CN103595861A - Method for enabling terminal to identify phone number and automatically dial or send text message - Google Patents
Method for enabling terminal to identify phone number and automatically dial or send text message Download PDFInfo
- Publication number
- CN103595861A CN103595861A CN201310501585.0A CN201310501585A CN103595861A CN 103595861 A CN103595861 A CN 103595861A CN 201310501585 A CN201310501585 A CN 201310501585A CN 103595861 A CN103595861 A CN 103595861A
- Authority
- CN
- China
- Prior art keywords
- telephone number
- identification
- note
- software
- character
- 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
- Character Discrimination (AREA)
Abstract
The invention discloses a method for enabling a terminal to identify a phone number and automatically dial or send a text message. According to the method, the scanning terminal (1) automatically identifies the phone number on a phone number sample (2) and automatically dials or sends the text message. The method comprises the steps that a camera in the scanning terminal (1) shoots image information on the phone number sample (2), phone number identifying software in the scanning terminal (1) identifies the phone number on the phone number sample (2), automatic dialing software in the scanning terminal (1) completes automatic dialing, the scanning terminal (1) is connected with a user terminal (3) through a communication network (4) to achieve communication, and alternatively automatic message sending software in the scanning terminal (1) sends a message to the user terminal (3) through the communication network (4). According to the method, the phone number can be identified automatically and rapidly, and automatic dialing or message sending can be achieved.
Description
Technical field
The present invention relates to a kind of method of terminal recognition telephone number auto dialing or transmission note, relate in particular to a kind of method of automatic identification telephone number, auto dialing or transmission note.Belong to character recognition and wireless communication field in computer vision.
Background technology
Market Research:
Along with ecommerce these several years is in Chinese rise, also there is fast development in associated express delivery industry.The numeral of announcing according to official of national post system general bureau network address, 2012, nationwide above Courier Service business event amount completed 56.9 hundred million, increases by 54.8% on a year-on-year basis.That is to say that corresponding logistics distribution in 2012 reaches 56.9 hundred million times, national courier need to send altogether getting of 56.9 hundred million notes or Advise By Wire express delivery this year.But, at present artificial Advise By Wire or SMS notification mode need manual dialing to notify each picking people, under normal circumstances, although single manual dialup only has more than ten second, for national courier, annual delay just reaches 56,900,000,000 seconds in the time of dialling phone number, not only time-consuming, and require great effort, directly affected the dispensing efficiency of whole express delivery sector terminal, increased huge human cost and time cost.For the industry of this stressing practical results property of express delivery, such time cost is very huge.
Technical background is introduced:
Image recognition is the important branch of artificial intelligence, and it utilizes optical system or other imaging systems to obtain image information, and a large amount of image information of then utilizing computer to process to obtain, to replace the mankind to complete the task of Images Classification or identification.Word in image is an important sources of picture material, to people, provides brief and important information, so image text is identified in field of image recognition and occupies an important position.From the producing method of text, divide, image text can be divided into artificial text and scene text, and artificial text refers to the text being manually added on image, and scene text is the text itself existing on image.
Researcher both domestic and external had just started in the extraction of image Chinese word part and the research of identification before very early, and had obtained huge achievement.
(1) 20 twenties in century
Nineteen twenty-nine, first has proposed OCR(Optical Character Recognize Germany scientist Tausheck, optical character identification) concept, and applied for patent.
(2) 20 sixties in century
Since phase early 1960s, there is first generation OCR(optical character identification) product starts, and improves again through the development of more than 30 years, and the research of various OCR technology (comprising handwritten form) has all obtained the achievement attracting people's attention.
(3) 20 centuries 70, the eighties
IBM Corporation is the company that develops the earliest OCR product, and nineteen sixty-five, New York World's Fair Shang， IBM Corporation has put on display their OCR product---IBMl287.This product can only be identified block letter English alphabet, numeral and part symbol at that time, in addition, only to the font of appointment, can identify.Nineteen eighty-three, Toshiba has issued the system OCRV595 that can identify block letter japanese character, and recognition speed is 70～100 Chinese character/seconds, and discrimination reaches 99.5%.After this, the Study of recognition of handwritten form japanese character has been set about carrying out by Toshiba.
(4) development of Chinese character OCR technology
Chinese Character Recognition can be traced back to the sixties in 20th century the earliest.The Casey of 1966 Nian， IBM Corporations and Nagy have delivered first piece of paper about Chinese Character Recognition, have proposed to utilize simple template matching method to identify, and have identified 1000 printed Chinese characters.Since the 1970's, for Chinese character OCR system, the representational printed Chinese character monocase recognition system that can identify 2000 Chinese characters that has Toshiba's comprehensive study in 1977 to research and develop; The initial stage eighties, leading Chinese Character Recognition at that time be the printed Chinese characters recognition system of the wild electric research institute of Musashi research and development.
(5) research of China to OCR
China just starts the recognition technologies such as numeral, letter and symbol to be studied at 20 century 70s, to the research of Chinese Character Recognition, is to start to walk at the end of the seventies.1989 Nian， Tsing-Hua University have worked out Chinese OCR software---and the logical TH-OCR1.0 version of Tsing-Hua University's literary composition, this is the first Chinese OCR software systems of China.
From then on start, Chinese OCR system formally develops into the market stage from the experimental stage.Tsing-Hua University, after the printed Chinese character recognition software system issue of research and development, continues to have issued TH-OCR 92, thereafter, the TH-OCR 94 releasing in 1994, this is the block letter text recognition system of high performance Chinese-English mixing.The middle and later periods nineties 20th century, department of electronic engineering, tsinghua university is to Chinese character Study of recognition work, makes the research of Chinese Character Recognition obtain significant achievement in fields such as the Chinese character of block letter, hand script Chinese input equipment, offline handwriting, numbers and symbols identifications.
Summary of the invention
The object of the invention is in order to overcome the deficiencies in the prior art, and propose a kind of automatic mobile phone dialing and note transmission method based on character recognition.
The object of the invention is to be achieved through the following technical solutions.
1, the image information on the picked-up of the camera in end of scan 1 telephone number sample 2;
2, the telephone number on the interior telephone number identification software identification of end of scan 1 telephone number sample 2;
3, judge whether it is effective telephone-number; If invalid telephone number turns back to step 1;
4, the auto dialing software in end of scan 1 completes auto dialing, by communication network 4, connects user terminal 3 calls; Or the automatic transmission note software in end of scan 1 sends note by communication network 4 to user terminal 3.
Described end of scan 1 can be that inside is provided with camera, telephone number identification software, auto dialing software, automatically sends the mobile phone of note software.Described user terminal 3 can be the telephone plant with call function or note receiving function that connects communication network 4.
End of scan 1 identification telephone number adopts the character recognition technologies in graphical analysis, identifies.Not only can identify the 0-9 number of block letter, can also identify handwritten form 0-9 number.Its algorithm comprises two steps, character locating and character recognition.Character locating selectes approximate region by user, carries out the accurate orientation and segmentation of each character; And character recognition is identified each character of Accurate Segmentation.Character locating mainly comprises: zone location, preliminary treatment and three steps of Character segmentation, after the selected digital approximate region of user, according to clustering algorithm, accurately locate number region; Next, region is carried out to the pretreatment operation such as denoising, self-adaption binaryzation and morphology processing; Finally, realize the Accurate Segmentation of each character, single character region is indicated with minimum boundary rectangle frame.And the step of character recognition is according to the different in kind of character and difference.When the character 0-9 numeral that is block letter, set up a material database that comprises various printing type faces and big or small 0-9 numeral, use Tesseract-ocr algorithm to train, can realize the identification of printing digital.When the character 0-9 numeral that is handwritten form, set up equally the material database of various hand-written scripts and big or small 0-9 numeral, during training, extract statistics and architectural feature is trained svm classifier device; During identification, by extracting same feature, then can realize identification by the SVM model training.
Beneficial effect: the present invention is by character recognition technologies, identifies telephone number automatically fast, dial-in direct or transmission note.In speed, be better than manual dial or send note.
Accompanying drawing explanation
Fig. 1 is system architecture of the present invention.
Fig. 2 is flow chart of the present invention.
Fig. 3 is character recognition general flow chart of the present invention.
Fig. 4 is Handwritten Digit Recognition process flow diagram.
Embodiment
The present invention is described in detail by the following examples and by reference to the accompanying drawings.
If Fig. 1 is as shown in system architecture of the present invention, the present invention is based on being comprised of end of scan 1, telephone number sample 2, communication network 4 and user terminal 3.Telephone number on end of scan 1 scanning telephone number sample 2 carries out telephone number identification, and carries out auto dialing or generate note in end of scan 1, by communication network 4 and user terminal 3, carries out communication.
General end of scan 1 can be mobile phone, and its camera-enabled can scan in dual-purpose capture; Telephone number sample 2 can contain the telephone number of block letter, or the telephone number of handwritten form, as mail front cover, postal delivery parcel form; Communication network 4 can comprise the communication networks such as existing telecommunications, movement, UNICOM; User terminal 3 can be telephone terminal corresponding on telephone number sample 2.
If Fig. 2 is as shown in flow process of the present invention, the steps include:
1, start on end of scan 1 and distinguish software, open camera and read the image on telephone number sample 2;
2, the telephone number on identification telephone number sample 2;
3, end of scan 1 automatic poking is taken on the telephone or is sent note to user terminal 3.
As shown in character recognition general flow chart as of the present invention in Fig. 3, the process of end of scan 1 telephone number comprises two steps: character locating and character recognition.Character locating selectes approximate region by user, carries out the accurate orientation and segmentation of each character; And character recognition is identified each character of Accurate Segmentation.Character locating mainly comprises: zone location, preliminary treatment and three steps of Character segmentation, after the selected digital approximate region of user, according to clustering algorithm, accurately locate number region; Next, region is carried out to the pretreatment operation such as denoising, self-adaption binaryzation and morphology processing; Finally, realize the Accurate Segmentation of each character, single character region is indicated with minimum boundary rectangle frame.And the step of character recognition is according to the different in kind of character and difference.
When the character 0-9 numeral that is block letter, set up a material database that comprises various printing type faces and big or small 0-9 numeral, use Tesseract-ocr algorithm to train, can realize the identification of printing digital.When the character 0-9 numeral that is handwritten form, set up equally the material database of various hand-written scripts and big or small 0-9 numeral, during training, extract statistics and architectural feature is trained svm classifier device; During identification, by extracting same feature, then can realize identification by the SVM model training.
For handwritten word identification, be the flow chart of handwritten word identification as shown in Figure 4.
Handwritten word identifying is mainly divided into two large divisions: off-line learning and Real time identification, the latter is based upon on the former basis.Off-line learning, is mainly divided into three parts: (1) sample material database is set up; (2) feature extraction; (3) grader training.Material database sample is by handwritten word composition of sample, the handwritten word sample that the present invention is Arabic numerals.Extract statistics and the architectural feature of sample, form characteristics dictionary, and the input of the training using feature as grader, by the process optimization grader of Classification and Identification-verification-improvement grader, the grader of the handwritten word optimum that finally obtains classifying.Real time identification, mainly comprises three parts: (1) cuts apart single character; (2) extract feature; (3) Classification and Identification.By individual digit is cut apart, and extract statistics and architectural feature, first according to characteristics dictionary, feature is carried out to local optimum, then send into the grader that off-line learning partly generates, just can obtain classification results, be i.e. final number output.
The process of carrying out block letter identification with Tesseract-ocr algorithm is as follows:
(1) connected component analysis; (2) single character recognition; (3) repeatedly identification.By connected component analysis, find out single character.Next single character is identified, and the not good character of first recognition effect is cut apart and associated, for multiple possibility, the distance of application dictionary is calculated, and selects best possibility.Finally, by repeatedly identifying, self adaptation, using the good character of recognition effect as training sample, is again identified other bad characters, thereby is constructed adaptive grader.
Claims (6)
1. terminal recognition telephone number auto dialing or send the method for note, its feature comprises the following steps:
Step 1: the image information on the camera picked-up telephone number sample (2) in end of scan (1);
Step 2: the telephone number on the interior telephone number identification software identification telephone number sample of end of scan (1) (2);
Step 3: judge whether it is effective telephone-number; If invalid telephone number turns back to step 1;
Step 4: the auto dialing software in end of scan (1) completes auto dialing, connects user terminal (3) call by communication network (4); Or the automatic transmission note software in end of scan (1) sends note by communication network (4) to user terminal (3).
2. a kind of terminal recognition telephone number auto dialing according to claim 1 or send the method for note, is characterized in that described end of scan (1) can be that inside is provided with camera, telephone number identification software, auto dialing software, automatically sends the mobile phone of note software.
3. the method for a kind of terminal recognition telephone number auto dialing according to claim 1 or transmission note, is characterized in that described user terminal (3) can be the telephone plant with call function or note receiving function that connects communication network (4).
4. a kind of terminal recognition telephone number auto dialing according to claim 1 or send the method for note, is characterized in that described telephone number identification software includes following steps:
Step 1: character locating: according to by the image information on telephone number sample (2), select telephone number region, carry out, after preliminary treatment, carrying out Character segmentation;
Step 2: character recognition: the 0-9 numeral for block letter, adopts Tesseract-ocr algorithm to identify; For the 0-9 numeral of handwritten form, adopt statistics and the identification of architectural feature training svm classifier device.
5. the method for a kind of terminal recognition telephone number auto dialing according to claim 4 or transmission note, is characterized in that described Tesseract-ocr algorithm comprises the following steps: (1) connected component analysis; (2) single character recognition; (3) repeatedly identification.
6. the method for a kind of terminal recognition telephone number auto dialing according to claim 1 or transmission note, is characterized in that described statistics and architectural feature training svm classifier device algorithm comprise the following steps: off-line learning and Real time identification;
Described off-line learning comprises: (1) sample material database is set up; (2) feature extraction; (3) grader training;
Described Real time identification comprises: (1) cuts apart single character; (2) extract feature; (3) Classification and Identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310501585.0A CN103595861A (en) | 2013-10-23 | 2013-10-23 | Method for enabling terminal to identify phone number and automatically dial or send text message |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310501585.0A CN103595861A (en) | 2013-10-23 | 2013-10-23 | Method for enabling terminal to identify phone number and automatically dial or send text message |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103595861A true CN103595861A (en) | 2014-02-19 |
Family
ID=50085833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310501585.0A Pending CN103595861A (en) | 2013-10-23 | 2013-10-23 | Method for enabling terminal to identify phone number and automatically dial or send text message |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103595861A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104935726A (en) * | 2015-04-23 | 2015-09-23 | 刘晓建 | Method for rapidly initiating voice call according to figure including called information |
| CN106470258A (en) * | 2015-08-19 | 2017-03-01 | 腾讯科技（深圳）有限公司 | Auto dialing or the method and apparatus sending note |
| CN106487975A (en) * | 2016-10-27 | 2017-03-08 | 北京小米移动软件有限公司 | Dialing method and device |
| CN107273531A (en) * | 2017-06-28 | 2017-10-20 | 百度在线网络技术（北京）有限公司 | Telephone number classifying identification method, device, equipment and storage medium |
| CN107977594A (en) * | 2016-10-25 | 2018-05-01 | 深圳市寒武纪智能科技有限公司 | A kind of interactive robot and its method of telling a story |
| WO2019076368A1 (en) * | 2017-10-20 | 2019-04-25 | 捷开通讯(深圳)有限公司 | Method for processing typed in phone number, mobile terminal and storage medium |
| CN110210488A (en) * | 2019-06-14 | 2019-09-06 | 上海中通吉网络技术有限公司 | The recognition methods of bar code and cell-phone number and device on a kind of express waybill |
| CN110598684A (en) * | 2019-07-19 | 2019-12-20 | 珠海格力电器股份有限公司 | Method, system, terminal device and storage medium for identifying telephone number in picture |
| CN111985311A (en) * | 2020-07-08 | 2020-11-24 | 福建亿能达信息技术股份有限公司 | Method, device, equipment and medium for identifying mobile phone number |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1750016A (en) * | 2004-09-15 | 2006-03-22 | 北京中星微电子有限公司 | Optical character identifying treating method for mobile terminal with camera |
| US20100054585A1 (en) * | 2008-09-03 | 2010-03-04 | Jean-Pierre Guillou | Text localization for image and video OCR |
| CN102474902A (en) * | 2009-10-05 | 2012-05-23 | 索尼公司 | Mobile device visual input systems and methods |
| CN102761668A (en) * | 2012-07-13 | 2012-10-31 | 青岛海信传媒网络技术有限公司 | Method and device for dialing telephone call based on telephone number information in pictures |
| CN103327155A (en) * | 2013-06-13 | 2013-09-25 | 青岛百灵信息科技有限公司 | Novel mobile phone mode for courier |
-
2013
- 2013-10-23 CN CN201310501585.0A patent/CN103595861A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1750016A (en) * | 2004-09-15 | 2006-03-22 | 北京中星微电子有限公司 | Optical character identifying treating method for mobile terminal with camera |
| US20100054585A1 (en) * | 2008-09-03 | 2010-03-04 | Jean-Pierre Guillou | Text localization for image and video OCR |
| CN102474902A (en) * | 2009-10-05 | 2012-05-23 | 索尼公司 | Mobile device visual input systems and methods |
| CN102761668A (en) * | 2012-07-13 | 2012-10-31 | 青岛海信传媒网络技术有限公司 | Method and device for dialing telephone call based on telephone number information in pictures |
| CN103327155A (en) * | 2013-06-13 | 2013-09-25 | 青岛百灵信息科技有限公司 | Novel mobile phone mode for courier |
