Deskewing

Scanned documents often become skewed (slanted) during scanning because of misfeeds or other alignment errors. Skew is the amount of rotation necessary to return an image to horizontal and vertical alignment. Skew is measured in degrees. Deskewing is a process whereby skew is removed by rotating an image by the same amount as its skew but in the opposite direction. This results in a horizontally and vertically aligned image where the text runs across the page rather than at an angle.

When an image is not aligned correctly, optical character recognition (OCR) is more difficult and becomes slower and less accurate. Deskewing the documents beforehand can make the OCR process faster and more accurate.

LEADTOOLS offers the following functionality for automatically straightening scanned documents:

L_DeskewBitmap

L_DeskewBitmapExt

The deskewing functions make it possible to::

Use the L_DeskewBitmapExt function when a lot of rotation may be necessary (as much as 45 degrees in either direction) or when it is desired to specify the maximum angle/resolution of deskewing.

Use the L_DeskewBitmap function to automatically straighten scanned documents of any kind. Images can be rotated by as much as 20 degrees in either direction to remove the skew. It can also be used when performing bank check scanning to obtain more accuracy than provided by the L_DeskewBitmapExt. To perform bank check scanning, send the following flags to the uFlags parameter (DSKW_USE_CHECK_DESKEW| DSKW_USE_CHECK_DESKEW_DETECT_LINES).

Different algorithms can be used when deskewing, as follows:

The following images demonstrate the deskewing of color, grayscale, and 1-bit images.

The original color image is shown in the following figure:

The deskewed color image is shown in the following figure:

The original grayscale image is shown in the following figure:

The deskewed grayscale image is shown in the following figure:

The original 1-bit image is shown in the following figure:

The deskewed 1-bit image is shown in the following figure: