METHOD AND SYSTEM FOR VERIFICATION DURING READING Russian patent published in 2015 - IPC G06K9/03 

Abstract RU 2571396 C2

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to a method and a system for determining incorrect recognition of symbols and a corresponding computer-readable medium. A method for determining incorrect recognition of symbols obtained from a text image recognition process includes: obtaining, by a processor, a set of uncertainly recognised symbols obtained from a text image recognition process, which includes an image of an uncertainly recognised symbol, a hypothesis on the uncertainly recognised symbol and the level of uncertainty associated with said hypothesis; calling, by the processor, a display device to display the image of the uncertainly recognised symbol on top of proofreading text, wherein the proofreading text differs from the image text; obtaining, by the processor, tagging data for the uncertainly recognised symbol and changing the level of uncertainty relating to the hypothesis on the uncertainly recognised symbol, based on the obtained tagging data in order to obtain a confirmed hypothesis on the uncertainly recognised symbol.

EFFECT: high accuracy of determining incorrectly recognised symbols.

24 cl, 12 dwg

Similar patents RU2571396C2

Title Year Author Number
POP-UP VERIFICATION PANEL 2014
  • Voronko Artem Nikolaevich
RU2665274C2
METHOD OF DETECTING NECESSITY OF STANDARD LEARNING FOR VERIFICATION OF RECOGNIZED TEXT 2014
  • Krivosheev Mikhail Viktorovich
  • Kolodkina Natalya Aleksandrovna
  • Makushev Aleksandr Sergeevich
RU2641225C2
VERIFICATION OF OPTICAL CHARACTER RECOGNITION RESULTS 2016
  • Kalyuzhny Aleksey Ivanovich
RU2634194C1
HANDWRITING RECOGNITION USING NEURAL NETWORKS 2020
  • Andrey Upshinskiy
RU2757713C1
METHOD AND SYSTEM FOR EXTRACTING DATA FROM IMAGES OF SEMISTRUCTURED DOCUMENTS 2015
  • Kostyukov Mikhail Valerievich
RU2613846C2
VIDEO CAPTURE IN DATA INPUT SCENARIO 2015
  • Isaev Andrey Anatolievich
RU2613734C1
IDENTIFICATION OF FIELDS ON AN IMAGE USING ARTIFICIAL INTELLIGENCE 2018
  • Kalenkov Maksim Petrovich
RU2695489C1
NEURAL NETWORK TRAINING BY MEANS OF SPECIALIZED LOSS FUNCTIONS 2018
  • Aleksey Alekseevich Zhuravlev
RU2707147C1
OPTICAL CHARACTER RECOGNITION USING SPECIALIZED CONFIDENCE FUNCTIONS, IMPLEMENTED ON THE BASIS OF NEURAL NETWORKS 2018
  • Aleksey Alekseevich Zhuravlev
RU2703270C1
IMAGE SEGMENTATION FOR DATA VERIFICATION 2015
  • Kanivets Diana Viktorovna
RU2605078C2

RU 2 571 396 C2

Authors

Pakhchanian Aram Bengurovich

Pogosskiy Michael Pavlovich

Dates

2015-12-20Published

2014-03-26Filed