CN102063611B - Method and system for inputting characters - Google Patents
Method and system for inputting characters Download PDFInfo
- Publication number
- CN102063611B CN102063611B CN 201010034478 CN201010034478A CN102063611B CN 102063611 B CN102063611 B CN 102063611B CN 201010034478 CN201010034478 CN 201010034478 CN 201010034478 A CN201010034478 A CN 201010034478A CN 102063611 B CN102063611 B CN 102063611B
- Authority
- CN
- China
- Prior art keywords
- image
- literal line
- literal
- bianry image
- name
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000006243 chemical reaction Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 10
- 230000011218 segmentation Effects 0.000 claims description 10
- 230000009182 swimming Effects 0.000 claims description 10
- 239000000284 extract Substances 0.000 claims description 9
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 230000013011 mating Effects 0.000 claims description 2
- 238000003672 processing method Methods 0.000 claims description 2
- 238000012015 optical character recognition Methods 0.000 abstract description 18
- 238000005516 engineering process Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000002950 deficient Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001915 proofreading effect Effects 0.000 description 2
- 238000005314 correlation function Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
Landscapes
- Character Input (AREA)
Abstract
The invention provides a method and a system for inputting characters, and belongs to the technical field of processing of electric data. The method comprises the following steps of: acquiring a two-dimensional image of characters on a scanned medium by an image collector; transmitting the acquired two-dimensional image to a data processing unit by using a universal serial bus (USB) interface in real time; performing binary processing on the two-dimensional image by the data processing unit to acquire a binary image; splicing the two-dimensional image in real time on the basis of image registration by the data processing unit to acquire a panoramic binary image of the characters on the scanned medium; correcting the panoramic binary image by the data processing unit; extracting a character line to be identified from the panoramic binary image by the data processing unit; and performing optical character recognition (OCR) on the characters to be identified and displaying by the data processing unit. The method and the system improve the image processing speed, scanning speed of a scanning pen and the entering speed.
Description
Technical field
The invention belongs to electric technical field of data processing, particularly a kind of character input method.
Background technology
Current, computer input method mainly contains the types such as five-stroke input method, spelling input method, hand-writing input method.When we carry out Data acquisition,, often need to be with the words input in a large amount of paper documents in computing machine, if carry out typing by above-mentioned several input methods, time-consuming again effort.By optical character identification (OCR, Optical Character Recognition) technology, the literal of needs is taken as image, then image is identified, can realize Rapid input.
But there is the slower defective of sweep velocity, input speed of image processing speed, wand in prior art.
Summary of the invention
Technical matters to be solved by this invention is to provide a kind of improved computer scanning input method for the defective of prior art existence, realizes the Rapid input of document information.
For solving the problems of the technologies described above, the invention provides a kind of character input method, this character input method may further comprise the steps:
S101: obtain the two dimensional image that is scanned literal on the medium by image acquisition device;
S102: the described two dimensional image that will obtain carries out subsequent treatment by the USB interface real-time Transmission to data processing unit;
S103: described two dimensional image is carried out binary conversion treatment, obtain bianry image;
S104: based on image registration described bianry image is spliced in real time, obtain to be scanned the panorama bianry image of literal on the medium;
S105: the panorama bianry image is proofreaied and correct;
S106: by extracting literal line to be identified in the panorama bianry image after the described correction;
S107: described literal line to be identified is carried out OCR identification and demonstration.
Preferably, among the described step S103, by the adaptive threshold binary processing method described two dimensional image is carried out binary conversion treatment.
Preferably, the image registration among the described step S104 is that the displacement relation between adjacent two two field pictures is realized before and after calculating by the overlapping region of mating adjacent image interframe in the described bianry image.
Preferably, also comprise before the described step S105: judge whether the panorama bianry image exists black matrix wrongly written or mispronounced character image, if so, then identify described black matrix wrongly written or mispronounced character image and carry out inverse; Otherwise, direct execution in step S105.
Preferably, among the described step S105, blacken the name of disposal route by the segmentation distance of swimming correction of the literal line in the described panorama bianry image is horizontality.
Describedly blacken the name of disposal route by the segmentation distance of swimming and the literal line in the described panorama bianry image is proofreaied and correct as horizontality comprised:
Merge the literal line that obtains in the described panorama bianry image by connected component labeling;
From described literal line, select delegation as the correction reference literal line;
Blacken the name of method by the segmentation distance of swimming and obtain blackening the name of the benchmark literal line;
By the described benchmark literal line of blackening the name of is sampled and is calculated the buckling curve of literal line;
According to described buckling curve described panorama bianry image being proofreaied and correct is horizontality.
