A robust and fast skew detection algorithm for generic documents

Abstract A robust and fast skew detection algorithm based on hierarchical Hough transform is proposed. It is capable of detecting the skew angle for various document images, including technical articles, postal labels, handwritten text, forms, drawings and bar codes. The algorithm is robust even when black margins introduced by photocopying are present in the image and when the document is scanned at a low resolution of 50 dpi. The algorithm consists of two steps. In the first step we quickly extract the centroids of connected components using a graph data structure. Then, a hierarchical Hough transform (at two different angular resolutions) is applied to the selected centroids. The skew angle corresponds to the location of the highest peak in the Hough space. The performance of the algorithm is shown on a number of document images collected from various application domains. The algorithm is not very sensitive to algorithmic parameters. For an A4 size document image scanned at 50 dpi (typically 413 × 575 pixels), our algorithm is able to detect the skew angle with an accuracy of 0.1° in 0.4s of CPU time on a SunSparc 20 workstation.

[1]  Jake K. Aggarwal,et al.  Range image understanding , 1992, Image and Vision Computing.

[2]  Josef Kittler,et al.  A survey of the hough transform , 1988, Comput. Vis. Graph. Image Process..

[3]  S. Shapiro.  Properties of transforms for the detection of curves in noisy pictures , 1978 .

[4]  S.C. Hinds,et al.  A document skew detection method using run-length encoding and the Hough transform , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[5]  Robert M. Haralick,et al.  An automatic algorithm for text skew estimation in document images using recursive morphological transforms , 1994, Proceedings of 1st International Conference on Image Processing.

[6]  정성종,et al.  문서영상의 기울어짐 교정 알고리즘 ( An Algorithm for the Skew Normalization of Document Image ) , 1994 .

[7]  Norihiro Hagita,et al.  Automated entry system for printed documents , 1990, Pattern Recognit..

[8]  Yasuto Ishitani,et al.  Document skew detection based on local region complexity , 1993, Proceedings of 2nd International Conference on Document Analysis and Recognition (ICDAR '93).

[9]  Azriel Rosenfeld,et al.  A method of detecting the orientation of aligned components , 1986, Pattern Recognit. Lett..

[10]  Stephen D. Shapiro,et al.  Feature space transforms for curve detection , 1978, Pattern Recognition.

[11]  Lawrence O'Gorman,et al.  Document Image Analysis , 1996 .

[12]  Yuan Yan Tang,et al.  Document skew detection based on the fractal and least squares method , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[13]  Lawrence O'Gorman,et al.  The Document Spectrum for Page Layout Analysis , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Baozong Yuan,et al.  Isothetic polygon representation for contours , 1992, CVGIP Image Underst..

[15]  Dana H. Ballard,et al.  Generalizing the Hough transform to detect arbitrary shapes , 1981, Pattern Recognit..

[16]  Gilles Burel,et al.  Cooperation of multi-layer perceptrons for the estimation of skew angle in text document images , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[17]  Jiang Liu,et al.  An efficient method for the skew normalization of a document image , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol. III. Conference C: Image, Speech and Signal Analysis,.

[18]  Henry S. Baird,et al.  The skew angle of printed documents , 1995 .

[19]  Bin Yu.  Automatic understanding of symbol-connected diagrams , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.