CN101059840A - Words input method using mobile phone shooting style - Google Patents
Words input method using mobile phone shooting style Download PDFInfo
- Publication number
- CN101059840A CN101059840A CN 200710074531 CN200710074531A CN101059840A CN 101059840 A CN101059840 A CN 101059840A CN 200710074531 CN200710074531 CN 200710074531 CN 200710074531 A CN200710074531 A CN 200710074531A CN 101059840 A CN101059840 A CN 101059840A
- Authority
- CN
- China
- Prior art keywords
- character
- mobile phone
- photo
- literal
- input method
- 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
Landscapes
- Telephone Function (AREA)
Abstract
The invention discloses a character input method via mobile phone camera method, which uses general camera mobile phone to shoot a target character picture and uses the software of the mobile phone to recognize the character in the picture into text character. And the method comprises that using mobile phone to shoot a target character material into a plurality of pictures, a character recognize module analyzes each picture, to draw out each character block on the picture according to the character profile, analyzing the strokes in the character, selecting similar characters from a character group automatically or hand-operated, when right character is not in a candidate character list, user inputs the character via hand. The invention is based on the basic hardware of present mobile phone, which can be realized by developing special software, with low cost and high practicality. The inventive method can be processed by general low-resolution mobile phone, with which user can freely collect information via mobile phone.
Description
[technical field]
The present invention relates to cellphone information input field, relate in particular to a kind of shooting style method for inputting characters with great of on shooting mobile phone, using.
[background technology]
Nowadays mobile phone becomes the information processing instrument of modern's indispensability, but input characters is a difficult problem on mobile phone, also has a huge market.Especially the fast development of note has more stimulated the fast development of mobile phone literal input technology.In cellphone inputting method, generally can be divided into keyboard input and handwriting input two big classes, these input methods generally are used for the input method of user when create message or address list.Also sometimes, Word message on other existing medium that the user needs to see is at any time noted, as newspaper, outdoor advertising, business card etc., but word for word import with hand-written or keyboard, may be inconvenient, as there isn't enough time, or crowded out of doors, or can't import by the operating handset keyboard by other restrictions, most convenient is filmed target text exactly with mobile phone at this moment, store as photo earlier, check later on.Sometimes the text conversion in the photo that also needs to take is an alphabetic character, is convenient in other software, use or send short messages to other people, as the business card of taking, need literal is wherein extracted just and can join in the address list.But it is manual still inconvenient according to typing again, especially when being inconvenient to use computer, on mobile phone while seeing the photo typing, efficient is very low, and many mobile phones do not provide multitask function yet, can only select one as photo function of browse and editing address list function, use one of them will close another down, like this, the user just can't realize seeing photo typing literal on mobile phone.
[summary of the invention]
The objective of the invention is to overcome the defective that present mobile phone exists in the ready-made Word message of typing, a kind of character input method with the mobile phone photograph mode is proposed, with common shooting mobile phone photographic subjects literal photo, by software built-in in the mobile phone literal in the photo is identified as text character again, realize this method, comprise an optical photographing module and a photo literal identification module in the mobile phone, the character set that comprises GB2312 in the literal identification module, system sets up a concordance list to the character in the character set from high to low according to its usage frequency in social public publication, and the concrete steps of this method are as follows:
Steps A: with mobile phone the target text material being taken is multiple pictures, when taking, and the distance that adjusting need be taken pictures, make the literal of shooting can be clear substantially, otherwise mobile phone just is difficult to identification, in order to reach shooting effect preferably, on mobile phone the microshot mode can be set;
Step B: mobile phone literal identification module is analyzed one by one to every photo, delineate out each character zone scope on the photo according to character outline, it is character block, here mainly be that the border of each character that will discern is delineated out earlier, in general be that a character is a square, comprised all strokes of character;
Step C: the literal identification module is analyzed one by one to each character block in every photo, analyze the stroke lines of character in the character block, and from character set, select a plurality of close characters, arrange from high to low by similarity degree, and select text character after first the highest character of similarity degree is defaulted as the identification of character block correspondence automatically, and this is the committed step of character recognition, this technology is comparative maturity also, be widely used in the literal identification OCR software, as the OCR of Qing Hua Ziguang, minister OCR etc.;
Step D: system is given tacit consent to automatically the character of selecting mistake, the user can select in other candidate characters by hand, if correct character is not in the candidate tabulation, import by hand by the user, be subjected to the influence of sharpness of target text own and shooting quality, sometimes handset identity get not necessarily accurate, need user's manual process, as the stroke on taking pictures lack or when not knowing, the user can replenish by handwriting mode and lack or unclear stroke, in addition, the candidate characters of user's manual setting, first candidate characters when next time discerning similar character.
