, with black engraved characters.
Character zone acquisition module 222 is used for black white image contiguous pixels zone is scanned, and obtains character zone.
In whole black white image, be not All Ranges all be character, may have the zone of non-character, this just need to remove the zone of non-character, only obtain character zone.
Among this embodiment, the continuity of character zone acquisition module 222 scanning black-white image black pixels is removed non-character zone according to the continuity of black pixel, obtains character zone.
Because the pixel of character has certain continuity, and larger continuous blocks and less continuous blocks are not characters, thereby so that character zone acquisition module 222 can be removed non-character zone according to the continuity of black pixel, obtain character zone.Simultaneously, character zone acquisition module 222 is according to the feature of character own, such as the distribution density of pixel, rules degree, size etc., also can further remove non-character zone.
Consider that character zone mostly is arranging according to ranks of rule, therefore according to the ranks feature of rule, the character zone ranks that break are operated, single character can be split off, thereby each character is extracted.
Among this embodiment, character extraction module 223 at first carries out every trade to character zone to be cut apart, and again to the ranks column split of whenever advancing, separates single character, and each character is extracted.
Among this embodiment, the size that is characterized as character of character and the quantity of character pixels point.Because there is the difference of font size in a plurality of characters, there is again the difference of runic and light face type in the character of identical font size.For ease of identification character, reduce workload, need to process each character.
230 pairs of characters of character processing server carry out refinement, extract the skeleton of character, obtain the pixel of character.The skeleton that extracts character namely is to represent this character with minimum pixel.Character processing server 230 all zooms to character and sets size, obtains the size of character.
Among this embodiment, feature database server 240 is searched size and the pixel of character in the feature database 241 according to size and the pixel of each character, mates, and determines the coded message that pixel is corresponding, identifies each character, thereby identifies text message.In preferred embodiment, feature database server 240 is for having the server cluster of a plurality of servers, and feature database server 240 carries out characteristic matching to character on a plurality of servers, and the identification text message improves recognition efficiency and discrimination.
The text message of 250 pairs of identifications of error correction server carries out error correction.
Because when text message identified, some character on the image can cause identification to make mistakes when identification, so also need to carry out error correction owing to the reason such as fuzzy.Error correction server 250 mates the phrase in the text message with the phrase of storing in the feature database 241, carry out error correction.
Transmitting/receiving server 210 is sent to client 100 with the text message of identification.
Text message is after identification, and transmitting/receiving server 210 is sent to client 100 with text message, for the user directly, need not manual input.Transmitting/receiving server 210 can send a plurality of clients 100 simultaneously.
Above-mentioned text message recognition methods and system, client with image uploading to Cloud Server, identifying and Cloud Server all carry out at Cloud Server, Cloud Server has powerful computing power and extended capability, performance can satisfy the requirement of feature database, so that feature database and recognition capability are not subjected to the restriction of subscriber computer, thereby can identify text message accurately, simple, efficient, discrimination improves greatly.The user only need get final product by the client upload image, and Cloud Server just can be simultaneously for mass users provides service, and is greatly convenient for users.
The above embodiment has only expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but can not therefore be interpreted as the restriction to claim of the present invention.Should be pointed out that for the person of ordinary skill of the art without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (10)
1. a text message recognition methods comprises the steps:
Client is obtained the image that comprises text message, and described image is sent to Cloud Server;
Described Cloud Server receives described image, and described image is processed, and extracts the character of described image Chinese version information;
Described Cloud Server is processed described character, obtains the feature of character;
Described Cloud Server is according to the feature of described character, and inquiry being arranged on feature database on the described Cloud Server, carries out characteristic matching with character in the feature database, character is identified, and then the identification text message;
Described Cloud Server is sent to client with the text message of identification.
2. text message recognition methods according to claim 1 is characterized in that, described Cloud Server receives described image, and described image is processed, and the step that extracts the character in the described image is specially:
Image is carried out binary conversion treatment to set brightness value as standard, image is become black white image;
Black white image contiguous pixels zone is scanned, obtained character zone;
To break ranks operation of character zone, each character is extracted.
