CN103065146A - Character recognition method for power communication machine room dumb equipment signboards - Google Patents
Character recognition method for power communication machine room dumb equipment signboards Download PDFInfo
- Publication number
- CN103065146A CN103065146A CN2012105684605A CN201210568460A CN103065146A CN 103065146 A CN103065146 A CN 103065146A CN 2012105684605 A CN2012105684605 A CN 2012105684605A CN 201210568460 A CN201210568460 A CN 201210568460A CN 103065146 A CN103065146 A CN 103065146A
- Authority
- CN
- China
- Prior art keywords
- character
- picture
- image
- sign board
- library
- 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 discloses a character recognition method for power communication machine room dumb equipment signboards. The character recognition method comprises the following steps: images of the dumb equipment signboards are extracted, and binarization processing is carried out on the images for obtaining optimized pictures; according to pixel points bonded horizontally in the optimized pictures, the images of each line of characters are sliced; according to pixel points bonded longitudinally in the images of each row of the characters, the image of each individual character is sliced; individual character pictures from an initial row to a termination row form character group pictures, the individual character pictures in the character group pictures are contrasted with sample characters of a signboard feature library, and then identification characters can be ensured and character strings can be obtained according to a contrastive accuracy rate; and the character strings and information sorting items of the signboard feature library are matched, and matching results are displayed and stored into an equipment resource management file. By adopting the character recognition method, manual handling efficiency can be improved. Thus, the character recognition method has wide practical application value and important academic significance on dumb equipment management in an electric system.
Description
Technical field
The present invention relates to the character recognition technology of Sign Board, particularly relate to the character recognition method for power communication machine room dummy device Sign Board.
Background technology
Power system device is of a great variety, because a lot of power communication machine rooms are early stage construction, a lot of equipment all do not adopt various advanced technologies to manage, cause the large number quipments register information not enough, although having, equipment component manually sticks one dimension code label, much do not carry out the facility information situations such as work and daily servicing real-time synchronization facility information of filing but exist, thereby to cause more and more " dummy device " to exist.In order to solve the dummy device problem, the method that generally adopts at present re-registers management by artificial scene exactly, but this will be a very huge engineering for electric system, do not have more outstanding technological means to assist and will cause a large amount of manpower and materials to drop into, and the cycle be also very long.Therefore how how rapidly and efficiently these dummy devices of management maintenance in electric system effectively alleviate manpower and materials with dummy device intelligent management means and drop into, and have become one and have been badly in need of very much the significant problem that solves.
Character recognition technology is used very ripe in every field, such as license plate recognition technology, books character translation technology.The main object of character recognition is Chinese character, letter and number, and its structure of Chinese character is very complicated, and its identifying complexity is far above numeral and alphabetical.Literary composition symbol commonly used recognition methods has: artificial neural network method: must look for feature, have the problems such as large, the expensive source of calculated amount; Template matching method: the large discrimination that affects of, character change high on noise requirements; Architectural feature character recognizing method: high to performance requirement, storage resources is large, recognition speed is slow; The statistical nature character recognizing method: the higher character recognition of phase knowledge and magnanimity is poor, sometimes also need seek feature, and discrimination is low in the large situation of changing features.Distinguish the pictograph practical technique and optical character identification (OCR) technology is a kind of high speed, whole sector application is the widest, the most ripe.But the application of OCR technology still faces many technological challenges at present, existing OCR technology generally all is in outdoor good brightness or adopts high-definition camera to take, and identification text object architectural feature is more fixing, changes fewerly, so the recognition capability requirement does not need too high.And for communications equipment room dummy device label, because of building environment light deficiency or dummy device label literal sharpness inadequate, and there be accuracy rate and the recognition speed that all can affect OCR identification in the situations such as pattern is many, color is many, background shading is many, size, literal is big or small, font type is many in the dummy device label.In addition, present intelligent mobile phone system, panel computer operating system, their processing speed is well below the operating system of PC, in this case, how to use the OCR technology at mobile phone or panel computer, and improve its recognition speed and accuracy? this just becomes the bottleneck that the OCR technology further develops.