Preferably, among the described step S106, obtain a plurality of complete literal line in the panorama bianry image after the described correction by horizontal projection, and according to described a plurality of complete literal lines separately positional information identification and extract literal line to be identified.
Preferably, described step S101 is specially: obtain the two dimensional image that is scanned literal on the medium by wand.
The present invention also provides a kind of character input system, and this literal input system comprises:
Image acquisition device gathers the two dimensional image of literal to be entered;
The USB transmitter sends described two dimensional image to image processor;
Image processor successively carries out binary conversion treatment, splices in real time, proofreaies and correct and extracts literal line to be identified the two dimensional image from described USB transmitter;
The OCR recognizer carries out OCR identification and demonstration to described literal line to be identified;
Controller is controlled described image acquisition device, USB transmitter, image processor and OCR recognizer execution operation separately.
Preferably, described image acquisition device is wand.
Compared with prior art, the advantage of technical solution of the present invention is: the two dimensional image by literal that image acquisition device is obtained is real-time transmitted in the graphics processing unit, carried out the processing such as binaryzation, registration, splicing, correction and identification by graphics processing unit, and image registration wherein is based on bianry image and carries out, like this, image processing speed, the sweep velocity of wand, input speed have been improved.
Description of drawings
Fig. 1 is the process flow diagram according to character input method of the present invention;
Fig. 2 is the two dimensional image that obtains by wand according among the embodiment of character input method of the present invention;
Fig. 3 carries out the bianry image that obtains after the binary conversion treatment to two dimensional image among Fig. 2;
Fig. 4 is the panorama bianry image that the bianry image among Fig. 3 is spliced in real time rear acquisition;
Fig. 5 is by the result after the panorama bianry image is proofreaied and correct among Fig. 4;
Fig. 6 is the literal line to be identified that extracts by in the panorama bianry image after proofreading and correct among Fig. 5;
Fig. 7 is the structural representation according to character input system of the present invention.
Embodiment
The character input method that the present invention proposes is described as follows in conjunction with the accompanying drawings and embodiments.
Character input method provided by the invention comprises following steps:
Image acquisition: utilize image acquisition device to gather image; Particularly, described image acquisition device can be wand, namely utilizes the photoinduction sensor of wand to obtain the two dimensional image that is scanned literal.Wand has one with the photoinduction sensor of optical lens, and such as ccd sensor, it can record gray scale or color 2 D image.By wand, can obtain the consecutive image sequence that is scanned literal on the medium, and adjacent image interframe has certain overlapping.
Image transmitting: the two dimensional image that image acquisition step is obtained is real-time transmitted in the data processing unit by USB interface.The data processing unit of present embodiment is computing machine.Particularly, connect by USB interface between image acquisition device and the computing machine, so that the two dimensional image that obtains is real-time transmitted in the computing machine by USB interface.
Image is processed: in computing machine, carry out first image binaryzation, then utilize the overlapping region of adjacent image interframe to carry out Image Mosaics, after finishing, splicing obtains being scanned the panorama bianry image of literal, the panorama bianry image is proofreaied and correct, extracted literal line to be identified the panorama bianry image after proofreading and correct.
Particularly, in computing machine, the scan image that receives is processed in real time: at first, image is carried out the adaptive threshold binary conversion treatment, obtain the bianry image of two dimensional image; Then bianry image is spliced, the core of bianry image splicing is to carry out image registration, a two field picture is with respect to the displacement of former frame image after namely calculating by the coupling overlapping region, for bianry image, available 0 expression background pixel value, represent the foreground pixel value with 1, the present invention proposes index---the registration coefficient of a tolerance bianry image similarity, namely in the overlapping region all pixels ask after the AND operation add and value and overlapping region in all pixels ask adding and the ratio of value after the inclusive-OR operation.The molecular moiety of registration coefficient is the form of image correlation function, can adopt FFT to carry out speed-up computation, the denominator part of registration coefficient can adopt integrogram to carry out speed-up computation, carries out image registration as the basis take this index, after obtaining registration position, just can realize Image Mosaics.After single pass finished, Image Mosaics also finished, and obtains the panorama bianry image of scan text.The panorama bianry image may be white gravoply, with black engraved characters, also may be the black matrix wrongly written or mispronounced character, for the situation of black matrix wrongly written or mispronounced character, need to carry out inverse and process; Literal line in the panorama bianry image is not level generally, has flexural deformation, it need to be corrected into horizontality; Except literal line to be identified, may also comprise some invalid literal lines in the panorama bianry image after the correction, therefore literal line to be identified need to be extracted, literal line to be identified generally is positioned at the position near the centre of multline text, can extract accordingly.
