Detection and Correction of Page Orientation in Monochrome Textual Document Image

While imaging a document with a scanner, it is quite possible that the document is fed into the scanner in portrait or landscape or portrait flipped or landscape flipped directions, in other words the direction of feed could be 0° or 90° or 180° or 270°. Additionally, a document suffers a skew also. Hence, correction of directional orientation followed by skew correction becomes the first phase of processing in Document Image Analysis. In this paper, we provide a solution to these problems by employing a non rotational approach in two stages. A macro level skew correction(if exists) is made to the document image in portrait or landscape directions. The direction of orientation detection and correction is done in the later stage. The performance on English text documents is presented based on extensive experimentation.