Summary of the invention
Based on this, be necessary for the problems referred to above, a kind of character recognition method for power communication machine room dummy device Sign Board is provided, management maintenance dummy device rapidly and efficiently, realize the intellectuality of dummy device management, the manpower that alleviates input is safeguarded, strengthens accuracy, the promptness of maintenance of information.
A kind of character recognition method for power communication machine room dummy device Sign Board comprises:
Extract dummy device sign board image, this image is carried out binary conversion treatment, obtain to optimize picture;
According to the horizontal pixel of adhesion in the described optimization picture, the image of every row word is cut; Again according to the vertical pixel of adhesion in the image of every row word, the image of each individual character is cut;
Form the literal picture group by initial row to the individual character picture of termination row, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determine identification character by the accuracy rate of contrast, obtain character string;
The information classification item of described character string and described sign board feature database is mated, show matching result and deposit the equipment resource management archives in.
Correspondingly, a kind of character recognition device for power communication machine room dummy device Sign Board comprises:
Pretreatment unit is used for extracting dummy device sign board image, and this image is carried out binary conversion treatment, obtains to optimize picture;
The picture cutter unit that links to each other with described pretreatment unit is used for according to the horizontal pixel of adhesion of described optimization picture, and the image of every row word is cut; Again according to the vertical pixel of adhesion in the image of every row word, the image of each individual character is cut;
The character contrast unit that links to each other with described picture cutter unit, be used for forming the literal picture group by initial row to the individual character picture of termination row, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determine identification character by the accuracy rate of contrast, obtain character string;
Coupling filing unit with described character contrast unit links to each other is used for the information classification item of described character string and described sign board feature database is mated, and shows matching result and deposits the equipment resource management archives in.
Implement the present invention, have following beneficial effect:
The technology of the present invention will be carried out intelligent management to the dummy device of power communication machine room with OCR image analysis processing technology.By the label literal on panel computer or the smart mobile phone shooting dummy device, to form brand-new Internet of Things management mode after the Word message of OCR image analysis processing technology automatic analysis label photo and the bonding apparatus resource system associated application, such as dummy device file fast application, dummy device attribute fast query, dummy device two-dimension code label quick Fabrication etc., for dummy device provides outstanding, efficient technological means.
Description of drawings
Fig. 1 is the process flow diagram of the character recognition method of power communication machine room dummy device Sign Board of the present invention;
Fig. 2 is the embodiment process flow diagram of the character recognition method of power communication machine room dummy device Sign Board of the present invention;
Fig. 3 is the schematic diagram of the character recognition device of power communication machine room dummy device Sign Board of the present invention;
Fig. 4 is the embodiment schematic diagram of the character recognition device of power communication machine room dummy device Sign Board of the present invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with accompanying drawing.
Fig. 1 is the process flow diagram of the character recognition method of power communication machine room dummy device Sign Board of the present invention, comprising:
S101: extract dummy device sign board image, this image is carried out binary conversion treatment, obtain to optimize picture;
S102: according to the horizontal pixel of adhesion in the described optimization picture, the image of every row word is cut; Again according to the vertical pixel of adhesion in the image of every row word, the image of each individual character is cut;
S103: form the literal picture group by initial row to the individual character picture of termination row, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determine identification character by the accuracy rate of contrast, obtain character string;
S104: the information classification item of described character string and described sign board feature database is mated, show matching result and deposit the equipment resource management archives in.