Literal identification: literal line to be identified is sent into the OCR module identify.
Particularly, in computing machine, literal line to be identified is carried out OCR identification.
The below will enumerate a specific embodiment character input method of the present invention will be further elaborated, as shown in Figure 1.
At first wand is linked to each other with computing machine by USB interface, and the scan process software in the startup calculating, then carry out following steps:
S101: obtain the two dimensional image that is scanned literal on the medium by wand.
Particularly, wand is attached to is scanned on the medium, allow literal line to be identified be positioned at the central authorities of wand acquisition window as far as possible, then drag wand according to certain orientation and carry out image acquisition, wand is according to certain frequency acquisition image.In the present embodiment, frequency was 120 frame/seconds, namely obtained 120 two field pictures in the per second; Be illustrated in figure 2 as the two dimensional image that present embodiment obtains by wand.
S102: the described two dimensional image that will obtain is real-time transmitted in the computing machine by USB interface; The two dimensional image that wand obtains is real-time transmitted in the computing machine via USB interface; Computing machine can be notebook, personal desktop machine, also can be server.
S103: described two dimensional image is carried out binary conversion treatment, obtain bianry image; In the present embodiment, adopt OTSU adaptive threshold Binarization methods that image is carried out binaryzation.Be illustrated in figure 3 as two dimensional image among Fig. 2 is carried out the bianry image that obtains after the binary conversion treatment.
S104: based on image registration described bianry image is spliced in real time, obtain to be scanned the panorama bianry image of literal on the medium; Described image registration is the relative displacement relational implementation of adjacent two two field pictures before and after calculating by the coupling overlapping region.
Particularly, because bianry image can represent background pixel value with 0, represent the foreground pixel value with 1, with all pixels in the overlapping region ask after the AND operation add and value and overlapping region in all pixels ask after the inclusive-OR operation add and the ratio of value as the Measure Indexes of overlapping region similarity, and obtain the optimal registration position according to this tolerance by pixel search; After obtaining the optimal registration position, only need the Non-overlapping Domain in the former frame image in the adjacent image frame is recorded; Since the first two field picture, the non-overlapped part in the former frame image in adjacent two two field pictures is spliced in the panorama bianry image goes, carry out so successively, to the last a two field picture; A two field picture place in the end owing to there has not been thereafter successive image, goes so it all is spliced in the panorama bianry image.After splicing is finished, obtain being scanned the panorama bianry image of literal; Be illustrated in figure 4 as the panorama bianry image that the bianry image among Fig. 3 is spliced in real time rear acquisition.
S105: the panorama bianry image is proofreaied and correct.
Because the panorama bianry image may be white gravoply, with black engraved characters, it also may be the black matrix wrongly written or mispronounced character, before carrying out the correction of panorama bianry image, black matrix wrongly written or mispronounced character image will be carried out inverse processes, for this reason, need to judge whether the panorama bianry image needs to carry out inverse, the panorama bianry image is generally rectangular image, generally not strict rectangle and be scanned character area, therefore the panorama bianry image is general larger than being scanned character area, the part that goes out greatly generally is filled to white, only needs utilization to be scanned character area image is judged; Determination methods comprises: in the every delegation image pixel in being scanned character area, divide into groups continuously as one group take every N continuous image vegetarian refreshments, then calculating pixels in all groupings and all be the group of deceiving pixel, to count B and pixel all be that the group of white pixel is counted W, if W＞B, judge that then image is white gravoply, with black engraved characters, otherwise judge that image is the black matrix wrongly written or mispronounced character; For the panorama bianry image that is identified as the black matrix wrongly written or mispronounced character, wherein the character area that is scanned is carried out inverse and processes; Through judging that panoramic picture shown in Fig. 4 is white gravoply, with black engraved characters, do not need to carry out inverse and process.
Literal line in the panorama bianry image is not level generally, has flexural deformation, it need to be corrected into horizontality.Concrete grammar comprises:
(1) merges the literal line that obtains in the panorama bianry image by connected component labeling;
(2) from above-mentioned literal line, select delegation as the correction reference literal line;
(3) utilize the segmentation distance of swimming to blacken the name of technology (segmented run-length smearing) and obtain blackening the name of the benchmark literal line;
(4) by sampling and calculate the buckling curve of literal line blackening the name of the benchmark literal line;
(5) according to buckling curve the panorama bianry image is proofreaied and correct.
Be as shown in Figure 5 the result after the panorama bianry image is proofreaied and correct among Fig. 4.
S106: by extracting literal line to be identified in the panorama bianry image after the described correction;
By finding out among Fig. 5, except containing literal line to be identified, may also contain other literal lines in the panorama bianry image, therefore, literal line to be identified need to be extracted.Literal line in the panorama bianry image after the correction is in horizontality substantially, along continuous straight runs carries out projection, obtain the projection histogram of black pixel, utilize the minimal value position of projection histogram the literal line in the panorama bianry image can be separated, can obtain each the complete literal line in the panorama bianry image.After obtaining all complete literal lines, according to minute other positional information identification of these complete literal lines and extract literal line to be identified, particularly, get in these literal lines literal line near the centre position as literal line to be identified.Contain two style of writing words among Fig. 5, wherein near that behavior literal line to be identified in image centre position, extract by row, finally obtain the literal line to be identified shown in Fig. 6.
S107: described literal line to be identified is carried out OCR identification and demonstration; Literal line to be identified is sent into the literal that the OCR module is identified and Identification display goes out, particularly, can adopt existing OCR recognizer SDK (Software Development Kit).
As shown in Figure 7, embodiments of the invention also provide a kind of character input system be used to realizing above-mentioned character input method, and this system comprises:
Image acquisition device gathers the two dimensional image of literal to be entered;
The USB transmitter sends described two dimensional image to image processor;
Image processor successively carries out binary conversion treatment, splices in real time, proofreaies and correct and extracts literal line to be identified the two dimensional image from described USB transmitter;
The OCR recognizer carries out OCR identification and demonstration to described literal line to be identified;
Controller is controlled described image acquisition device, USB transmitter, image processor and OCR recognizer execution operation separately.
Particularly, described image acquisition device can be wand, described USB transmitter is USB interface and transmission line, and described image processor, OCR recognizer and controller are to be arranged at that execution in the computing machine is processed with above-mentioned image, the software of the corresponding operation of OCR identification step.
Claims (8)
1. a character input method is characterized in that, may further comprise the steps:
S101: obtain the two dimensional image that is scanned literal on the medium by image acquisition device;
S102: the described two dimensional image that will obtain is real-time transmitted to data processing unit by USB interface and carries out subsequent treatment;
S103: described two dimensional image is carried out binary conversion treatment, obtain bianry image;
S104: based on image registration described bianry image is spliced in real time, obtain to be scanned the panorama bianry image of literal on the medium;
S105: the panorama bianry image is proofreaied and correct;
S106: by extracting literal line to be identified in the panorama bianry image after the described correction;
S107: described literal line to be identified is carried out OCR identification and demonstration;
Among the described step S105, blacken the name of disposal route by the segmentation distance of swimming correction of the literal line in the described panorama bianry image is horizontality;
Describedly blacken the name of disposal route by the segmentation distance of swimming and the literal line in the described panorama bianry image is proofreaied and correct as horizontality comprised:
Merge the literal line that obtains in the described panorama bianry image by connected component labeling;
From described literal line, select delegation as the correction reference literal line;
Blacken the name of method by the segmentation distance of swimming and obtain blackening the name of the benchmark literal line;
By the described benchmark literal line of blackening the name of is sampled and is calculated the buckling curve of literal line;
According to described buckling curve described panorama bianry image being proofreaied and correct is horizontality.
2. character input method as claimed in claim 1 is characterized in that, among the described step S103, by the adaptive threshold binary processing method described two dimensional image is carried out binary conversion treatment.
3. character input method as claimed in claim 1 is characterized in that, the image registration among the described step S104 is that the displacement relation between adjacent two two field pictures is realized before and after calculating by the overlapping region of mating adjacent image interframe in the described bianry image.
