 Handwriting is a personal biometric that is thought to be unique to an individual and hence can be used to identify an individual. As a result, the use of handwritten signatures has been, from early history, a legally accepted means of authenticating various documents. In addition, in some criminal cases, analysis of handwriting is often performed by forensic document examiners to determine the authorship of a questioned document. Forensic analysis is usually seeking to determine whether the questioned document is genuine (written by the person who claims to have written it), forged (written by someone trying to pass the document as being written by someone else) or disguised (written by a person in a different handwriting style to their own in order to deny writing the document at a future time).
The techniques used by forensic document examiners to reach their expert opinion are based on a set of techniques that are standardised and well documented in various texts. However, unlike other forms of forensic analysis (such as DNA testing and chemical analysis of blood and tissue samples) the analysis of handwriting does not have a strong scientific base.
 Recent court cases (for example United States vs Starzecypzel, 1995) have raised questions about the scientific basis of the expert opinion offered by forensic document examiners. In order for forensic document examination to retain its credibility and legal acceptability as a science there must be some statistically sound basis for the analysis. In addition, such scientific information would be useful for the future development of automatic off-line signature verification systems.
This research is concerned with the computer analysis of visible features in handwriting to determine which features (if any) are difficult to forge or disguise. This project extends Graham's recent research on tools to assist forensic document analysis (see recently completed research projects below) by studying the production and resulting visual features of handwriting to identify forgeries and disguised writing.
| Principal Investigator: |
Dr. Graham Leedham |
| Research student: |
Vladimir Pervouchine (PhD awarded 2006) |
| External collaborators: |
Prof Sargur Srihari, CEDAR, SUNY, Buffalo, USA and |
| |
Dr Ujjwal Bhattacharya, Indian Statistical Institute, Calcutta
|
Associated publications:
- Graham Leedham, Vladimir Pervouchine and Wei Kei Tan, Quantitative letter-level extraction and analysis of features used by document examiners, Journal of Forensic Document Examiners Vol. 16, pp. 21 -40, 2006.
- Vladimir Pervouchine & Graham Leedham, Extraction and analysis of document examiner features from vector skeletons of grapheme `th', Lecture Notes in Computer Science 3872, Springer, Horst Bunke & Lawrence Spitz (Eds.), pp.196-207, 2006.
 Powerpoint presentation for this paper
- Vladimir Pervouchine & Graham Leedham, The validity and usefulness of forensic document examiner features for automatic writer identification, Pattern Recognition, (In Press, February 2006).
- Philip Jonathan Sutanto, Graham Leedham & Vladimir Pervouchine, Study of the consistency of some discriminatory features used by document examiners in the analysis of handwritten letter 'a', Proceedings of the 7th International Conference of Document Analysis and Recognition, ISBN 0-7695-1960-1, pp.1091-1095, Edinburgh, Scotland, 3-6 August 2003.
- Graham Leedham & Sumit Chachra, Writer identification using innovative binarised features of handwritten numerals, Proceedings of the 7th International Conference of Document Analysis and Recognition, ISBN 0-7695-1960-1, pp. 413-417, Edinburgh, Scotland, 3-6 August 2003.
- A K Khor, C G Leedham & D J Maskell, Collision and impact force computation for virtual reality applications, Proceedings of TENCON 2003 - The IEEE Region 10 Technical Conference on Convergent Technologies for the Asia-Pacific, Bangalore, India, 14-17 October 2003.
- Sargur N Srihari & Graham Leedham, A survey of computer methods in forensic document examination, Proceedings of the 11th International Graphonomics Society Conference, ISBN 0-9746365-0-9, pp. 278-281, Scottsdale, USA, 2-5 November 2003.
- Graham Leedham, Vladimir Pervouchine, Tan Wei Kei, & Arun Jacob, Automatic quantitative letter-level extraction of features used by document examiners, Proceedings of the 11th International Graphonomics Society Conference, ISBN 0-9746365-0-9, pp. 291-294, Scottsdale, USA, 2-5 November 2003.
- Graham Leedham, Vladimir Pervouchine, Wei Kei Tan, & Arun Jacob, Assessment of the stability and usefulness of some handwriting features used by document examiners to identify authorship, Proceedings of the 11th International Graphonomics Society Conference, ISBN 0-9746365-0-9, pp. 316-319, Scottsdale, USA, 2-5 November 2003.
- Vladimir Pervouchine, Graham Leedham, and Konstantin Melikhov, Handwritten character skeletonisation for forensic document analysis, Proceedings of the 20th Annual ACM Symposium on Applied Computing, Vol. 1 pp. 754-758, Santa Fe, New Mexico, 13-17 March 2005.
- Vladimir Pervouchine & Graham Leedham, Three-stage handwriting stroke extraction method with hidden loop recovery, Proceedings of the 8th International Conference on Document Analysis and Recognition, pp. 307-311, Seoul, Korea, 29 Aug - 1 Sep 2005.
- Vladimir Pervouchine & Graham Leedham, Document examiner feature extraction: thinned vs. skeletonised handwriting images, Proceedings of The IEEE Region 10 Technical Conference, (TENCON05), Melbourne, Australia, 21-24 November 2005.
Powerpoint presentation for this paper
- Vladimir Pervouchine & Graham Leedham, Extraction and Analysis of Document Examiner Features from Vector Skeletons of Grapheme `th', Proceedings of the 7th IAPR Workshop on Document Analysis Systems, pp.196-207, 13-15 February 2006, Nelson, New Zealand.
- Graham Leedham & Vladimir Pervouchine, Validating the use of Handwriting as a Biometric and its Forensic Analysis, Invited lecture at the International Workshop on Document Analysis (IWDA-2005), Indian Statistical Institute, Calcutta, 8-11 March 2005. Published in Document Analysis, Eds. Umapada Pal, Swapan K Parui & B. B. Chaudhuri, Allied Publishing Ltd, Chennai, pp. 175-192, ISBN 81-7784-9, 2005.
Powerpoint presentation for this paper
|
