DOCUMENT DIGITALIZATION ARCHITECTURE BY MULTI-MODEL DEEP LEARNING AND DOCUMENT IMAGE PROCESSING PROGRAM

• Main Author: HOSSAIN SHARIAR SHEIKH
• Format: Patent
• Language: English; Japanese
• Published: 08.07.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; PHYSICS

To convert a character string included in a document image to text data by a technique different from conventional optical character recognition.SOLUTION: An electronic document generation device comprises: a document image acquisition unit which acquires a document image obtained by making a picture of a document; a character recognition unit which subjects character strings included in the character image acquired by the document image acquisition unit to character recognition using a character string learning model which learned a correspondence between the character image and character strings included in the document image and outputs text data related to the character string; and an output unit which outputs the text data as a text of an electronic medium.SELECTED DRAWING: Figure 3 【課題】文書画像に含まれる文字列を、従来の光学文字認識とは異なる手法によって、テキストデータに変換することを目的とする。【解決手段】 電子文書生成装置は、文書を画像化した文書画像を取得する文書画像取得部と、文書画像と当該文書画像に含まれる文字列との対応関係を学習した文字列学習モデルを用いて、文書画像取得部に取得された文書画像に含まれる文字列を文字認識し、当該文字列に係るテキストデータを出力する文字列認識部と、テキストデータを電子媒体のテキストとして出力する出力部とを備える。【選択図】図３