3. text message recognition methods according to claim 2, it is characterized in that, described Cloud Server receives described image, described image is processed, the step that extracts the character in the described image also comprises: the image background look is judged, be black, text message for the image transitions of white is that background is white with background, text message is the step of the image of black.
4. text message recognition methods according to claim 1 is characterized in that, the size that is characterized as character and the pixel of described character; Described Cloud Server is processed described character, and the step of obtaining character feature is specially:
Character is carried out refinement, extract the skeleton of character, obtain the pixel of character;
Character scale to setting size, is obtained the size of character.
5. text message recognition methods according to claim 1, it is characterized in that, described method also comprises the step that described Cloud Server carries out error correction to the text message of identifying, be specially: described Cloud Server is with the phrase in the text message of identification, mate with the phrase of storing in the feature database, carry out error correction, be sent to client after the error correction.
6. a text message recognition system is characterized in that, comprises client and Cloud Server,
Described client is used for obtaining the image that comprises text message, and described image is sent to described Cloud Server;
Described Cloud Server comprises:
Transmitting/receiving server is used for receiving described image;
Image processing server is used for described image is processed, and extracts the character of described image Chinese version information;
The character processing server is used for described character is processed, and obtains character feature;
The feature database server, according to the feature of described character, inquiry being arranged on feature database on the feature database server, carries out characteristic matching with character in the feature database, character is identified, and then the identification text message; The feature database server is transferred to transmitting/receiving server with the text message of identification, and transmitting/receiving server is sent to described client with the text message of identification.
7. text message recognition system according to claim 6 is characterized in that, described image processing server comprises:
Binarization block is used for image is carried out binary conversion treatment to set brightness value as standard, and image is become black white image;
The character zone acquisition module is used for black white image contiguous pixels zone is scanned, and obtains character zone;
The character extraction module for ranks operation that character zone is broken, extracts each character.
8. text message recognition system according to claim 7, it is characterized in that, described binarization block also is used for the image background look is judged, is that black, text message are that background is that white, text message are the image of black for the image transitions of white with background.
9. text message recognition system according to claim 6 is characterized in that, the size that is characterized as character and the pixel of described character; Described character processing server is used for character is carried out refinement, extracts the skeleton of character, obtains the pixel of character, and described character processing server to setting size, obtains the size of character with character scale.
10. text message recognition system according to claim 6 is characterized in that, described Cloud Server also comprises the error correction server that the text message of identifying is carried out error correction; The phrase that described error correction server is used for storing in the phrase of the text message of identification and the feature database mates, and carries out error correction, transfers to transmitting/receiving server after the error correction and is sent to described client.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102199124A CN102915437A (en) | 2011-08-02 | 2011-08-02 | Text information identification method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102199124A CN102915437A (en) | 2011-08-02 | 2011-08-02 | Text information identification method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102915437A true CN102915437A (en) | 2013-02-06 |