Step e: the character that affirmation was used during to the identification input, system increases a unit with its usage frequency automatically, this step mainly is the learning functionality of increase system, mobile phone is when dispatching from the factory, the character usage frequency concordance list of installing is the general purpose table of a linguistics mechanism statistics, but the user in actual use, hobby because of the user, the actual usage frequency of character can be variant in the data that occupational characteristic etc. are collected, if therefore system progressively adapts to user's characteristics, the user uses the handset identity input just more and more handy, and here the usage frequency by automatic adjustment character set just can achieve the goal.
Further improvement in the technical proposal is: shooting mobile phone can be regulated brightness of image automatically when taking the literal photo, makes it reach literal identification predetermined value, and this predetermined value has default value when mobile phone dispatches from the factory, and the user can also revise.
Further improvement in the technical proposal is: shooting mobile phone can be regulated shooting focal length automatically when taking the literal photo, makes it reach literal identification predetermined value.Can improve the sharpness of photo by focusing, help to improve the accuracy rate of character recognition, common Digital Zoom camera lens mobile phone, only regulate just passable by software, want the effect that reaches desirable, also need the support of optical zoom, have only some high-end mobile phones that this function is just arranged.
Further improvement in the technical proposal is: photo literal identification module can be divided into Chinese Character Recognition when identification character, digit recognition, the identification of capital and small letter English alphabet, Symbol recognition, the user can specify one or more mixing identifications wherein, be convenient to improve recognition efficiency like this, in general, the scope of choosing character is wide more, recognition accuracy is low more, by calculating like this, the recognition accuracy of numeral is the highest, English alphabet takes second place, is-symbol then, the Chinese Character Recognition effect is the poorest, if mix identification, effect is all lower than the discrimination of any single type, therefore the user selects single character type identification as far as possible, and selected identification types is consistent with the classification of target photo Chinese words as far as possible.Wherein, the user can add self-defined symbol in Symbol recognition, and this self-defined symbol is from other texts of mobile phone or the hand-written symbol of creation certainly of user.
Further improvement in the technical proposal is: increase a miniature magnifier on the picture taking lens of mobile phone, this magnifier can be designed to movable, is removing in photographed, only when taking literal class photo, shift miniature magnifier onto mirror before.And miniature magnifier has the focal length micromatic setting, is equivalent to optical zoom like this, literal can be taken very clear, is convenient to identification.
The beneficial effect that the present invention can bring is:
The present invention realizes on the basic hardware basis of existing mobile phone, only just can realize by developing specific software, cost is low, but practical value is very high, common low pixel mobile phone just can be accomplished, the camera function of mobile phone can be filled part like this and excavate out, also allow the user become possibility with the mobile phone data collection whenever and wherever possible.
[description of drawings]
No accompanying drawing
[embodiment]
Implement the present invention better for the ease of those skilled in the art, the present invention is described in further detail below in conjunction with embodiment.
Here with the taking pictures and discern and be input as example explanation of business card, business card is the data that the modal cellphone subscriber of needs imports, and quantity of information is less, and discernible character range is little, is one of ideal application of handset identity input.Consider that the word on some business card is smaller, can increase a miniature magnifier on the picture taking lens of mobile phone, small character can be amplified demonstration like this, the effect of taking is just relatively good.
Before using mobile phone photograph, to debug the correlation parameter of mobile phone earlier, mainly regulate the brightness of taking pictures, if mobile phone has automatic adjusting brightness function, just need not change.Before also will moving miniature magnifier to camera lens, make and take pictures, obtain clear photograph, the focal length fine adjustment function is arranged, also will regulate its focal length if amplify by magnifier, make photo literal on the mobile phone screen show when the most clear till.
When choosing business card and begin to take pictures, as far as possible with in the photo of the content imaging on the business card,, only clap into many, but the content on photo preferably there is certain integrality if content is more on the business card.Can clap a name cards and discern immediately, also can once multiple name cards have been taken, unified again identification.
After beginning identification, mobile phone literal identification module comparison film carries out aggregate analysis earlier, delineate out each character zone scope on the photo according to character outline, it is character block, just the character that will discern is one by one found out earlier, the correct description separately between adjacent character is if the adjacent character system of connecting together that has can't accurately discern, need user's manual process, correctly separately with it.
The literal identification module is analyzed one by one to each character block in every photo then, mainly is the stroke lines of analyzing character in the character block, and character is correctly identified.In order to improve the accuracy rate of identification, can adopt some succinct technological means, consider that the character on the business card has certain rule, system can fully discern according to rule when analyzing.The following several classes of character classification on the business card can be discerned:
(1) name: generally be business card independently two to four Chinese characters in unilateral, and also bigger than other characters, first character is the character in the One Hundred Family Names, according to these characteristics, just the name character recognition can be come out.
