A device and a method for deep learning based character recognition
The present invention provides a deep learning-based character recognition device and a deep learning-based character recognition method, which may overlap and divide a corresponding part of an image to be converted into text data, and reassemble characters by learning the relationship between the c...
Gespeichert in:
Hauptverfasser: | , |
---|---|
Format: | Patent |
Sprache: | eng ; kor |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The present invention provides a deep learning-based character recognition device and a deep learning-based character recognition method, which may overlap and divide a corresponding part of an image to be converted into text data, and reassemble characters by learning the relationship between the characters and images included in the divided part. The deep learning-based character recognition device may include an image data pre-processing module, an image segmentation module, a character extraction and optimization recognition module, and a character recognition and natural language processing module.
본 발명은 문자 데이터로 전환하고자 하는 이미지의 해당 부분을 중첩 분할하고, 분할된 부분에 포함된 문자와 이미지 사이의 연관 관계를 학습하여 문자를 재조립하는 딥러닝 기반 문자 인지 장치 및 딥러닝 기반 문자 인지 방법을 제안한다. 상기 딥러닝 기반 문자 인지 장치는, 이미지 데이터 전처리 모듈, 이미지 분할모듈, 문자 추출 & 최적화인식 모듈 및 글자인식 & 자연어처리 모듈을 포함한다. |
---|