Title :
Skew Estimation and Correction of Text Using Bounding Box
Author :
Sarfraz, Muhammad ; Rasheed, Zeehasham
Author_Institution :
Dept. of Inf. Sci., Kuwait Univ., Safat
Abstract :
This paper, in addition to reporting some existing techniques, proposes a new technique for skew correction. It includes a novel document skew detection algorithm based on bounding box technique. The algorithm works quite efficiently for detecting the skew and then correcting it. The method has been experimented on various text documents and very promising results have been achieved given more than 97% accuracy. A comparative study has been reported to provide a detailed analysis of the proposed method together with some other existing methods in the literature.
Keywords :
document image processing; optical character recognition; text analysis; OCR; bounding box; document skew detection algorithm; image processing; skew correction; skew estimation; text correction; text documents; Character recognition; Computer graphics; Computer science; Detection algorithms; Feature extraction; Image processing; Information science; Optical character recognition software; Petroleum; Visualization; Image processing; character recognition; connected component; skew estimation; tilt correction;
Conference_Titel :
Computer Graphics, Imaging and Visualisation, 2008. CGIV '08. Fifth International Conference on
Conference_Location :
Penang
Print_ISBN :
978-0-7695-3359-9
DOI :
10.1109/CGIV.2008.10