OCR(Optical Character Recognition, optical character identification) refers to that electronic equipment (for example panel computer, smart mobile phone, scanner or digital camera) checks the character of printing on the paper, by detecting its shape of dark, bright mode decision, then with character identifying method shape is translated into the process of computword; Namely text information is scanned, then image file is carried out analyzing and processing, obtain the process of literal and layout information.The technology of the present invention will be carried out intelligent management to the dummy device of power communication machine room with OCR image analysis processing technology.By the label literal on panel computer or the smart mobile phone shooting dummy device, to form brand-new Internet of Things management mode after the Word message of OCR image analysis processing technology automatic analysis label photo and the bonding apparatus resource system associated application, such as dummy device file fast application, dummy device attribute fast query, dummy device two-dimension code label quick Fabrication etc., for dummy device provides outstanding, efficient technological means.
The technology of the present invention is studied the character recognition technologies in the equipment label in the electric system, and each module of the character recognition system in the equipment label is studied and realizes.Whole scheme is extracted three parts of decomposition by visual pre-service, character recognition, facility information and is formed, the technology input of scheme is individual dummy device label image, and pretreatment portion is divided into picture extraction, binaryzation and inclination rectification, with mixed other engine implementation character recognition and the label information extraction decomposition match management distinguished.Particularly, when carrying out visual pre-service, adopt image filtering processing, Gray Histogram correction and gray level image correlation technique that the part of facility information areal map is highlighted.In order effectively to identify, can also adopt threshold method to carry out the label image binaryzation.By to its label binaryzation, extract the steel seal information in the label image.At last, carry out the rim detection of label image and cut apart, adopt again based on the technology of optical character identification (OCR) and MATLAB software algorithm and finish identification to the label character.
What need supplementary notes is, the present invention has considered that the format characteristic of dummy device label identifies.Mainly comprise: on the one hand, a lot of steel seal information are arranged in the label or print paster information, when extracting the label feature by OCR, need complete its characteristic quantity of extraction.On the other hand, the character in the label is a lot, and especially related Chinese character is abundanter, and the declaration of will that the present invention also forms by Chinese character just realizes intelligentized coupling.
Fig. 2 is the embodiment process flow diagram of the character recognition method of power communication machine room dummy device Sign Board of the present invention.Compared to Figure 1, Fig. 2 is the schematic diagram of specific embodiment.This specific embodiment will carry out Android OCR technological development and the technological development of resource identification flow processing based on the Android mobile phone operating system, will be by the image pre-service for mobile phone OCR technology, character recognition, facility information extracts and decomposes three part compositions, and the input of the technology of whole scheme will be taken individual dummy device label image with the camera of Android mobile phone or Android panel computer, carrying out picture extracts, binaryzation and rectifications of tilting wait operation, and finally carrying out quick character recognition and label resource information at the Android mobile phone operating system, to extract decomposition operation associated.
S201: extract dummy device sign board image, this image is carried out binary conversion treatment, obtain to optimize picture;
S202: according to the horizontal pixel of adhesion in the described optimization picture, the image of every row word is cut; Again according to the vertical pixel of adhesion in the image of every row word, the image of each individual character is cut;
S203: form the literal picture group by initial row to the individual character picture of termination row, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determine identification character by the accuracy rate of contrast, obtain character string;
S204: described information classification item comprises at least a in model, parameter and the kind, from the bebinning character of described character string, and the beginning phrase that is formed by several bebinning characters that search and described model are complementary;
Judge and search for successfully, when being, change S205 over to; When no, change S206 over to;
S205: will start content behind the phrase as the details of parameter query corresponding device;
S206: from the final character of described character string, the ending phrase that is formed by several final characters that search and described kind are complementary;
Judge that the match is successful, changes S207 over to when being; When no, change S208 over to;
S207: the content before the phrase that will end up is as the details of parameter query corresponding device;
S208: with the details of described character string as the parameter query corresponding device;
S209: show matching result and deposit the equipment resource management archives in.
Present technique will be camera function of exploitation of Android mobile phone or Android panel computer, and the main API that takes pictures that realizes calling Android operating system carries out on-the-spot dummy device label position and takes pictures, and finish will be stored in the SD card after taking pictures.
