CN101281592A - Method for estreating words based on camera and character recognition technique - Google Patents
Method for estreating words based on camera and character recognition technique Download PDFInfo
- Publication number
- CN101281592A CN101281592A CNA2008100372278A CN200810037227A CN101281592A CN 101281592 A CN101281592 A CN 101281592A CN A2008100372278 A CNA2008100372278 A CN A2008100372278A CN 200810037227 A CN200810037227 A CN 200810037227A CN 101281592 A CN101281592 A CN 101281592A
- Authority
- CN
- China
- Prior art keywords
- literal
- camera
- interested
- cpu
- text image
- 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
Landscapes
- User Interface Of Digital Computer (AREA)
Abstract
The invention relates to a character extracting method based on the camera and the character identification technique, specifically relates to a method for character extraction using the text image shot by the camera, belonging to the technical field of the character input using the computer to combining the character identification technique. A program for controlling and commanding the CPU to execute the following operations is stored in the storage of the intelligent device having the text image analysis and character identifying function: affirming the users to select the interesting word area in the accustomed selecting way on the display of the intelligent device for showing the interface of the text image; analyzing and character identifying the character image in the selected area, commanding the display to show the identification result, reversing the identification result, completing the text extracting. The method has the advantage that the users can select the interesting character on the text image in the accustomed selecting way, thereby completing the text extracting.
Description
Technical field
The present invention relates to a kind of literal extracts method, definitely say, relate to the method that a kind of text image of taking with camera carries out the literal extracts, belong to the technical field of carrying out the literal input with camera in conjunction with character recognition technology based on camera and character recognition technology.
Background technology
In recent years, along with the application of camera on smart mobile phone and microcomputer is more and more, the application of carrying out the literal input in conjunction with character recognition technology based on the text image of taking with camera also more and more widely, as name card recognition technique based on camera, text image analysis and recognition technology etc.One of important use is exactly the literal extracts: the user sees passage at newspapers and periodicals or other places, perhaps a few words wherein, very interested, want to take passages, preserve for share in the future or with friend, he just can take these literal with camera, discerns and put in order interested those literal with character recognition technology from the text image that camera is taken then.
At present such software that utilizes camera to carry out the literal extracts has had the function as " the Chinese king extracts " software in the N800 of the Amoisonic mobile phone just has.The user utilizes this software can select to allow software analyze the text image that camera is taken automatically, and identify wherein literal, also can select to allow the automatic analysis user of software drawing certain the selected specific literal section of frame or other modes or the text image of literal line, and identify literal wherein.
Utilize existing software, the user can obtain: the identification literal of whole section text image; The identification literal of one full line text image; Identification literal with the regional interior text image of choosing with red rectangle frame.To obtain the identification literal of the text image of the inter-bank that red line marks such as the user, he/her have to utilize earlier software to obtain identifying to contain the identification literal of text image of the two style of writing words of the words, recycle other the text edit tool and do the operation that butt truncates, just can obtain the literal in the words at last.
In actual life, people get used to selecting interested literal with following dual mode:
One of mode is on the paper media such as books, newspapers and magazines, draws a line at interested sentence place with pen, and this line can be straight, curved, or even ripple, can draw on word, also can draw below word.
When two of mode was to use edit tool such as microsoft word or reads electronic document with browsers such as internet explore, the user can choose interested sentence by mouse-based operation easily.To select one or several sentences of inter-bank such as the user, the left button that only need to pin mouse moves to the final position of choosing sentence to mouse from the original position of choosing sentence just passable, the path independence that selection and mouse move, mouse also needn't inswept all sentences of choosing, and the literal that edit tool or browser can be found out between original position and the final position automatically comes.
The shortcoming of background technology is that the user can not take passages thereby finish text with selected mode selected interested character area on text image of custom.
Summary of the invention
The objective of the invention is to release a kind of literal extracts method based on camera and character recognition technology.The advantage of this method be the user can with the selected mode of custom the selected interested literal of text image, thereby finish the text extracts.
To achieve these goals, the present invention by the following technical solutions.Be placed with the program that control and commander CPU carry out following operation at the store memory of the smart machine of existing analysis with text image and character identification function: show that at the display of described smart machine the interface of text image confirms the zone of the interested literal that the user selectes with the selected mode of custom, analyze and the interior character image of literal identification selection area, order display Identification display result, preserve recognition result, finish the text extracts.
Now be described with reference to the accompanying drawings technical scheme of the present invention.
