Robust Perspective Rectification of Camera-Captured Document Images

Autor: Yusuke Takezawa, Salvatore Tabbone, Makoto Hasegawa
Přispěvatelé: Tokyo Denki University, Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)
Jazyk: angličtina
Rok vydání: 2017
Předmět:
Zdroj: 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)
2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Nov 2017, Kyoto, France. pp.27-32, ⟨10.1109/ICDAR.2017.345⟩
7th International Workshop on Camera-Based Document Analysis and Recognition, 14th IAPR International Conference on Document Analysis and Recognition, CBAR@ICDAR
7th International Workshop on Camera-Based Document Analysis and Recognition, 14th IAPR International Conference on Document Analysis and Recognition, CBAR@ICDAR, Nov 2017, Kyoto, France. pp.27-32, ⟨10.1109/ICDAR.2017.345⟩
CBDAR@ICDAR
DOI: 10.1109/ICDAR.2017.345⟩
Popis: International audience; A correction method for perspective distortions on document images is discussed. In documents, lines and line feeds give rise to many horizontal and vertical lines, then two vanishing points generated by these lines can be computed based on the Radon transform. However, some noisy lines are unrelated to vanishing points and can be removed using RANSAC algorithm in the Radon domain. In this perspective, we propose a unique scheme on how to apply RANSAC algorithm to our method. The distortion is rectified by the perspective mapping determined with two vanishing points. Experimental results show that our method can remove line noise effectively and is competitive compared to the state-of-art especially for camera-captured inside document images where contours are not clearly defined.
Databáze: OpenAIRE