Camera function will be realized the auxiliary raising photo imaging effect of taking of following technology:
1) camera function is automatically controlled and whether is automatically opened the auxiliary shooting of flashlamp because machine room shooting area light is not enough, improves the photo lamplight brightness, promotes discrimination;
2) camera function is controlled the automatic Identifiaction plates photo of camera focus automatically in the situation that the most clear and location label central point is finished shooting automatically.
First, dummy device label image extracts with image binaryzation and processes
Adopt panel computer, smart mobile phone to be taken pictures in current dummy device label position because some machine room area light is not enough, can utilize terminal with auxiliary shooting of flashlamp.Finish after taking pictures and to carry out the optimization process of label image.
The situations such as dummy device label pattern is many, color is many, background shading is many, size difference is large adopt the black and white conversion regime will seriously lose image information.The technology of the present invention will be changed as prerequisite adopts the image binaryzation gray scale to obtain meticulous image information, fully improve discrimination.In order more to adapt to dummy device label characteristics, present technique will be carried out the cutting of 3 * 3 piecemeals to the dummy device label simultaneously, and the effect of cutting mainly is to adopt multithreading to carry out fast processing, and less its discrimination of image more can improve; Enforcement is carried out the binaryzation gradation conversion by piece, and the advantage of image binaryzation gradation conversion is to avoid the situations such as the image border color is overweight, the background dispersion is overlapping to cause the binary picture image distortion, affects discrimination.
In the process of taking pictures, photo can often occur tilting or skew, will affect the pattern recognition success ratio.System will be based on monochrome pixels equilibrium level value after alternately the some analog computation tilts to correct, finish rectification with the highest equilibrium level as base value, and then guarantee the correctness that label graphical analysis and row are cut apart.
This discovery technique provides following treatment technology to finish label photo inclination and offset correction process:
1) the equilibrium level value after alternately the some analog computation tilts to correct based on monochrome pixels to the label photo, with the highest equilibrium level as correcting base value;
2) carry out the inclination and offset processing to correct base value;
3) carry out the overweight identification in edge and the processing of cutting edge.
Second portion, character recognition
Owing to comprising Chinese, English, numeral, punctuate in the dummy device label, may also have other state's outer languages, the technology of the present invention is only applicable to the identifications such as simplified form of Chinese Character, English, numeral, punctuate at present.Also there is variation in the character font pattern in the dummy device label, and some is very large, and some is considerably less, and some also has the character background look.And some character fonts are also adopted the meeting Song typeface, black matrix, variety, regular script etc.Therefore, multi-font multi-mode Suresh Kumar mixed distinguished the maximum characteristics that other engine is the technology of the present invention.
Mixed distinguish other engine and will cut out each character to the label image.The good label small images of binaryzation is carried out every row monochrome pixels point identification, calculate the starting and ending coordinate figure of every line character, finish every line character image and cut figure; Every style of writing word image is continued vertically to cut by column word processing, equally with the monochrome pixels point, find out each literal starting and ending coordinate of capable image, finish and to obtain the independent image of each word.
To build an electric power calculator room equipment label dictionary of information storehouse in the technical program, dictionary library will contain the dictionary values such as the pattern, picture structure, Related product data characteristics of various labels.To carry out the literal identification learning for the single character image that cuts out in dictionary library processes, mixed distinguish other engine graphical analysis goes out eigenvalue cluster to single character, then the eigenvalue cluster with the dictionary library master sample mates, and will screen from high to low by the accuracy rate value.
Third part, the matching analysis
Present technique has been collected a large amount of electric power calculator room equipment label items of information storehouse, will carry out the character mining analysis in label dictionary of information storehouse to the character data that OCR identifies, and analyzes the label dictionary of information group that meets most.
As there is " model " attribute item in the label dictionary of information group that analyzes, and present technique is the character of " model " two word latter range so, carries out the model matching inquiry and obtains current dummy device details thereby submit to CRS as the model value.