A kind of literal extracts method based on camera and character recognition technology, need in the smart machine of analysis with text image and character identification function, to implement, the user selects interested literal according to the literal that is presented on the display, it is characterized in that, described literal is a text image, also deposit the program that control and commander CPU carry out following operation in the described smart machine, described being operating as confirmed the zone of user with the selected interested literal of the selected mode of custom, the concrete operations step on the interface of the display videotex image of described smart machine:
The display videotex image of the described smart machine of the first step, the operating process of beginning execution contexts extracts, the user selects the mode in the zone of the selected interested literal be accustomed to;
Is second step selected: with the zone of line mode at the selected interested literal of text image? in this way, entered for the 4th step, as no, entered for the 3rd step;
The 3rd goes on foot to specify the zone of mode selected interested literal in text image of original position and final position;
The 4th step CPU analyzes and identifies the character image of user's appointment;
The 5th step cpu command display shows recognition result;
The 6th step customer inspection, correction, editor and preservation recognition result;
The 7th EOS: finish the text extracts.
Compare with background technology, the present invention has following advantage:
The present invention imitates people in the enterprising row labels of paper media such as books, newspapers and magazines, and the user's literal interested in the text image that the custom of selected interested literal comes camera is taken during view electronic documents is taken passages, both saved the time of the manual input of user, meet daily operating habit again, convenient and practical.
Description of drawings
Fig. 1 is that the user is with the zone of line mode selected interested literal in text image.
Fig. 2 is the user selectes interested literal text image in the mode that moves to final position from reference position zone.
Fig. 3 is the user selectes interested literal in text image in the mode of clicking reference position and final position zone.
Fig. 4 is that the user is to select the zone of interested literal in the mode of reference position and each standardized symbol of final position place in text image.
Fig. 5 is the operational flowchart of text extracts method of the present invention.
Embodiment
Now further specify technical scheme of the present invention with accompanying drawing in conjunction with the embodiments.All embodiment all operate according to the operating procedure of above-mentioned text extracts method.Each embodiment is only enumerated crucial technical data.
Embodiment 1
Present embodiment needs to implement in based on the touch-screen mobile phone dopod 830 of windows mobile5.0 operating system, the operating frequency of the CPU of this mobile phone is 200MHz, memory size is 48M, camera resolution is the 2M pixel, display is the touch-screen of 2.8 inches 320*240 pixel, with writing pencil, in the second step, selection is, the user streaks interested character area in the text image with writing pencil at touch-screen, and after CPU confirmed, the order display was with inverse, highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
Embodiment 2
Except for the following differences, other and embodiment 1 are identical.In the second step, select no, in the 3rd step, the user with writing pencil contact on touch-screen from text image the original position in the zone of interested literal draw final position.
Embodiment 3
Except for the following differences, other and embodiment 1 are identical.In the second step, select noly, in the 3rd step, the user is clicking the original position of interested character area with writing pencil in text image on the touch-screen, and then clicks the final position of interested character area.
Embodiment 4
Except for the following differences, other and embodiment 1 are identical.In the second step, select noly, in the 3rd step, the user is clicking standardized symbol of original position of interested character area with writing pencil on the touch-screen or with mouse in text image, and then at standardized symbol of interested literal final position.
Embodiment 5
Present embodiment needs to implement in being connected to the microcomputer of camera, the operating frequency of the CPU of this microcomputer is 200MHz, and memory size is 256M, and camera resolution is the 2M pixel, with mouse, in the second step, selection is that the user streaks interested character area in the text image with mouse, after CPU confirms, the order display is with inverse, and is highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
Embodiment 6
Except for the following differences, other and embodiment 5 are identical.In the second step, select no, the 3rd the step in, the user draws final position with the original position in mouse zone of interested literal from text image.
Embodiment 7
Except for the following differences, other and embodiment 5 are identical.In the second step, select noly, in the 3rd step, the user clicks the original position of interested character area with mouse in text image, and then clicks the final position of interested character area.
Embodiment 8
Except for the following differences, other and embodiment 5 are identical.In second step, select not, in the 3rd step, the user uses mouse standardized symbol of the reference position of interested character area in text image, and then at standardized symbol of interested literal final position.