4. character input method as claimed in claim 1 is characterized in that, also comprises before the described step S105:
Judge whether the panorama bianry image exists black matrix wrongly written or mispronounced character image, if so, then identify described black matrix wrongly written or mispronounced character image and carry out inverse; Otherwise, direct execution in step S105.
5. character input method as claimed in claim 1, it is characterized in that, among the described step S106, obtain a plurality of complete literal line in the panorama bianry image after the described correction by horizontal projection, and according to described a plurality of complete literal lines separately positional information identification and extract literal line to be identified.
6. such as the described character input method of each claim among the claim 1-5, it is characterized in that described step S101 is specially: obtain the two dimensional image that is scanned literal on the medium by wand.
7. a character input system is characterized in that, comprising:
Image acquisition device gathers the two dimensional image of literal to be entered;
The USB transmitter sends described two dimensional image to image processor;
Image processor successively carries out binary conversion treatment, splices in real time, proofreaies and correct and extracts literal line to be identified the two dimensional image from described USB transmitter;
The OCR recognizer carries out OCR identification and demonstration to described literal line to be identified;
Controller is controlled described image acquisition device, USB transmitter, image processor and OCR recognizer execution operation separately;
Described image processor is blackened the name of disposal route by the segmentation distance of swimming and will be spliced in real time literal line in the panorama bianry image that obtains and proofread and correct and to be horizontality;
Describedly blacken the name of disposal route by the segmentation distance of swimming and the literal line in the described panorama bianry image is proofreaied and correct as horizontality comprised:
Merge the literal line that obtains in the described panorama bianry image by connected component labeling;
From described literal line, select delegation as the correction reference literal line;
Blacken the name of method by the segmentation distance of swimming and obtain blackening the name of the benchmark literal line;
By the described benchmark literal line of blackening the name of is sampled and is calculated the buckling curve of literal line;
According to described buckling curve described panorama bianry image being proofreaied and correct is horizontality.
8. character input system as claimed in claim 7 is characterized in that, described image acquisition device is wand.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010034478 CN102063611B (en) | 2010-01-21 | 2010-01-21 | Method and system for inputting characters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010034478 CN102063611B (en) | 2010-01-21 | 2010-01-21 | Method and system for inputting characters |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102063611A CN102063611A (en) | 2011-05-18 |
| CN102063611B true CN102063611B (en) | 2013-05-29 |
Family
ID=43998880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010034478 Expired - Fee Related CN102063611B (en) | 2010-01-21 | 2010-01-21 | Method and system for inputting characters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102063611B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102324027B (en) * | 2011-05-27 | 2013-05-29 | 汉王科技股份有限公司 | Scanning and identifying device and method |
| CN102982326B (en) * | 2011-09-02 | 2016-05-25 | 汉王科技股份有限公司 | Literal processing method, device and electronic translation pen |
| CN103019834B (en) * | 2011-09-22 | 2016-08-10 | 汉王科技股份有限公司 | A kind of scan pattern changing method based on recording device and recording device |
| CN102360353A (en) * | 2011-09-27 | 2012-02-22 | 汉王科技股份有限公司 | Storage method and device of scanned file as well as scanning equipment |
| CN102360427A (en) * | 2011-09-30 | 2012-02-22 | 汉王科技股份有限公司 | Scanning-mode-adjustable text collection method and scanning device |
| CN103559512B (en) * | 2013-11-01 | 2017-11-03 | 中国联合网络通信集团有限公司 | A kind of Text region output intent and system |
| CN104951741A (en) * | 2014-03-31 | 2015-09-30 | 阿里巴巴集团控股有限公司 | Character recognition method and device thereof |
| CN105631486A (en) * | 2014-10-27 | 2016-06-01 | 深圳Tcl数字技术有限公司 | image character recognition method and device |
| CN105160958A (en) * | 2015-07-24 | 2015-12-16 | 苏州点通教育科技有限公司 | Intelligent pen answering system and method |
