The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.
The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.
Patent No.:
Date of Patent:
Aug. 03, 2021
Filed:
Apr. 25, 2019
International Business Machines Corporation, Armonk, NY (US);
Kohichi Kamijoh, Tokyo, JP;
Akira Shimizu, Kanagawa, JP;
Dai Sakai, Kawasaki, JP;
Yuta Ishii, Tokyo, JP;
International Business Machines Corporation, Amronk, NY (US);
Abstract
A computer-implemented method for increasing a recognition rate of an optical character recognition (OCR) system is provided. The method includes preprocessing by receiving an image, and extracting all vertical lines from the image. The method includes adding vertical lines at character areas of the image, extracting all horizontal lines from the image, and creating an unlined image removing all the vertical/horizontal lines from the image. The method further includes determining a border of a vertical direction of the unlined image based on the total of pixels of rows in each column, and adding vertical/horizontal auxiliary lines between characters of the unlined image. The method also includes postprocessing by receiving garbled words of OCR output, removing noise after morphologically analyzing, replacing garbled letters with correct ones based on a frequent edit operation, and outputting the correct word, weighting results of image distance calculations based on machine learning.