Claims (9)
1, a kind of literal extracts method based on camera and character recognition technology, need in the smart machine of analysis with text image and character identification function, to implement, the user selects interested literal according to the literal that is presented on the display, it is characterized in that, described literal is a text image, also deposit the program that control and commander CPU carry out following operation in the described smart machine, described being operating as confirmed the zone of user with the selected interested literal of the selected mode of custom, the concrete operations step on the interface of the display videotex image of described smart machine:
The display videotex image of the described smart machine of the first step, the operating process of beginning execution contexts extracts, the user selects the mode in the zone of the selected interested literal be accustomed to;
Is second step selected: with the zone of line mode at the selected interested literal of text image? in this way, entered for the 4th step, as no, entered for the 3rd step;
The 3rd goes on foot to specify the zone of mode selected interested literal in text image of original position and final position;
The 4th step CPU analyzes and identifies the character image of user's appointment;
The 5th step cpu command display shows recognition result;
The 6th step customer inspection, correction, editor and preservation recognition result;
The 7th EOS: finish the text extracts.
2, literal extracts method based on camera and character recognition technology according to claim 1, it is characterized in that, this method needs to implement in based on the touch-screen mobile phone dopod 830 of windows mobile5.0 operating system, the operating frequency of the CPU of this mobile phone is 200MHz, memory size is 48M, camera resolution is the 2M pixel, display is the touch-screen of 2.8 inches 320*240 pixel, with writing pencil, in the second step, selection is, the user streaks interested character area in the text image with writing pencil at touch-screen, and after CPU confirmed, the order display was with inverse, highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
3, literal extracts method based on camera and character recognition technology according to claim 1, it is characterized in that, this method needs to implement in based on the touch-screen mobile phone dopod 830 of windows mobile5.0 operating system, the operating frequency of the CPU of this mobile phone is 200MHz, memory size is 48M, camera resolution is the 2M pixel, display is the touch-screen of 2.8 inches 320*240 pixel, with writing pencil, in the second step, select no, in the 3rd step, the user with writing pencil contact on touch-screen from text image the original position in the zone of interested literal draw final position, after CPU confirms, the order display is with inverse, and is highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
4, literal extracts method based on camera and character recognition technology according to claim 1, it is characterized in that, this method needs to implement in based on the touch-screen mobile phone dopod 830 of windows mobile5.0 operating system, the operating frequency of the CPU of this mobile phone is 200MHz, memory size is 48M, camera resolution is the 2M pixel, display is the touch-screen of 2.8 inches 320*240 pixel, with writing pencil, in the second step, select no, the 3rd the step in, the user uses writing pencil in the original position of clicking interested character area on the touch-screen in text image, and then clicking the final position of interested character area, after CPU confirmed, the order display was with inverse, highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
5, literal extracts method based on camera and character recognition technology according to claim 1, it is characterized in that, this method needs to implement in based on the touch-screen mobile phone dopod 830 of windows mobile5.0 operating system, the operating frequency of the CPU of this mobile phone is 200MHz, memory size is 48M, camera resolution is the 2M pixel, display is the touch-screen of 2.8 inches 320*240 pixel, with writing pencil, in the second step, select no, the 3rd the step in, the user with writing pencil on the touch-screen or in text image, click standardized symbol of original position of interested character area with mouse, and then at standardized symbol of interested literal final position, after CPU confirmed, the order display was with inverse, highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
6, the literal extracts method based on camera and character recognition technology according to claim 1, it is characterized in that, this method needs to implement in being connected to the microcomputer of camera, the operating frequency of the CPU of this microcomputer is 200MHz, memory size is 256M, camera resolution is the 2M pixel, with mouse, in the second step, selection is that the user streaks interested character area in the text image with mouse, after CPU confirms, the order display is with inverse, and is highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
7, according to the literal extracts method based on camera and character recognition technology under the claim 1, it is characterized in that, this method needs to implement in being connected to the microcomputer of camera, the operating frequency of the CPU of this microcomputer is 200MHz, memory size is 256M, camera resolution is the 2M pixel, with mouse, in the second step, select no, in the 3rd step, the user draws final position with the original position in mouse zone of interested literal from text image, and after CPU confirmed, the order display was with inverse, highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
