TOP > クイック検索 > 外国特許検索 > PROGRAM, INFORMATION STORAGE MEDIUM, AND CHARACTER STRING RECOGNITION DEVICE

PROGRAM, INFORMATION STORAGE MEDIUM, AND CHARACTER STRING RECOGNITION DEVICE NEW

外国特許コード F170009112
整理番号 (S2016-0209-N0)
掲載日 2017年7月7日
出願国 世界知的所有権機関(WIPO)
国際出願番号 2016JP087558
国際公開番号 WO 2017104805
国際出願日 平成28年12月16日(2016.12.16)
国際公開日 平成29年6月22日(2017.6.22)
優先権データ
  • 特願2015-247379 (2015.12.18) JP
発明の名称 (英語) PROGRAM, INFORMATION STORAGE MEDIUM, AND CHARACTER STRING RECOGNITION DEVICE NEW
発明の概要(英語) This character string recognition device performs character recognition on a character pattern that is inputted via handwritten input, generates a plurality of recognition candidates, and outputs a score for each recognition candidate, then selects one recognition candidate from the plurality of recognition candidates and determines whether the selected recognition candidate is the correct answer or an incorrect answer. At least two of the following recognition processes are performed: a first recognition process, in which recognition character types are restricted by comparing with words and phrases; a second recognition process, in which recognition character types are not restricted and the context is evaluated; a third recognition process, in which recognition character types are not restricted and the context is not evaluated; and a fourth recognition process, in which, if there is a recognition candidate among the recognition candidates generated during the second recognition process that matches the correct answer, the score of that recognition candidate is outputted, and if there is not a recognition candidate that matches the correct answer, the lowest score is outputted. One recognition candidate is then selected on the basis of the scores for each generated recognition candidate.
特許請求の範囲(英語) [claim1]
1. Handwritten, doing character recognition vis-a-vis the letter pattern which is input, it forms plural recognition candidacy, certain of each recognition candidacy seems the recognition section which outputs the score which is shown and,
On the basis of the score of each recognition candidacy, the selective section which selects one recognition candidacy from plural recognition candidacy and,
Functioning point to the computer as the decision section which is decided whether the recognition candidacy which is selected is correct answer, or is error answer to,
As for the aforementioned recognition section,
Limiting recognition letter kind due to the collation with the phrase which is remembered in the memory section it forms recognition candidacy, it does not limit the recognition letter kind with the collation with the 1st recognition processing and the aforementioned phrase which output score at the same time certain of context does not seem appraising, it forms recognition candidacy, it does not limit the recognition letter kind with the collation with the 2nd recognition processing and the aforementioned phrase which output score at the same time certain of context does not seem without appraising it formed recognition candidacy, it was formed with 3rd recognition processing and the aforementioned 2nd recognition processing which output scoreWhen there is no recognition candidacy where it searches correct answer from recognition candidacy, when there is recognition candidacy which matches correct answer, it outputs the score of particular recognition candidacy, when there is no recognition candidacy which matches correct answer, it searches correct answer from the recognition candidacy which is formed with 4th recognition processing and the aforementioned 3rd recognition processing which output specified smallest score as score of recognition candidacy, when there is recognition candidacy which matches correct answer, outputs the score of particular recognition candidacy, matches correct answer, as a score of recognition candidacy specified smallestAmong the 5th recognition processing which output score, two recognition processing are executed at least,
As for the aforementioned selective section,
The program which features that one recognition candidacy is selected from plural recognition candidacy the description above on the basis of the score of each recognition candidacy which at least is formed with two recognition processing.
[claim2]
2. In claim 1,
As for the aforementioned recognition section,
The description above which includes the aforementioned 3rd recognition processing two recognition processing are executed at least,
As for the aforementioned selective section,
The program which features that one recognition candidacy is selected from plural recognition candidacy the description above which includes the aforementioned 3rd recognition processing on the basis of the score of each recognition candidacy which is formed at least with two recognition processing.
[claim3]
3. Claim in 1 or 2,
As for the aforementioned recognition section,
It executes with the aforementioned 2nd recognition processing and the aforementioned 3rd recognition processing and the aforementioned 4th recognition processing,
As for the aforementioned selective section,
The program which features that one recognition candidacy is selected from plural recognition candidacy on the basis of the score of each recognition candidacy which is formed with the aforementioned 2nd recognition processing, aforementioned 3rd recognition processing and aforementioned 4th recognition processing.
[claim4]
