Please use this identifier to cite or link to this item: https://hdl.handle.net/1959.11/9654
Title: Validating the use of Handwriting as a Biometric and its Forensic Analysis
Contributor(s): Leedham, Graham  (author); Pervouchine, Vladimir (author)
Publication Date: 2005
Handle Link: https://hdl.handle.net/1959.11/9654
Abstract: For centuries handwriting (mostly in the form of a handwritten signature or mark) has been used as a means of authorizing documents. The assumption is that handwriting is a unique biometric identifying the writer and signatory. When the authenticity of handwriting or a signature is questioned Forensic Document Examiners are called in to determine whether the handwriting/signature is genuine, forged, or disguised. In recent years the uniqueness or individuality of handwriting has been questioned and the techniques used by Forensic Document Examiners have been found lacking in scientific rigour. In this paper we describe some of the recent work to scientifically determine the individuality of handwriting and validate some of the techniques used by Forensic Document Examiners. The Letter-level microfeatures used by document examiners are studied in detail for three letters and one two-letter grapheme from 165 writers. The results indicate that some of the features used by document examiners do haw the ability to discriminate between writers. The features extracted from graphemes or groups of letters have more discriminative power than features from individual letters. This study shows that handwriting analysis call identify an unknown writer against a set of known writers with a high degree of certainty. It is not possible to determine whether handwriting is a biometric which uniquely identifies an individual.
Publication Type: Book Chapter
Source of Publication: Proceedings of the International Workshop on Document Analysis, p. 175-192
Publisher: Allied Publishers Pvt Ltd
Place of Publication: Chennai, India
ISBN: 8177647849
9788177647846
Field of Research (FOR): 080109 Pattern Recognition and Data Mining
080104 Computer Vision
080106 Image Processing
HERDC Category Description: B1 Chapter in a Scholarly Book
Statistics to Oct 2018: Visitors: 291
Views: 373
Downloads: 0
Appears in Collections:Book Chapter
School of Science and Technology

Files in This Item:
3 files
File Description SizeFormat 
Show full item record

Page view(s)

136
checked on May 3, 2019
Google Media

Google ScholarTM

Check


Items in Research UNE are protected by copyright, with all rights reserved, unless otherwise indicated.