(2) telephone number: characteristics are that a plurality of numerical characters are arranged continuously, the front that has also has printed words such as " phone ", " mobile phone ", " number ", also identify easily, also can advance any in the phone and be divided into " mobile phone ", " office telephone ", " fax ", " exchange " etc., certain characteristics are all arranged, have a plurality of phones sometimes in the business card simultaneously, can be according to the space of analyzing between the identification continuous number, meet etc., a plurality of phones are correctly separated.
(3) mailbox: maximum characteristics are exactly that symbol " @ " is arranged in the character string, are easy to identification.
(4) network address: maximum characteristics are exactly that letter " WWW " is arranged in the character string, also are easy to identification.
(5) Business Name: a bit of continuous Chinese character, ending place have " company ", " group ", " printed words such as " center ", " office ", " institute ", " institute ", " university is ", " school " are also than being easier to identification.
(6) address: continuous a bit of Chinese character, characteristics are that beginning place has " address ", place name in the character string, the end of string place have " mansion ", " garden ", " number ", printed words such as " chamber ", " room ", also identification easily.
(7) post: mainly be some post class characters, as printed words such as " president ", " managers ".
Above-mentioned classification has been arranged, the content on the name cards can have been identified basically and is filled up to automatically in the address list, the place wrong, but user's manual setting system identification.
The identification input of comparison film Chinese words always has certain error, this mainly is that the low institute of similarity degree readability high and that take pictures of character in the photo causes, in order to correct this error, character of the every identification of system, all be listed as a plurality of characters similar on the stroke structure, and will be the most close one be arranged in the foremost, character as the acquiescence selection, when being difficult to differentiate in the close system of similarity degree in addition, with usage frequency high come the front, as " people " and " going into ", " potato " and " alpine yarrow ", sometimes system just is difficult to distinguish, just can with usage frequency high be arranged in the front, can improve recognition accuracy effectively.
This recognition system also has the intelligence learning function, it mainly is automatic adjustment to identification candidate characters collection usage frequency, when mobile phone dispatches from the factory, the character usage frequency concordance list of installing is the general purpose table of a linguistics mechanism statistics, but the user in actual use, the usage frequency of character can be variant in the data of collecting because of user's hobby, occupational characteristic etc., if therefore system progressively adapts to user's characteristics, the user uses the handset identity input just more and more handy, and here the usage frequency by automatic adjustment character set just can achieve the goal.As the selected mistake of certain character once, its frequency value just increases by 1.Wrong to system identification in addition character, the user can select by hand, the manual character of selecting of user, the increasable amplitude of its frequency is big, as increases by 2.In continuous use, system will more and more be fit to user's custom like this, and recognition accuracy improves constantly.
Claims (10)
1, a kind of character input method with the mobile phone photograph mode, with common shooting mobile phone photographic subjects literal photo, by software built-in in the mobile phone literal in the photo is identified as text character again, realize this method, comprise an optical photographing module and a photo literal identification module in the mobile phone, the character set that comprises GB2312 in the literal identification module, system sets up a concordance list to the character in the character set from high to low according to its usage frequency in social public publication, and the concrete steps of this method are as follows:
Steps A: with mobile phone the target text material being taken is multiple pictures;
Step B: mobile phone literal identification module is analyzed one by one to every photo, delineates out each character zone scope, i.e. character block on the photo according to character outline;
Step C: the literal identification module is analyzed one by one to each character block in every photo, analyze the stroke lines of character in the character block, and from character set, select a plurality of close characters, arrange from high to low by similarity degree, and select text character after first the highest character of similarity degree is defaulted as the identification of character block correspondence automatically;
Step D: system is given tacit consent to the character of selecting mistake automatically, and the user can select in other candidate characters by hand, if correct character is imported by hand by the user not in the candidate tabulation;
Step e: when input identification confirmed used character, every use once, system increases a unit with its usage frequency automatically.
2, character input method according to claim 1 is characterized in that described shooting mobile phone adopts the microshot mode when photographic subjects literal photo.
3, character input method according to claim 1, its feature has hand input device at shooting mobile phone, the stroke on taking pictures lack or when not knowing, the user can replenish by handwriting mode and lack or unclear stroke.
4, character input method according to claim 1 is characterized in that described shooting mobile phone can regulate brightness of image automatically when taking the literal photo, makes it reach literal identification predetermined value.