| CN106651769B (en) * | 2015-11-03 | 2020-06-02 | 上海蓝渺智能工程有限公司 | Splicing method for scanning and amplifying images of printed files |
| CN105913104B (en) * | 2016-04-13 | 2018-09-21 | 上海厚吉智能科技有限公司 | Calligraphy code generating method and system, analytic method and system and calligraphy code label |
| CN105975560B (en) * | 2016-04-29 | 2020-01-03 | 广东小天才科技有限公司 | Question searching method and device of intelligent equipment |
| CN108038825B (en) * | 2017-12-12 | 2020-08-04 | 维沃移动通信有限公司 | Image processing method and mobile terminal |
| CN108024026B (en) * | 2017-12-14 | 2019-06-28 | 广东金赋科技股份有限公司 | A kind of document fast scanning method |
| CN109889729A (en) * | 2019-03-21 | 2019-06-14 | 国网山东省电力公司经济技术研究院 | A kind of communication system of power grids equipment is taken pictures identification device and method |
| CN113408522A (en) * | 2020-03-16 | 2021-09-17 | 北京字节跳动网络技术有限公司 | Text processing method, device, terminal and storage medium |
| CN111612696B (en) * | 2020-05-21 | 2024-04-02 | 网易有道信息技术（北京）有限公司 | Image stitching method, device, medium and electronic equipment |
| CN114663901A (en) * | 2020-12-04 | 2022-06-24 | 北京搜狗科技发展有限公司 | Image processing method, image recognition device, electronic apparatus, and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1755707A (en) * | 2004-09-30 | 2006-04-05 | 德鑫科技股份有限公司 | Automatic correction method for tilted image |
| CN101008985A (en) * | 2007-01-18 | 2007-08-01 | 章毅 | Seal identification system and controlling method thereof |
| CN101086766A (en) * | 2007-06-04 | 2007-12-12 | 北京航空航天大学 | Fingerprint image frame sequence combination method based on wave form match |
| CN101241395A (en) * | 2007-08-22 | 2008-08-13 | 中国人民解放军国防科学技术大学 | Scan inputting method |
-
2010
- 2010-01-21 CN CN 201010034478 patent/CN102063611B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1755707A (en) * | 2004-09-30 | 2006-04-05 | 德鑫科技股份有限公司 | Automatic correction method for tilted image |
| CN101008985A (en) * | 2007-01-18 | 2007-08-01 | 章毅 | Seal identification system and controlling method thereof |
| CN101086766A (en) * | 2007-06-04 | 2007-12-12 | 北京航空航天大学 | Fingerprint image frame sequence combination method based on wave form match |
| CN101241395A (en) * | 2007-08-22 | 2008-08-13 | 中国人民解放军国防科学技术大学 | Scan inputting method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102063611A (en) | 2011-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102063611B (en) | Method and system for inputting characters | |
| CN102982330B (en) | Character identifying method and identification device in character image | |
| US11482023B2 (en) | Method and apparatus for detecting text regions in image, device, and medium | |
| CN109948510B (en) | Document image instance segmentation method and device | |
| CN102360419B (en) | Method and system for computer scanning reading management | |
| Luo et al. | Design and implementation of a card reader based on build-in camera | |
| US20180101726A1 (en) | Systems and Methods for Optical Character Recognition for Low-Resolution Documents | |
| CN101667251B (en) | OCR recognition method and device with auxiliary positioning function | |
| CN111127339B (en) | Method and device for correcting trapezoidal distortion of document image | |
| CN101719142B (en) | Method for detecting picture characters by sparse representation based on classifying dictionary | |
| CN102622593B (en) | Text recognition method and system | |
| CN105260733A (en) | Method and device for processing image information | |
| CN113486828B (en) | Image processing method, device, equipment and storage medium | |
| CN102360427A (en) | Scanning-mode-adjustable text collection method and scanning device | |
| CN110807454B (en) | Text positioning method, device, equipment and storage medium based on image segmentation | |
| CN103577818A (en) | Method and device for recognizing image characters | |
| Malik et al. | An efficient skewed line segmentation technique for cursive script OCR | |
| US9679218B2 (en) | Method and apparatus for image matching | |
| CN103489254A (en) | Lottery recognition method and lottery recognition system | |
| CN112819004A (en) | Image preprocessing method and system for OCR recognition of medical bills | |
| CN103400130B (en) | File image inclination degree based on energy minimization framework detects and correcting method | |
| CN102682457A (en) | Rearrangement method for performing adaptive screen reading on print media image | |
| CN116704523B (en) | Text typesetting image recognition system for publishing and printing equipment | |
| CN114220091A (en) | Image text detection method and system based on fast Rcnn | |
| KR102562170B1 (en) | Method for providing deep learning based paper book digitizing service |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130529 |