As the label dictionary of information group that analyzes can not find attribute item, only has related content, such as the 70P038011ODF block, present technique is with the precedence parse Chinese character so, find out the relevant information that belongs to the ODF block from label dictionary of information storehouse, obtain the backward CRS of corresponding device kind and submit ODF unit type 70P03801 to, thereby obtain current dummy device details.Otherwise, with the details of whole character string as the parameter query corresponding device.
At last, show matching result and deposit the equipment resource management archives in.
Fig. 3 is the schematic diagram of the character recognition device of power communication machine room dummy device Sign Board of the present invention, comprising:
Pretreatment unit is used for extracting dummy device sign board image, and this image is carried out binary conversion treatment, obtains to optimize picture;
The picture cutter unit that links to each other with described pretreatment unit is used for according to the horizontal pixel of adhesion of described optimization picture, and the image of every row word is cut; Again according to the vertical pixel of adhesion in the image of every row word, the image of each individual character is cut;
The character contrast unit that links to each other with described picture cutter unit, be used for forming the literal picture group by initial row to the individual character picture of termination row, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determine identification character by the accuracy rate of contrast, obtain character string;
Coupling filing unit with described character contrast unit links to each other is used for the information classification item of described character string and described sign board feature database is mated, and shows matching result and deposits the equipment resource management archives in.
Fig. 3 is corresponding with Fig. 1, identical among the figure in the method for operation of unit and the method.
Fig. 4 is the embodiment schematic diagram of the character recognition device of power communication machine room dummy device Sign Board of the present invention.
Described information classification item comprises at least a in model, parameter and the kind, and as shown in Figure 4, described coupling filing unit comprises:
The model matching unit is used for from the bebinning character of described character string, the beginning phrase that is comprised of several bebinning characters that search and described model are complementary; When searching, with the details of the content behind the beginning phrase as the parameter query corresponding device;
The kind matching unit that links to each other with described model matching unit is used for when not searching, from the final character of described character string, and the ending phrase that is formed by several final characters that search and described kind are complementary; When the match is successful, with the details of the content before the ending phrase as the parameter query corresponding device; When it fails to match, with the details of described character string as the parameter query corresponding device.
As shown in Figure 4, pretreatment unit also comprises:
The slant correction unit is used for the alternately analog computation equilibrium level value based on monochrome pixels point, image is carried out slant correction process.
Described sign board feature database comprises Song typeface character library, black matrix character library, variety character library, regular script character library, and as shown in Figure 4, described character contrast unit comprises:
Character library screening unit is used for the individual character picture of described literal picture group is compared with the sample word of Song typeface character library, black matrix character library, variety character library, regular script character library respectively, screens from high to low by accuracy rate, determines identification character.
Fig. 4 is corresponding with Fig. 2, identical among the figure in the method for operation of unit and the method.
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 (8)
1. the character recognition method of a power communication machine room dummy device Sign Board is characterized in that, comprising:
Extract dummy device sign board image, this image is carried out binary conversion treatment, obtain to optimize picture;
According to the horizontal pixel of adhesion in the described optimization picture, the image of every row word is cut; Again according to the vertical pixel of adhesion in the image of every row word, the image of each individual character is cut;
Form the literal picture group by initial row to the individual character picture of termination row, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determine identification character by the accuracy rate of contrast, obtain character string;
The information classification item of described character string and described sign board feature database is mated, show matching result and deposit the equipment resource management archives in.
2. the character recognition method of power communication machine room dummy device Sign Board according to claim 1, it is characterized in that, described information classification item comprises at least a in model, parameter and the kind, step with the information classification item of described character string and described sign board feature database mates comprises:
From the bebinning character of described character string, the beginning phrase that is formed by several bebinning characters that search and described model are complementary;
When searching, with the details of the content behind the beginning phrase as the parameter query corresponding device;
When not searching, from the final character of described character string, the ending phrase that is formed by several final characters that search and described kind are complementary; When the match is successful, with the details of the content before the ending phrase as the parameter query corresponding device; When it fails to match, with the details of described character string as the parameter query corresponding device.