5, character input method according to claim 1 is characterized in that described shooting mobile phone can regulate shooting focal length automatically when taking the literal photo, makes it reach literal identification predetermined value.
6, character input method according to claim 1, it is characterized in that photo literal identification module can be divided into Chinese Character Recognition, digit recognition, the identification of capital and small letter English alphabet, Symbol recognition when identification character, the user can specify one or more mixing identifications wherein.
7, character input method according to claim 6 is characterized in that, the user can add self-defined symbol in Symbol recognition, and this self-defined symbol is from other texts of mobile phone or the hand-written symbol of creation certainly of user.
8, character input method according to claim 1 is characterized in that, in the text character identifying of photo, and the candidate characters of user's manual setting, first candidate characters when next time discerning similar character.
9, character input method according to claim 1, it is characterized in that increase a miniature magnifier on the picture taking lens of mobile phone, this magnifier can be designed to movable, removing in photographed, only when taking literal class photo, shift miniature magnifier onto mirror before.
10, character input method according to claim 1 is characterized in that described miniature magnifier has the focal length micromatic setting.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200710074531 CN101059840A (en) | 2007-05-24 | 2007-05-24 | Words input method using mobile phone shooting style |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200710074531 CN101059840A (en) | 2007-05-24 | 2007-05-24 | Words input method using mobile phone shooting style |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101059840A true CN101059840A (en) | 2007-10-24 |
Family
ID=38865938
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 200710074531 Pending CN101059840A (en) | 2007-05-24 | 2007-05-24 | Words input method using mobile phone shooting style |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101059840A (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102591477A (en) * | 2012-01-18 | 2012-07-18 | 邓晓波 | Character selection method and character selection device for typing in short sentence |
CN102799379A (en) * | 2012-07-13 | 2012-11-28 | 北京小米科技有限责任公司 | Character input method and character input device |
CN101873364B (en) * | 2009-04-27 | 2012-12-05 | 华硕电脑股份有限公司 | Folding type electronic equipment |
CN103533503A (en) * | 2013-10-22 | 2014-01-22 | 上海斐讯数据通信技术有限公司 | Method and system for recording and acquiring bus time table through photographing |
CN103678286A (en) * | 2012-09-14 | 2014-03-26 | 腾讯科技(深圳)有限公司 | Translation method and system |
CN104182381A (en) * | 2013-05-23 | 2014-12-03 | 欧姆龙株式会社 | character input method and system |
CN104881149A (en) * | 2015-06-24 | 2015-09-02 | 北京奇虎科技有限公司 | Input method and device based on video stream |
CN105653160A (en) * | 2016-02-25 | 2016-06-08 | 努比亚技术有限公司 | Text determining method and terminal |
CN106557766A (en) * | 2016-11-22 | 2017-04-05 | 宇龙计算机通信科技(深圳)有限公司 | Ambiguous characters processing method, system and electronic equipment |
CN103970452B (en) * | 2014-03-31 | 2017-09-22 | 联想(北京)有限公司 | A kind of information processing method and device |
WO2017185613A1 (en) * | 2016-04-29 | 2017-11-02 | 乐视控股(北京)有限公司 | Method and device for inputting character string, and electronic device |
CN108009546A (en) * | 2016-10-28 | 2018-05-08 | 北京京东尚科信息技术有限公司 | information identifying method and device |
CN108650464A (en) * | 2018-05-15 | 2018-10-12 | 深圳市沃特沃德股份有限公司 | Business card recording method and intelligent terminal |
CN109617946A (en) * | 2018-11-06 | 2019-04-12 | 国电电力发展股份有限公司 | The method and photographing device that Text region result data customized in image is uploaded automatically |
CN111381683A (en) * | 2018-12-28 | 2020-07-07 | 薛康泰华 | Photographing recognition input method and software |
-
2007
- 2007-05-24 CN CN 200710074531 patent/CN101059840A/en active Pending
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101873364B (en) * | 2009-04-27 | 2012-12-05 | 华硕电脑股份有限公司 | Folding type electronic equipment |
CN102591477A (en) * | 2012-01-18 | 2012-07-18 | 邓晓波 | Character selection method and character selection device for typing in short sentence |
CN102799379A (en) * | 2012-07-13 | 2012-11-28 | 北京小米科技有限责任公司 | Character input method and character input device |
CN103678286A (en) * | 2012-09-14 | 2014-03-26 | 腾讯科技(深圳)有限公司 | Translation method and system |
CN103678286B (en) * | 2012-09-14 | 2017-06-06 | 腾讯科技(深圳)有限公司 | A kind of interpretation method and system |
CN104182381A (en) * | 2013-05-23 | 2014-12-03 | 欧姆龙株式会社 | character input method and system |
CN103533503A (en) * | 2013-10-22 | 2014-01-22 | 上海斐讯数据通信技术有限公司 | Method and system for recording and acquiring bus time table through photographing |
CN103533503B (en) * | 2013-10-22 | 2017-08-04 | 上海斐讯数据通信技术有限公司 | It is a kind of based on photograph to record obtain bus timetable method and system |
CN103970452B (en) * | 2014-03-31 | 2017-09-22 | 联想(北京)有限公司 | A kind of information processing method and device |
CN104881149A (en) * | 2015-06-24 | 2015-09-02 | 北京奇虎科技有限公司 | Input method and device based on video stream |
CN104881149B (en) * | 2015-06-24 | 2018-04-20 | 北京奇虎科技有限公司 | Input method and device based on video flowing |
CN105653160A (en) * | 2016-02-25 | 2016-06-08 | 努比亚技术有限公司 | Text determining method and terminal |
WO2017185613A1 (en) * | 2016-04-29 | 2017-11-02 | 乐视控股(北京)有限公司 | Method and device for inputting character string, and electronic device |
CN108009546A (en) * | 2016-10-28 | 2018-05-08 | 北京京东尚科信息技术有限公司 | information identifying method and device |
CN106557766A (en) * | 2016-11-22 | 2017-04-05 | 宇龙计算机通信科技(深圳)有限公司 | Ambiguous characters processing method, system and electronic equipment |
CN106557766B (en) * | 2016-11-22 | 2020-05-19 | 宇龙计算机通信科技(深圳)有限公司 | Fuzzy character processing method and system and electronic equipment |
CN108650464A (en) * | 2018-05-15 | 2018-10-12 | 深圳市沃特沃德股份有限公司 | Business card recording method and intelligent terminal |
CN109617946A (en) * | 2018-11-06 | 2019-04-12 | 国电电力发展股份有限公司 | The method and photographing device that Text region result data customized in image is uploaded automatically |
CN111381683A (en) * | 2018-12-28 | 2020-07-07 | 薛康泰华 | Photographing recognition input method and software |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101059840A (en) | Words input method using mobile phone shooting style | |
US7787693B2 (en) | Text detection on mobile communications devices | |
US8988543B2 (en) | Camera based method for text input and keyword detection | |
CN110008944B (en) | OCR recognition method and device based on template matching and storage medium | |
CN101523346B (en) | Image layout constraint generation | |
US20050050165A1 (en) | Internet access via smartphone camera | |
US9531999B2 (en) | Real-time smart display detection system | |
US20090034844A1 (en) | System and method for capturing and processing business data | |
JPWO2007004519A1 (en) | Search system and search method | |
CN106462766A (en) | Image capturing parameter adjustment in preview mode | |
CN101667251A (en) | OCR recognition method and device with auxiliary positioning function | |
JP2011008752A (en) | Document operation system, document operation method and program thereof | |
CN103065146A (en) | Character recognition method for power communication machine room dumb equipment signboards | |
US9070038B2 (en) | Techniques including URL recognition and applications | |
EP2012510A1 (en) | Method and apparatus for connecting a cellular telephone user to the internet | |
CN104537339A (en) | Information identification method and information identification system | |
KR20130066819A (en) | Apparus and method for character recognition based on photograph image | |
CN1700646A (en) | Utilizing a scannable URL (universal resource locator) | |
CN1878182A (en) | Name card input recognition mobile phone and its recognizing method | |
WO2022062027A1 (en) | Wine product positioning method and apparatus, wine product information management method and apparatus, and device and storage medium | |
US20070019248A1 (en) | Method for collecting information of business cards in a mobile communication device | |
CN113159029A (en) | Method and system for accurately capturing local information in picture | |
CN2634766Y (en) | Character scanning and identifying device for cell phone | |
CN1734467A (en) | Mobile phone capable of inputting text by taking pictures and inputting method thereof | |
TWI420404B (en) | Character recognition system and method for the same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C57 | Notification of unclear or unknown address | ||
DD01 | Delivery of document by public notice |
Addressee: Chen Jia Document name: Notice of conformity |
|
C57 | Notification of unclear or unknown address | ||
DD01 | Delivery of document by public notice |
Addressee: Chen Jia Document name: Notification of Publication of the Application for Invention |
|
DD01 | Delivery of document by public notice |
Addressee: Chen Jia Document name: Notification of before Expiration of Request of Examination as to Substance |
|
DD01 | Delivery of document by public notice |
Addressee: Chen Jia Document name: Notification that Application Deemed to be Withdrawn |
|
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20071024 |