A general algorithm for document skew angle estimation

Abstract

We propose a new method for calculating the skew angle of scanned document images. The method is designed to be insensitive to document layout, line spacing, font, graphicshmages and, most importantly, the language or script of the document. This is achieved by examining the Fourier spectra of blocks of the document image for peak pairs corresponding to the angle of skew. From a histogram compiled over all blocks in the document image the correct skew angle can be determined to within approximately 0.5 3 regardless of document script, even when the image contains considerable graphical information.

Texture for script identification

“…This is due to character and word spacings and the strength of such a line varies with changes in language and script type. Peake and Tan expand on this method, breaking the document image into a number of small blocks, and calculating the dominant direction of each such block by finding the Fourier spectrum maxima [25].…”

“…The exact skew angle is then calculated by taking the average of all values within a specified range of this approximate. There is some evidence that this technique is invariant to document layout and will still function even in the presence of images and other noise [25]. A number of other techniques for the estimation of the skew angle have also been proposed [26], [27], [28].…”

An unconstrained handwriting recognition system

“…In the strong belief that the preprocessing stage of an OCR system is at least equally important as the recognition part, we paid special attention to all the preprocessing stages. Thus, our system, presented in Fig.1, copes with the following tasks: -Skew angle estimation and correction: several methods have been proposed to solve the problem of skew detection [8,9]. Analytical reviews regarding these methods can be found in [10,11].…”

“…The horizontal histogram of a properly oriented document page presents the maximum peaks and the most intense alternations between peaks and dips than any other histogram-by-angle of the same page [8,9]. In order to apply it, the WVD and its maximum intensity curve is calculated for each histogram.…”

Skew angle estimation in document processing using Cohen’s class distributions

“…The Projection Pro®le technique has been applied by Pavlidis and Zhou (1991) and Ciardiello et al (1988), while its combination with the Fourier transform is a possibility presented by Postl (1986) and Peake and Tan (1997). Shihari and Govindaraju (1989) and Le et al (1994) applied the Hough Transform.…”