Family
ID=47613798
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2011102199124A Pending CN102915437A (en) | 2011-08-02 | 2011-08-02 | Text information identification method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102915437A (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103248705A (en) * | 2013-05-20 | 2013-08-14 | 北京智谷睿拓技术服务有限公司 | Server, client and video treatment method |
CN103279754A (en) * | 2013-06-25 | 2013-09-04 | 觅林网络科技(上海)有限公司 | Business card cloud identification method and system |
CN104090878A (en) * | 2013-07-04 | 2014-10-08 | 腾讯科技(深圳)有限公司 | Multimedia checking method, terminal, server and system |
CN104200204A (en) * | 2014-09-02 | 2014-12-10 | 福建富士通信息软件有限公司 | Picture processing device and method |
CN104240068A (en) * | 2014-08-25 | 2014-12-24 | 小米科技有限责任公司 | Method and device for creating reminding event |
CN104598902A (en) * | 2015-01-29 | 2015-05-06 | 百度在线网络技术(北京)有限公司 | Method and device for identifying screenshot and browser |
CN104933429A (en) * | 2015-06-01 | 2015-09-23 | 深圳市诺比邻科技有限公司 | Method and device for extracting information from image |
CN105335163A (en) * | 2015-11-30 | 2016-02-17 | 上海斐讯数据通信技术有限公司 | Software code reading method and system |
CN105718855A (en) * | 2015-12-03 | 2016-06-29 | 王晓龙 | Online composition assessment method and system |
CN106412008A (en) * | 2016-08-26 | 2017-02-15 | 乐视控股(北京)有限公司 | Identifier correcting method and device |
CN107277602A (en) * | 2017-07-26 | 2017-10-20 | 联想(北京)有限公司 | Information acquisition method and electronic equipment |
CN107451582A (en) * | 2017-07-13 | 2017-12-08 | 安徽声讯信息技术有限公司 | A kind of graphics context identifying system and its recognition methods |
CN110032503A (en) * | 2018-11-05 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Data processing system, method, equipment and device based on UI automation and OCR |
CN110222193A (en) * | 2019-05-21 | 2019-09-10 | 深圳壹账通智能科技有限公司 | Scan text modification method, device, computer equipment and storage medium |
CN110647878A (en) * | 2019-08-05 | 2020-01-03 | 紫光西部数据(南京)有限公司 | Data processing method based on screen shot picture |
CN113065537A (en) * | 2021-06-03 | 2021-07-02 | 江苏联著实业股份有限公司 | OCR file format conversion method and system based on model optimization |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1632820A (en) * | 2004-12-30 | 2005-06-29 | 北京中星微电子有限公司 | Method for deciding background color according to area in optical character recognition of mobile terminal |
CN101782899A (en) * | 2009-01-19 | 2010-07-21 | 李茂武 | Central translation platform |
CN101807241A (en) * | 2010-03-17 | 2010-08-18 | 四川创立信息科技有限责任公司 | Cloud computing-based mobile terminal barcode recognition method |
CN101976148A (en) * | 2010-10-28 | 2011-02-16 | 广东开心信息技术有限公司 | Hand input system and method |
CN102122360A (en) * | 2011-03-01 | 2011-07-13 | 华南理工大学 | Cloud computing-based mobile terminal handwriting identification method |
-
2011
- 2011-08-02 CN CN2011102199124A patent/CN102915437A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1632820A (en) * | 2004-12-30 | 2005-06-29 | 北京中星微电子有限公司 | Method for deciding background color according to area in optical character recognition of mobile terminal |
CN101782899A (en) * | 2009-01-19 | 2010-07-21 | 李茂武 | Central translation platform |
CN101807241A (en) * | 2010-03-17 | 2010-08-18 | 四川创立信息科技有限责任公司 | Cloud computing-based mobile terminal barcode recognition method |
CN101976148A (en) * | 2010-10-28 | 2011-02-16 | 广东开心信息技术有限公司 | Hand input system and method |
CN102122360A (en) * | 2011-03-01 | 2011-07-13 | 华南理工大学 | Cloud computing-based mobile terminal handwriting identification method |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103248705A (en) * | 2013-05-20 | 2013-08-14 | 北京智谷睿拓技术服务有限公司 | Server, client and video treatment method |
CN103279754A (en) * | 2013-06-25 | 2013-09-04 | 觅林网络科技(上海)有限公司 | Business card cloud identification method and system |
CN104090878A (en) * | 2013-07-04 | 2014-10-08 | 腾讯科技(深圳)有限公司 | Multimedia checking method, terminal, server and system |
WO2015000433A1 (en) * | 2013-07-04 | 2015-01-08 | 腾讯科技(深圳)有限公司 | Multimedia search method, terminal, server and system |