4. Claim either 1 or 3 in 1 sections,
As for the aforementioned selective section,
To adjust the score of each recognition candidacy which is formed with the aforementioned 1st recognition processing with 1st weight, to adjust the score of each recognition candidacy which is formed with the aforementioned 2nd recognition processing with 2nd weight, to adjust the score of each recognition candidacy which is formed with the aforementioned 3rd recognition processing with 3rd weight, to adjust the score of recognition candidacy with the aforementioned 4th recognition processing with 4th weight, to adjust the score of recognition candidacy with the aforementioned 5th recognition processing with 5th weight, from plural recognition candidacy, after the adjustingThe program which features that the recognition candidacy whose score is highest is selected.
[claim5]
5. Claim either 1 or 4 in 1 sections,
As for the aforementioned decision section,
If score of recognition candidacy of 1st rank and difference of score of recognition candidacy of 2nd rank there is recognition candidacy which matches correct answer in the recognition candidacy which is formed to the case where it is lower than specified threshold, with the aforementioned 3rd recognition processing, the recognition candidacy which is not the recognition candidacy which matches correct answer in the recognition candidacy where as correct answer it decides, if there is no recognition candidacy which matches correct answer in the recognition candidacy which is formed with the aforementioned 4th recognition processing as error answer it decides, is formed with the aforementioned 3rd recognition processing and at the same time is formed with the aforementioned 4th recognition processingIf there is recognition candidacy which matches correct answer in the program which features that it does not decide.
[claim6]
6. Being computer reading possible information storage medium, claim either 1 or 5 the information storage medium which features that the program of 1 sections is remembered.
[claim7]
7. Handwritten, doing character recognition vis-a-vis the letter pattern which is input, it forms plural recognition candidacy, certain of each recognition candidacy seems the recognition section which outputs the score which is shown and,
On the basis of the score of each recognition candidacy, the selective section which selects one recognition candidacy from plural recognition candidacy and,
The decision section which is decided whether the recognition candidacy which is selected is correct answer, or is error answer, implication,
As for the aforementioned recognition section,
Limiting recognition letter kind due to the collation with the phrase which is remembered in the memory section it forms recognition candidacy, it does not limit the recognition letter kind with the collation with the 1st recognition processing and the aforementioned phrase which output score at the same time certain of context does not seem appraising, it forms recognition candidacy, it does not limit the recognition letter kind with the collation with the 2nd recognition processing and the aforementioned phrase which output score at the same time certain of context does not seem without appraising it formed recognition candidacy, it was formed with 3rd recognition processing and the aforementioned 2nd recognition processing which output scoreWhen there is no recognition candidacy where it searches correct answer from recognition candidacy, when there is recognition candidacy which matches correct answer, it outputs the score of particular recognition candidacy, when there is no recognition candidacy which matches correct answer, it searches correct answer from the recognition candidacy which is formed with 4th recognition processing and the aforementioned 3rd recognition processing which output specified smallest score as score of recognition candidacy, when there is recognition candidacy which matches correct answer, outputs the score of particular recognition candidacy, matches correct answer, as a score of recognition candidacy specified smallestAmong the 5th recognition processing which output score, two recognition processing are executed at least,
As for the aforementioned selective section,
The character string recognition device which features that one recognition candidacy is selected from plural recognition candidacy the description above on the basis of the score of each recognition candidacy which at least is formed with two recognition processing.
  • 出願人(英語)
  • ※2012年7月以前掲載分については米国以外のすべての指定国
  • TOKYO UNIVERSITY OF AGRICULTURE AND TECHNOLOGY
  • 発明者(英語)
  • ZHU BILAN
  • NAKAGAWA MASAKI
国際特許分類(IPC)
指定国 (WO2017104805)
National States: AE AG AL AM AO AT AU AZ BA BB BG BH BN BR BW BY BZ CA CH CL CN CO CR CU CZ DE DJ DK DM DO DZ EC EE EG ES FI GB GD GE GH GM GT HN HR HU ID IL IN IR IS JP KE KG KH KN KP KR KW KZ LA LC LK LR LS LU LY MA MD ME MG MK MN MW MX MY MZ NA NG NI NO NZ OM PA PE PG PH PL PT QA RO RS RU RW SA SC SD SE SG SK SL SM ST SV SY TH TJ TM TN TR TT TZ UA UG US UZ VC VN ZA ZM ZW
ARIPO: BW GH GM KE LR LS MW MZ NA RW SD SL SZ TZ UG ZM ZW
EAPO: AM AZ BY KG KZ RU TJ TM
EPO: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR
OAPI: BF BJ CF CG CI CM GA GN GQ GW KM ML MR NE SN ST TD TG
※ 国立大学法人東京農工大学では、先端産学連携研究推進センターにおいて、知的財産の創出・権利化・活用に取り組んでいます。上記の特許・技術の内容および導入に興味・関心がありましたら、当センターまでお気軽にお問い合わせください。

PAGE TOP

close
close
close
close
close
close