Skip to main content
zenodoopen

Brazilian Forensic Letter Database

<p>The BFL database&nbsp;composed of 315&nbsp;writers, three samples per writer, summing up 945 images. The motivation for using such database is the growing interest of the Brazilian forensic experts as well as the Brazilian&nbsp;Federal Police in writer verification/identification .</p><p>The samples were provided by undergraduate students inthree different sessions during one month. The texts werecollected on an A4 white sheet of paper with no pen-drawbaseline and then scanned in gray level with 300 dpi (3760 X&nbsp;2448). Each writer was allowed to use his/her own pen, which&nbsp;means that several different pens were used. The text is concise (131 words) and complete in the sense that it contains all&nbsp;characters (letters and numerals) and certain character combinations of interest. This makes it suitable for text-dependent&nbsp;writer identification as well. The image below shows the letter contents and a sample of the database.</p><p>&nbsp;</p><p>The BFL Database may be used for non-commercial research provided you acknowledge the source of the image by citing the following paper in publications about your research:</p><ul><li>C. Freitas, L. S. Oliveira, R. Sabourin, F. Bortolozzi <strong>Brazilian Forensic Letter Database</strong>, 11th International Workshop on Frontiers on Handwriting Recognition (IWFHR-11), Montreal, QC, Canada, August 19-21 2008. (<a href="http://www.cenparmi.concordia.ca/ICFHR2008/Proceedings/papers/cr1011.pdf">pdf</a>).</li></ul><h4>Related Papers:</h4><ul><li>Hanusiak, R. K., Justino, E., Oliviera, L. S., Sabourin, R. Writer Verification Using Texture-based Features, International Journal of Document Analysis and Recognition, 15(3):213-226, 2012. (<a href="http://www.inf.ufpr.br/lesoliveira/download/IJDAR2011.pdf">pdf</a>)</li><li>Bertolini, D.,Oliveira, L. S., Justino, E., Sabourin, R., Texture-based Descriptors for Writer Identification and Verification, Expert Systems With Applications, 40(6):2069-2080, 2013. (<a href="http://www.inf.ufpr.br/lesoliveira/download/ESWA2013a.pdf">pdf</a>).</li></ul>

ShareScore

36/100

Overall dataset sharing score

Score breakdown

These five areas show where the dataset supports — or may limit — practical reuse.

Stewardship
4
Harmonization
4
Access
16
Reuse readiness
8
Engagement
4

ScienceDex for faculty

Your academic CV and NIH biosketch

Make your CV in your school's format and your NIH biosketch from it, then download both as Word files.