CN104090878B (en) * | 2013-07-04 | 2017-09-05 | 腾讯科技(深圳)有限公司 | A kind of multimedia lookup method, terminal, server and system |
CN104240068A (en) * | 2014-08-25 | 2014-12-24 | 小米科技有限责任公司 | Method and device for creating reminding event |
CN104200204B (en) * | 2014-09-02 | 2017-10-03 | 福建富士通信息软件有限公司 | A kind of picture processing device and method |
CN104200204A (en) * | 2014-09-02 | 2014-12-10 | 福建富士通信息软件有限公司 | Picture processing device and method |
CN104598902A (en) * | 2015-01-29 | 2015-05-06 | 百度在线网络技术(北京)有限公司 | Method and device for identifying screenshot and browser |
CN104933429A (en) * | 2015-06-01 | 2015-09-23 | 深圳市诺比邻科技有限公司 | Method and device for extracting information from image |
CN105335163A (en) * | 2015-11-30 | 2016-02-17 | 上海斐讯数据通信技术有限公司 | Software code reading method and system |
CN105718855A (en) * | 2015-12-03 | 2016-06-29 | 王晓龙 | Online composition assessment method and system |
CN106412008A (en) * | 2016-08-26 | 2017-02-15 | 乐视控股(北京)有限公司 | Identifier correcting method and device |
CN107451582A (en) * | 2017-07-13 | 2017-12-08 | 安徽声讯信息技术有限公司 | A kind of graphics context identifying system and its recognition methods |
CN107277602A (en) * | 2017-07-26 | 2017-10-20 | 联想(北京)有限公司 | Information acquisition method and electronic equipment |
CN107277602B (en) * | 2017-07-26 | 2020-05-26 | 联想(北京)有限公司 | Information acquisition method and electronic equipment |
CN110032503A (en) * | 2018-11-05 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Data processing system, method, equipment and device based on UI automation and OCR |
CN110222193A (en) * | 2019-05-21 | 2019-09-10 | 深圳壹账通智能科技有限公司 | Scan text modification method, device, computer equipment and storage medium |
CN110647878A (en) * | 2019-08-05 | 2020-01-03 | 紫光西部数据(南京)有限公司 | Data processing method based on screen shot picture |
CN113065537A (en) * | 2021-06-03 | 2021-07-02 | 江苏联著实业股份有限公司 | OCR file format conversion method and system based on model optimization |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102915437A (en) | Text information identification method and system | |
US8355578B2 (en) | Image processing apparatus, image processing method, and storage medium | |
EP1588293B1 (en) | Image processing method, system, program, program storage medium and information processing apparatus | |
US20060221357A1 (en) | Information processing apparatus and method | |
CN103065146A (en) | Character recognition method for power communication machine room dumb equipment signboards | |
US20050286805A1 (en) | Image processing apparatus, control method therefor, and program | |
CN101324883A (en) | Method for extracting variation key word | |
CN103577818A (en) | Method and device for recognizing image characters | |
US8514462B2 (en) | Processing document image including caption region | |
CN110765740B (en) | Full-type text replacement method, system, device and storage medium based on DOM tree | |
JP2005352696A (en) | Image processing device, control method thereof, and program | |
Isheawy et al. | Optical character recognition (ocr) system | |
CN104603833A (en) | A method and system for linking printed objects with electronic content | |
CN111368511A (en) | PDF document analysis method and device | |
US8195626B1 (en) | Compressing token-based files for transfer and reconstruction | |
CN201222256Y (en) | Digitalization integration processing archive system | |
CN101751512A (en) | Recipe management system applied to communication device and method | |
CN102682457A (en) | Rearrangement method for performing adaptive screen reading on print media image | |
CN103455786A (en) | Image recognition method and system | |
CN113780276A (en) | Text detection and identification method and system combined with text classification | |
CN115630636A (en) | Text recognition method and device | |
JP6091552B2 (en) | Movie processing apparatus and movie processing system | |
CN114677700A (en) | Identification method and device of identity, storage medium and electronic equipment | |
CN113344096A (en) | Automatic bid document analysis method and system based on OCR technology | |
CN114359913A (en) | Text label determination method and related device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130206 |
|
RJ01 | Rejection of invention patent application after publication |