Document skew detection based on local region complexity
作者： Y. Ishitani
DOI: 10.1109/ICDAR.1993.395784
关键词: Pixel 、 Artificial intelligence 、 Orientation (computer vision) 、 Pattern recognition 、 Computer science 、 Skew 、 Image (mathematics)
摘要: A new method is proposed for detecting skew in document images which contain a mixture of text areas, photographs, figures, charts, and tables. Two basic ideas are introduced the method. One idea that parameter used detection to discern orientation lines images. This based on image complexity obtained from number transitions white black pixels or vice versa. The other detected local regions only expected. Such extracted automatically angle defined as overall skew. Document has been measured experiments with an error 0.12 degrees average all test documents. >
sci-hub.se 参考文章(3)
Teruo Akiyama, Norihiro Hagita, Automated entry system for printed documents Pattern Recognition. ,vol. 23, pp. 1141- 1154 ,(1990) , 10.1016/0031-3203(90)90112-X
S. Tsujimoto, H. Asada, Major components of a complete text reading system Proceedings of the IEEE. ,vol. 80, pp. 1133- 1149 ,(1992) , 10.1109/5.156475
Y. Nakano, Y. Shima, H. Fujisawa, J. Higashino, M. Fujinawa, An algorithm for the skew normalization of document image international conference on pattern recognition. pp. 8- 13 ,(1990) , 10.1109/ICPR.1990.119320