3. the character recognition method of power communication machine room dummy device Sign Board according to claim 1 and 2 is characterized in that, this image is carried out after the step of binary conversion treatment, obtains to optimize before the step of picture, also comprises:
Based on the alternately analog computation equilibrium level value of monochrome pixels point, image is carried out slant correction process.
4. the character recognition method of power communication machine room dummy device Sign Board according to claim 1 and 2, it is characterized in that, described sign board feature database comprises Song typeface character library, black matrix character library, variety character library, regular script character library, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determines that by the accuracy rate of contrast the step of identification character comprises:
Individual character picture in the described literal picture group is compared with the sample word of Song typeface character library, black matrix character library, variety character library, regular script character library respectively, screen from high to low by accuracy rate, determine identification character.
5. the character recognition device of a power communication machine room dummy device Sign Board is characterized in that, comprising:
Pretreatment unit is used for extracting dummy device sign board image, and this image is carried out binary conversion treatment, obtains to optimize picture;
The picture cutter unit that links to each other with described pretreatment unit is used for according to the horizontal pixel of adhesion of described optimization picture, and the image of every row word is cut; Again according to the vertical pixel of adhesion in the image of every row word, the image of each individual character is cut;
The character contrast unit that links to each other with described picture cutter unit, be used for forming the literal picture group by initial row to the individual character picture of termination row, the sample word of the individual character picture in the described literal picture group and sign board feature database is compared, determine identification character by the accuracy rate of contrast, obtain character string;
Coupling filing unit with described character contrast unit links to each other is used for the information classification item of described character string and described sign board feature database is mated, and shows matching result and deposits the equipment resource management archives in.
6. the character recognition device of power communication machine room dummy device Sign Board according to claim 5 is characterized in that, described information classification item comprises at least a in model, parameter and the kind, and described coupling filing unit comprises:
The model matching unit is used for from the bebinning character of described character string, the beginning phrase that is comprised of several bebinning characters that search and described model are complementary; When searching, with the details of the content behind the beginning phrase as the parameter query corresponding device;
The kind matching unit that links to each other with described model matching unit is used for when not searching, from the final character of described character string, and the ending phrase that is formed by several final characters that search and described kind are complementary; When the match is successful, with the details of the content before the ending phrase as the parameter query corresponding device; When it fails to match, with the details of described character string as the parameter query corresponding device.
7. according to claim 5 or the character recognition device of 6 described power communication machine room dummy device Sign Boards, it is characterized in that pretreatment unit also comprises:
The slant correction unit is used for the alternately analog computation equilibrium level value based on monochrome pixels point, image is carried out slant correction process.
