Publication: Handwritten Text Line Segmentation by Shredding Text into its Lines

Authors: Anguelos Nicolaou, Basilis Gatos

Published in: ICDAR 2009 (Barcelona)

Download pdf

In this paper, we propose a novel technique to segment handwritten document images into text lines by shredding their surface with local minima tracers. Our approach is based on the topological assumption that for each text line, there exists a path from one side of the image to the other that traverses only one text line. We first blur the image and then use tracers to follow the white-most and black-most paths from left to right as well as from right to left in order to shred the image into text line areas. We experimentally tested the proposed methodology and got promising results comparable to state of the art text line segmentation techniques.

Naive implementation of LLA:

This mfile is a script compatible with octave/matlab and provided under the GPL. Keep in mind that this mfile is not the full method but an implementation of the LLA segmentation as reported in the article with some minor differences.

Download seglines.m (octave/matlab source)

Usage assuming sample1.tif to be an image (preferably binarized and around 1-3 MP):

[res B LA LC LLA LLC lh]=seglines(In);
imshow(res./max(res(:)),'Colormap',[[0,0,0]' colormap('lines')' ]');
title('RES a simple result (no per component line assignment)');
imshow(LLC./max(LLC(:)),'Colormap',[[0,0,0]' colormap('lines')' ]');
title('LLC labeled line centers');
imshow(LLA./max(LLA(:)),'Colormap',[[0,0,0]' colormap('lines')' ]');
title('LLA labeled line areas');
title('LC line centers');
title('LA line areas');
title('B blured image');

bibtex entry :
title={Handwritten text line segmentation by shredding text into its lines},
author={Nicolaou, Anguelos and Gatos, Basilios},
booktitle={Document Analysis and Recognition, 2009. ICDAR'09. 10th International Conference on},

Creative Commons License
Any work in this page other than source code or program binaries is licensed under a Creative Commons Attribution 4.0 International License. When applicable atribution should be in the form of a citation.