Document Dewarping

Here we provide an on-line demo of our page dewarping algorithm that is based on the assumption that the original page contained only straight lines that were approximately equally spaced and sized. This is often true for book pages. Compare the publication

Adrian Ulges, Christoph H. Lampert, Thomas M. Breuel: Document Image Dewarping using Robust Estimation of Curled Text Lines, International Conference on Document Analysis and Recognition (ICDAR), pages 1001-1005, 2005.

Note that the algorithm has several limitations:

it does not work if the assumption is not true for (parts of) the document image, i.e. for headlines, paragraphs with spacing, figures, etc.
we (try to) use the largest box of text within the image only
we assume that the image is given in the correct orientation
the maximum angle that any part of the text line can deviate from the horizontal is currently set to 0.5 radians (about 30 degrees)
if there are large spaces between words, the line tracker sometimes tries to make two lines out of one; this results in visually strong distortions

You can either submit an image through the form interface, or you can submit it programmatically through HTTP.

Form Interface

If you do not have an image at hand or want to try some of our images, try one of these (note that results are cached, so this is faster than using a new image):

Programmatic Interface

To submit your image programmatically, you can simply POST to this URL; the image should be a parameter named "imagefile".

From the command line, you can do this using:

curl -D header.out -F 'imagefile=@input.jpg;type=image/jpeg' http://quito.informatik.uni-kl.de/dewarp/dewarp.php > output.jpg

You can also do this easily using the HTTP implementation in your favorite programming language (C#, Python, Java, Perl, etc.).