Non-Patent Citations (2)
| Title |
|---|
| VIEWCODE: "tesseract OCR Engine overview字符识别学习", 《BLOG.CSDN.NET/VIEWCODE/ARTICLE/DETAILS/7790065》, 26 July 2012 (2012-07-26) * |
| 孟凡栋: "基于主动学习svm的字符识别方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, 15 November 2008 (2008-11-15) * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104935726A (en) * | 2015-04-23 | 2015-09-23 | 刘晓建 | Method for rapidly initiating voice call according to figure including called information |
| CN106470258A (en) * | 2015-08-19 | 2017-03-01 | 腾讯科技（深圳）有限公司 | Auto dialing or the method and apparatus sending note |
| CN107977594A (en) * | 2016-10-25 | 2018-05-01 | 深圳市寒武纪智能科技有限公司 | A kind of interactive robot and its method of telling a story |
| CN106487975A (en) * | 2016-10-27 | 2017-03-08 | 北京小米移动软件有限公司 | Dialing method and device |
| CN107273531A (en) * | 2017-06-28 | 2017-10-20 | 百度在线网络技术（北京）有限公司 | Telephone number classifying identification method, device, equipment and storage medium |
| CN107273531B (en) * | 2017-06-28 | 2021-01-08 | 百度在线网络技术（北京）有限公司 | Telephone number classification identification method, device, equipment and storage medium |
| WO2019076368A1 (en) * | 2017-10-20 | 2019-04-25 | 捷开通讯(深圳)有限公司 | Method for processing typed in phone number, mobile terminal and storage medium |
| CN110210488A (en) * | 2019-06-14 | 2019-09-06 | 上海中通吉网络技术有限公司 | The recognition methods of bar code and cell-phone number and device on a kind of express waybill |
| CN110598684A (en) * | 2019-07-19 | 2019-12-20 | 珠海格力电器股份有限公司 | Method, system, terminal device and storage medium for identifying telephone number in picture |
| CN111985311A (en) * | 2020-07-08 | 2020-11-24 | 福建亿能达信息技术股份有限公司 | Method, device, equipment and medium for identifying mobile phone number |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103595861A (en) | Method for enabling terminal to identify phone number and automatically dial or send text message | |
| CN108664996B (en) | Ancient character recognition method and system based on deep learning | |
| CN109948510B (en) | Document image instance segmentation method and device | |
| Shahab et al. | ICDAR 2011 robust reading competition challenge 2: Reading text in scene images | |
| CN105630915A (en) | Method and device for classifying and storing pictures in mobile terminals | |
| CN111160352A (en) | Workpiece metal surface character recognition method and system based on image segmentation | |
| CN113313111B (en) | Text recognition method, device, equipment and medium | |
| CN102663435B (en) | Junk image filtering method based on semi-supervision | |
| CN102193620A (en) | Input method based on facial expression recognition | |
| CN105631039A (en) | Picture browsing method | |
| CN106815588B (en) | Junk picture filtering method and device | |
| CN102098449B (en) | A kind of method utilizing Mark Detection to carry out TV programme automatic inside segmentation | |
| CN113963147A (en) | Key information extraction method and system based on semantic segmentation | |
| CN110647956A (en) | Invoice information extraction method combined with two-dimensional code recognition | |
| He et al. | Context-aware mathematical expression recognition: An end-to-end framework and a benchmark | |
| CN103701987A (en) | Method for fast sending express delivery notice by intelligent mobile phone | |
| CN105246149B (en) | Geographical position identification method and device | |
| CN111882625A (en) | Method and device for generating dynamic graph, electronic equipment and storage medium | |
| CN106991438A (en) | One kind is based on the interactive facial image attribute labeling methods of MFC | |
| WO2022062027A1 (en) | Wine product positioning method and apparatus, wine product information management method and apparatus, and device and storage medium | |
| CN110858291A (en) | Character segmentation method and device | |
| CN115311664A (en) | Method, device, medium and equipment for identifying text type in image | |
| CN115830610A (en) | Multi-mode advertisement recognition method and system, electronic equipment and storage medium | |
| CN114022300A (en) | Social dynamic information publishing method and device, storage medium and electronic equipment | |
| CN112633283A (en) | Method and system for identifying and translating English mail address |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140219 |