HOME    SCOPE    VENUE    COMMITTEE    GUIDELINES    AWARD    PAPERS     CONFERENCES
PAPERS
THE SIXTH INTERNATIONAL CONFERENCE ON FORENSIC COMPUTER SCIENCE

Print ISBN 978-85-65069-07-6 - Online ISBN 978-85-65069-05-2, pp 115-121
DOI: 10.5769/C2011012 and http://dx.doi.org/
10.5769/C2011012


OCR errors and their effects on computer forensics


By Mateus de Castro Polastro, and Nalvo Franco de Almeida Jr




To download this paper, click here.
ABSTRACT

The use of Optical Character Recognition (OCR) technology is an alternative when it is desired to search by keywords in image documents. In the field of computer forensics, this technology was recently incorporated into the version 3.1 of Access Data Forensic Toolkit (FTK). In this paper, we propose a method to evaluate the effects of OCR errors on information retrieval in Portuguese and English texts using this FTK feature. The method is described in detail and tools and public data were used. The experiments results showed that keywords search hits in OCRed texts are directly affected by the type of degradation suffered by the images. Success rates in searches of the English texts were around 95% and below 80% in Portuguese texts.


KEYWORDS

Computer forensics; OCR; image degradation; keyword search; FTK.

To return to the "Published Papers" main page, click here.