8. according to claim 5 or the character recognition device of 6 described power communication machine room dummy device Sign Boards, it is characterized in that, described sign board feature database comprises Song typeface character library, black matrix character library, variety character library, regular script character library, and described character contrast unit comprises:
Character library screening unit is used for the individual character picture of described literal picture group is compared with the sample word of Song typeface character library, black matrix character library, variety character library, regular script character library respectively, screens from high to low by accuracy rate, determines identification character.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012105684605A CN103065146A (en) | 2012-12-24 | 2012-12-24 | Character recognition method for power communication machine room dumb equipment signboards |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012105684605A CN103065146A (en) | 2012-12-24 | 2012-12-24 | Character recognition method for power communication machine room dumb equipment signboards |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103065146A true CN103065146A (en) | 2013-04-24 |
Family
ID=48107769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012105684605A Pending CN103065146A (en) | 2012-12-24 | 2012-12-24 | Character recognition method for power communication machine room dumb equipment signboards |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103065146A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678291A (en) * | 2013-12-12 | 2014-03-26 | 苏州市峰之火数码科技有限公司 | Mobile phone translation method |
| CN104268814A (en) * | 2014-10-11 | 2015-01-07 | 北京网秦天下科技有限公司 | Method and device for generating electronic medical record |
| CN105590101A (en) * | 2015-12-28 | 2016-05-18 | 杭州淳敏软件技术有限公司 | Hand-written answer sheet automatic processing and marking method and system based on mobile phone photographing |
| CN107480243A (en) * | 2017-08-10 | 2017-12-15 | 刘燕燕 | Electric power calculator room equipment application method |
| CN108805115A (en) * | 2018-04-24 | 2018-11-13 | 五八同城信息技术有限公司 | Recognition methods, device, equipment and the storage medium of word in a kind of picture |
| CN109271993A (en) * | 2018-09-28 | 2019-01-25 | 徐常柏 | A kind of document scanning system, computer equipment and readable storage medium storing program for executing |
| CN109389084A (en) * | 2018-10-09 | 2019-02-26 | 郑州云海信息技术有限公司 | A kind of method and device handling image information |
| CN109766891A (en) * | 2018-12-14 | 2019-05-17 | 北京上格云技术有限公司 | Obtain the method and computer readable storage medium of installations and facilities information |
| CN110345954A (en) * | 2018-04-03 | 2019-10-18 | 奥迪股份公司 | Navigation system and method |
| CN110991450A (en) * | 2019-12-03 | 2020-04-10 | 厦门亿合恒拓信息科技有限公司 | Method for collecting power grid equipment information and computer readable storage medium |
| CN112861861A (en) * | 2021-01-15 | 2021-05-28 | 珠海世纪鼎利科技股份有限公司 | Method and device for identifying nixie tube text and electronic equipment |
| CN112926568A (en) * | 2021-03-05 | 2021-06-08 | 北京全路通信信号研究设计院集团有限公司 | Interlocking table blueprint identification method and device, electronic equipment and storage medium |
| CN113591066A (en) * | 2021-07-23 | 2021-11-02 | 广州绿怡信息科技有限公司 | Equipment identity identification method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1877598A (en) * | 2005-06-06 | 2006-12-13 | 英华达(上海)电子有限公司 | Method for gathering and recording business card information in mobile phone by using image recognition |
| CN101662581A (en) * | 2009-09-09 | 2010-03-03 | 谭洪舟 | Multifunctional certificate information collection system |
| CN102156853A (en) * | 2010-02-11 | 2011-08-17 | 深圳鼎识科技有限公司 | Certificate reading system and certificate reading method |
-
2012
- 2012-12-24 CN CN2012105684605A patent/CN103065146A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1877598A (en) * | 2005-06-06 | 2006-12-13 | 英华达(上海)电子有限公司 | Method for gathering and recording business card information in mobile phone by using image recognition |
| CN101662581A (en) * | 2009-09-09 | 2010-03-03 | 谭洪舟 | Multifunctional certificate information collection system |
| CN102156853A (en) * | 2010-02-11 | 2011-08-17 | 深圳鼎识科技有限公司 | Certificate reading system and certificate reading method |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678291A (en) * | 2013-12-12 | 2014-03-26 | 苏州市峰之火数码科技有限公司 | Mobile phone translation method |
| CN104268814A (en) * | 2014-10-11 | 2015-01-07 | 北京网秦天下科技有限公司 | Method and device for generating electronic medical record |
