Recursive XY-cut

From HandWiki
Revision as of 15:31, 6 March 2021 by imported>Unex (simplify)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

The recursive X-Y cut is a top-down page segmentation technique that decomposes a document image recursively into a set of rectangular blocks. The algorithm works by projecting the document bitmap (i.e. summing up all the pixels in a line) to the sides of the document page. By this method, a white space density graph is produced, with peaks for vertical or horizontal whitespace lines. These peaks define the cuts of the document and are used top-down to segment the document into smaller pieces.

References

  • G. Nagy and S. Seth. Hierarchical representation of optically scanned documents. In Proc. of the 17th Conf. on Pattern Recognition, pp. 347–349, 1984.