8, literal extracts method based on camera and character recognition technology according to claim 1, it is characterized in that, this method needs to implement in being connected to the microcomputer of camera, the operating frequency of the CPU of this microcomputer is 200MHz, memory size is 256M, camera resolution is the 2M pixel, with mouse, in the second step, select no, in the 3rd step, the user clicks the original position of interested character area with mouse in text image, and then clicks the final position of interested character area, after CPU confirms, the order display is with inverse, and is highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
9, literal extracts method based on camera and character recognition technology according to claim 1, it is characterized in that, this method needs to implement in being connected to the microcomputer of camera, the operating frequency of the CPU of this microcomputer is 200MHz, memory size is 256M, camera resolution is the 2M pixel, with mouse, in the second step, select no, in the 3rd step, the user uses mouse standardized symbol of the original position of interested character area in text image, and then at standardized symbol of interested literal final position, after CPU confirms, the order display is with inverse, and is highlighted, or other modes highlight literal between original position and the final position; Then carry out continuously~the seven step of the 4th step.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008100372278A CN101281592A (en) | 2008-05-09 | 2008-05-09 | Method for estreating words based on camera and character recognition technique |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008100372278A CN101281592A (en) | 2008-05-09 | 2008-05-09 | Method for estreating words based on camera and character recognition technique |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101281592A true CN101281592A (en) | 2008-10-08 |
Family
ID=40014052
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2008100372278A Pending CN101281592A (en) | 2008-05-09 | 2008-05-09 | Method for estreating words based on camera and character recognition technique |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101281592A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101859225A (en) * | 2010-05-31 | 2010-10-13 | 济南恒先科技有限公司 | Method for implementing words and tables rapidly entering by digital red drawing |
CN101937312A (en) * | 2010-09-15 | 2011-01-05 | 中兴通讯股份有限公司 | Method for marking electronic book and mobile terminal |
CN101908136B (en) * | 2009-06-08 | 2013-02-13 | 比亚迪股份有限公司 | Table identifying and processing method and system |
CN103093217A (en) * | 2013-01-06 | 2013-05-08 | 北京百度网讯科技有限公司 | Interactive image and character recognition method and device |
WO2015000429A1 (en) * | 2013-07-05 | 2015-01-08 | 腾讯科技(深圳)有限公司 | Intelligent word selection method and device |
CN106951893A (en) * | 2017-05-08 | 2017-07-14 | 奇酷互联网络科技(深圳)有限公司 | Text information acquisition methods, device and mobile 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 |
-
2008
- 2008-05-09 CN CNA2008100372278A patent/CN101281592A/en active Pending
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101908136B (en) * | 2009-06-08 | 2013-02-13 | 比亚迪股份有限公司 | Table identifying and processing method and system |
CN101859225A (en) * | 2010-05-31 | 2010-10-13 | 济南恒先科技有限公司 | Method for implementing words and tables rapidly entering by digital red drawing |
CN101937312A (en) * | 2010-09-15 | 2011-01-05 | 中兴通讯股份有限公司 | Method for marking electronic book and mobile terminal |
CN103093217A (en) * | 2013-01-06 | 2013-05-08 | 北京百度网讯科技有限公司 | Interactive image and character recognition method and device |
WO2015000429A1 (en) * | 2013-07-05 | 2015-01-08 | 腾讯科技(深圳)有限公司 | Intelligent word selection method and device |
CN104281394A (en) * | 2013-07-05 | 2015-01-14 | 腾讯科技(深圳)有限公司 | Method and device for intelligently selecting words |
CN106951893A (en) * | 2017-05-08 | 2017-07-14 | 奇酷互联网络科技(深圳)有限公司 | Text information acquisition methods, device and mobile 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 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105247540B (en) | Manage real-time handwriting recognition | |
JP4297442B2 (en) | Handwritten information input device | |
US7966352B2 (en) | Context harvesting from selected content | |
CN101281592A (en) | Method for estreating words based on camera and character recognition technique | |
CN105830011B (en) | For overlapping the user interface of handwritten text input | |
EP1683075B1 (en) | Boxed and lined input panel | |
EP3246796A1 (en) | Device and method for inputting note information into image of photographed object | |
US20060001656A1 (en) | Electronic ink system | |
CN101533317A (en) | Fast recording device with handwriting identifying function and method thereof | |
CN102339275B (en) | Comment processing method and device for electronic book | |
CN105378753B (en) | apparatus, method and computer-readable data medium for handwriting input | |
US20070098263A1 (en) | Data entry apparatus and program therefor | |
US7962846B2 (en) | Organization of annotated clipping views | |
US20110252316A1 (en) | Translating text on a surface computing device | |
JP2003303047A (en) | Image input and display system, usage of user interface as well as product including computer usable medium | |
EP1719109A2 (en) | Ink correction pad | |
CN102289322A (en) | Method and system for processing handwriting | |
US20050114351A1 (en) | Document information management system | |
US20130117161A1 (en) | Method for selecting and providing content of interest | |
CN109074223A (en) | For carrying out the method and system of character insertion in character string | |
TW201322049A (en) | Electronic device and read guiding method thereof | |
CN108885530A (en) | Visual user interface control is operated with ink order | |
CN102301366A (en) | Information processing device | |
CN106776488A (en) | E-book digest is included and the method and system for presenting | |
US20130011066A1 (en) | System, Method, and Product for Handwriting Capture and Storage |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Open date: 20081008 |