| CN105590101A (en) * | 2015-12-28 | 2016-05-18 | 杭州淳敏软件技术有限公司 | Hand-written answer sheet automatic processing and marking method and system based on mobile phone photographing |
| CN107480243A (en) * | 2017-08-10 | 2017-12-15 | 刘燕燕 | Electric power calculator room equipment application method |
| CN110345954A (en) * | 2018-04-03 | 2019-10-18 | 奥迪股份公司 | Navigation system and method |
| CN108805115A (en) * | 2018-04-24 | 2018-11-13 | 五八同城信息技术有限公司 | Recognition methods, device, equipment and the storage medium of word in a kind of picture |
| CN109271993A (en) * | 2018-09-28 | 2019-01-25 | 徐常柏 | A kind of document scanning system, computer equipment and readable storage medium storing program for executing |
| CN109389084A (en) * | 2018-10-09 | 2019-02-26 | 郑州云海信息技术有限公司 | A kind of method and device handling image information |
| CN109766891A (en) * | 2018-12-14 | 2019-05-17 | 北京上格云技术有限公司 | Obtain the method and computer readable storage medium of installations and facilities information |
| CN109766891B (en) * | 2018-12-14 | 2020-11-10 | 北京上格云技术有限公司 | Method for acquiring equipment facility information and computer readable storage medium |
| CN110991450A (en) * | 2019-12-03 | 2020-04-10 | 厦门亿合恒拓信息科技有限公司 | Method for collecting power grid equipment information and computer readable storage medium |
| CN112861861A (en) * | 2021-01-15 | 2021-05-28 | 珠海世纪鼎利科技股份有限公司 | Method and device for identifying nixie tube text and electronic equipment |
| CN112861861B (en) * | 2021-01-15 | 2024-04-09 | 珠海世纪鼎利科技股份有限公司 | Method and device for recognizing nixie tube text and electronic equipment |
| CN112926568A (en) * | 2021-03-05 | 2021-06-08 | 北京全路通信信号研究设计院集团有限公司 | Interlocking table blueprint identification method and device, electronic equipment and storage medium |
| CN112926568B (en) * | 2021-03-05 | 2024-02-13 | 北京全路通信信号研究设计院集团有限公司 | Interlocking table blueprint identification method and device, electronic equipment and storage medium |
| CN113591066A (en) * | 2021-07-23 | 2021-11-02 | 广州绿怡信息科技有限公司 | Equipment identity identification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103065146A (en) | Character recognition method for power communication machine room dumb equipment signboards | |
| US10783400B2 (en) | Generating searchable text for documents portrayed in a repository of digital images utilizing orientation and text prediction neural networks | |
| CN107633239B (en) | Bill classification and bill field extraction method based on deep learning and OCR | |
| KR20190123790A (en) | Extract data from electronic documents | |
| CN102915437A (en) | Text information identification method and system | |
| CN103488983A (en) | Business card OCR data correction method and system based on knowledge base | |
| CN109635805B (en) | Image text positioning method and device and image text identification method and device | |
| CN104182750A (en) | Extremum connected domain based Chinese character detection method in natural scene image | |
| WO2022089170A1 (en) | Caption area identification method and apparatus, and device and storage medium | |
| CN101901484A (en) | Document image segmentation system | |
| CN112508011A (en) | OCR (optical character recognition) method and device based on neural network | |
| CN103996055A (en) | Identification method based on classifiers in image document electronic material identification system | |
| CN112766255A (en) | Optical character recognition method, device, equipment and storage medium | |
| CN113901952A (en) | Print form and handwritten form separated character recognition method based on deep learning | |
| CN107766854A (en) | It is a kind of to realize that the quick page number knows method for distinguishing based on template matches | |
| CN110956167A (en) | Classification discrimination and strengthened separation method based on positioning characters | |
| Ikica et al. | An improved edge profile based method for text detection in images of natural scenes | |
| CN107958261B (en) | Braille point detection method and system | |
| CN112686243A (en) | Method and device for intelligently identifying picture characters, computer equipment and storage medium | |
| Arsenovic et al. | Deep learning driven plates recognition system | |
| Meng et al. | IRIS: smart phone aided intelligent reimbursement system using deep learning | |
| CN114694133B (en) | Text recognition method based on combination of image processing and deep learning | |
| CN108062548B (en) | Braille square self-adaptive positioning method and system | |
| CN112541505B (en) | Text recognition method, text recognition device and computer-readable storage medium | |
| CN113657162A (en) | Bill OCR recognition method based on deep learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130424 |