TWI456411B - Method of autometically detecting error by using language model for print - Google Patents

Method of autometically detecting error by using language model for print Download PDF

Info

Publication number
TWI456411B
TWI456411B TW100115891A TW100115891A TWI456411B TW I456411 B TWI456411 B TW I456411B TW 100115891 A TW100115891 A TW 100115891A TW 100115891 A TW100115891 A TW 100115891A TW I456411 B TWI456411 B TW I456411B
Authority
TW
Taiwan
Prior art keywords
printed
text
content
error
language model
Prior art date
Application number
TW100115891A
Other languages
Chinese (zh)
Other versions
TW201245981A (en
Inventor
Ping Cheng Lin
Jui Feng Yeh
Original Assignee
Univ Far East
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Univ Far East filed Critical Univ Far East
Priority to TW100115891A priority Critical patent/TWI456411B/en
Publication of TW201245981A publication Critical patent/TW201245981A/en
Application granted granted Critical
Publication of TWI456411B publication Critical patent/TWI456411B/en

Links

Landscapes

  • Accessory Devices And Overall Control Thereof (AREA)
  • Record Information Processing For Printing (AREA)

Claims (2)

一種印表機之利用語言模型自動偵測錯誤之方法，包含：對一列印文件之一列印內容進行一文字抽取之動作，以從具有一文字部分與一圖片部分之該列印內容中抽取出該文字部分；針對該列印文件之該列印內容之文字部分進行一斷詞之動作，其中該斷詞之動作係以字詞為一基本單元，將該列印內容之該文字部分之文字串依據該基本單元做斷詞；利用一包含一自然語言模式之語言模型針對該列印文件之該列印內容之該文字部分進行一錯誤偵測，其中進行該錯誤偵測後，若偵測該列印文件之該列印內容之該文字部分正確，則該列印文件直接被列印出來；反之，若偵測該列印文件之該列印內容之該文字部分有一錯誤，則產生一警訊或自動修正該錯誤。 A method for automatically detecting an error by using a language model of a printer comprises: performing a text extraction operation on printing a content of one of the printed documents to extract the text from the printed content having a text portion and a picture portion And performing a word breaking operation on the text portion of the printed content of the printed document, wherein the action of the word breaking is based on the word as a basic unit, and the text string of the text portion of the printed content is based on The basic unit performs a word break; and uses a language model including a natural language mode to perform an error detection on the text portion of the printed content of the printed file, wherein the error detection is performed if the column is detected If the text portion of the printed content of the printed document is correct, the printed document is directly printed; otherwise, if the text portion of the printed content of the printed document is detected to have an error, a warning is generated. Or fix the error automatically. 如申請專利範圍第1項所述之印表機之利用語言模型自動偵測錯誤之方法，其中該文字抽取之動作係將該列印內容之該文字部分與該圖片部分分離，並抽取出純文字的部份。 The method for automatically detecting an error by using a language model of the printer according to Item 1 of the patent application scope, wherein the action of extracting the text separates the text portion of the printed content from the image portion, and extracts pure Part of the text.
TW100115891A 2011-05-06 2011-05-06 Method of autometically detecting error by using language model for print TWI456411B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW100115891A TWI456411B (en) 2011-05-06 2011-05-06 Method of autometically detecting error by using language model for print

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW100115891A TWI456411B (en) 2011-05-06 2011-05-06 Method of autometically detecting error by using language model for print

Publications (2)

Publication Number Publication Date
TW201245981A TW201245981A (en) 2012-11-16
TWI456411B true TWI456411B (en) 2014-10-11

Family

ID=48094443

Family Applications (1)

Application Number Title Priority Date Filing Date
TW100115891A TWI456411B (en) 2011-05-06 2011-05-06 Method of autometically detecting error by using language model for print

Country Status (1)

Country Link
TW (1) TWI456411B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1163827A (en) * 1996-01-12 1997-11-05 三星电子株式会社 Printing system for automatically detecting paper length and control method thereof
TWI280924B (en) * 2004-08-27 2007-05-11 Seiko Epson Corp Printer and printer control method
TW200846939A (en) * 2006-12-05 2008-12-01 Microsoft Corp Web-based collocation error proofing
TW200944383A (en) * 2008-04-30 2009-11-01 Cal Comp Electronics & Comm Co Ltd Device and system of ink box detection device and system and ink box detection method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1163827A (en) * 1996-01-12 1997-11-05 三星电子株式会社 Printing system for automatically detecting paper length and control method thereof
TWI280924B (en) * 2004-08-27 2007-05-11 Seiko Epson Corp Printer and printer control method
TW200846939A (en) * 2006-12-05 2008-12-01 Microsoft Corp Web-based collocation error proofing
TW200944383A (en) * 2008-04-30 2009-11-01 Cal Comp Electronics & Comm Co Ltd Device and system of ink box detection device and system and ink box detection method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
T *

Also Published As

Publication number Publication date
TW201245981A (en) 2012-11-16

Similar Documents

Publication Publication Date Title
EP2444889A3 (en) Print processing apparatus, print processing apparatus control method, and storage medium
WO2008039927A3 (en) Typing candidate generating method for enhancing typing efficiency
JP2008276493A5 (en)
JP2009177726A5 (en)
TW200729041A (en) System, method and program for generating data for printing invisible information, and method for making physical media on which invisible information is printed
JP2009225181A5 (en)
JP2013197785A5 (en)
ES2555180R1 (en) Method implemented by computer to synchronize annotations between a printed document and an electronic document, computer readable support and corresponding system
WO2009091210A3 (en) Method of providing e-book service utilizing text information, and a system therefor
EP2230593A3 (en) Job management apparatus, control method, and program
EP2131566A3 (en) Image processing apparatus and image processing method
EP2746989A3 (en) Document processing device, image processing apparatus, document processing method and computer program product
JP2007535771A5 (en)
JP2012517635A5 (en) How to arrange print jobs into expressions that can be divided
EP2755143A3 (en) Automated language detection for domain names
JP2008186451A5 (en)
TWI456411B (en) Method of autometically detecting error by using language model for print
JP2009175450A5 (en)
JP2012221095A5 (en)
JP2012221372A5 (en) Form, form processing apparatus, and form processing method
JP2010240844A5 (en)
JP2015005161A5 (en) Control device and control method of control device
JP2007249692A5 (en)
JP2012173999A5 (en)
JP2008278307A5 (en)

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees