Title :
Fast and robust skew detection for scanned documents
Author :
Shi Na ; Pan Jinxiao
Author_Institution :
Dept. of Math., North Univ. of China, Taiyuan, China
Abstract :
A fast and robust skew detection for scanned document images based on the improved piecewise covering by parallelograms is proposed in this papper. Comparison with the original PCP algorithm, the skew detection of the proposed algorithm is more applicable on vertical textlines documents, and is not sensitive to the noise. The experiment results show that the computation time is short, while the detection accuracy is higher. The enhanced skew detection approach is efficient for horizontal and vertical textlines documents, and the complex documents with forms, tables and pictures.
Keywords :
document image processing; text analysis; detection accuracy; document image analysis; horizontal textline document; improved PCP algorithm; improved piecewise covering by parallelograms algorithm; robust skew detection; scanned document images; vertical textline document; Correlation; Estimation; Noise; Pattern recognition; Robustness; Slabs; Transforms; document image analysis; improved PCP; skew detection;
Conference_Titel :
Electronic and Mechanical Engineering and Information Technology (EMEIT), 2011 International Conference on
Conference_Location :
Harbin, Heilongjiang
Print_ISBN :
978-1-61284-087-1
DOI :
10.1109/EMEIT.2011.6023104
