New document skew detection and correction algorithm
Huichuan Duan · Jisuanji gongcheng yu sheji · 2007
During document scanning, the input document images inevitably introduce skew, while the algorithms for layout analysis and character recognition are sensitive to page skew. A document skew detection and correction based on GA algorithms is proposed. Firstly, the bounding box of the skew document is gotten, by using the characters of the bounding box, then GA is used to detect the skew angle. The experimental results show that the proposed algorithm guarantee the accuracy for skew document.