US 12,400,463 B2Granted

Information processing apparatus, information processing method, and recording medium

AssigneeCanon Kabushiki Kaisha

InventorShoji Tanaka

Life of this patent

Estimating the term…

Loading patent…

Abstract

The present disclosure prevents learning a miscorrection made by a user on an OCR result. Provided is an information processing apparatus including: a character recognition unit configured to extract a character string containing at least one character by performing character recognition on an image obtained by scanning of a document; a display control unit configured to display the extracted character string on a display apparatus; an obtaining unit configured to receive a user correction on the extracted character string and obtaining a corrected character string based on the user correction; and a learning unit configured to learn a correction characteristic of a difference portion between the extracted character string and the corrected character string. The learning unit does not learn the correction characteristic of the difference portion in a case where a reliability level of the character recognition on the difference portion is equal to or above a predetermined threshold.

Claims11

Dependent claims

An information processing apparatus comprising:

at least one memory that stores instructions;
and
at least one processor that executes the instructions to:
extract a character string containing at least one character by performing character recognition on an image obtained by scanning of a document, wherein each of the at least one character included in the extracted character string has a reliability level of the character recognition;
display the extracted character string on a display apparatus;
obtain a corrected character string by replacing a first character included in the extracted character string with a second character based on a user's correction instruction, wherein the corrected character includes the second character;
and
learn a correction characteristic of a difference portion between the extracted character string and the corrected character string in a case where the reliability level of the character recognition performed for the first character corresponding to the difference portion is below a predetermined threshold, wherein
the correction characteristic of the difference portion is not learned in a case where the reliability level of the character recognition performed for the first character corresponding to the difference portion is equal to or above the predetermined threshold.

6 more elements

Dependencies are inferred from the claim text.

Facts

Application: 17/687,758Filed 7 Mar 2022
Priority: JP 2021-0372119 Mar 2021
Published: US 2022/0292857 A115 Sep 2022
Granted: 26 Aug 2025
Term ends: Not estimatedThe source records don’t support an estimate.

Classification5 codes

G06V 30/133
Detection or correction of errors, e.g. by rescanning the pattern: evaluation of quality of the acquired characters
G06V 30/127
Detection or correction of errors, e.g. by rescanning the pattern: with the intervention of an operator
H04N 1/00331
With an apparatus processing optically-read information: with an apparatus performing optical character recognition
G06V
Image and video recognition