Handwriting identification using global and local features for forensic purposes

Greening, C.M.; Sagar, V.K.; Leedham, C.G.

doi:10.1049/cp:19950511

Cited by 3 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…tion. An overview of the complete system is shown in Figure 2.5 (Greening et al, 1995a(Greening et al, , 1995b. Feature extraction blocks are emphasised.…”

Section: Computational Featuresmentioning

confidence: 99%

Discriminative power of features used by forensic document examiners in the analysis of handwriting

Pervouchine¹

View full text Add to dashboard Cite

One of the tasks of a forensic document examiner is to analyse a questioned handwritten document in order to determine its author. The purpose of such analysis is to determine whether the document is genuine, forged, or disguised. Document analysis plays an important investigative and forensic role in many types of crime and is used as evidence of authorship or non-authorship in a court of law. The methods used by forensic experts in document analysis are intuitively reasonable and have been derived from experience. The experts look into characteristics of handwriting (features) that are consistent in a person's normal writing. These characteristics allow the experts to compare different handwritten samples and draw a conclusion about the authorship of the questioned sample. However, the methods used are based on experience rather then scientific facts. There is a lack of scientific basis to support their methods of analysis. The aim of this thesis is to answer the question whether the methods used by forensic document examiners for handwriting analysis allow us to distinguish different writers with reasonable accuracy when the documents are written in normal (not deliberately changed) writing. In this work a number of features commonly used by forensic document examiners were automatically extracted from greyscale images of samples of characters "d", "y", "f", "t", and grapheme "th". The discriminative power of the features was assessed. The feature extraction was performed in several iterations; in each iteration new features were added, some of the previous features were removed, and extraction algorithms were improved. Automatic feature extraction accuracy of 85% was achieved in the initial experiments and later was improved to 94% by using a novel method of handwriting stroke approximation (character skeletonisation). A new method of skeletonisation for approximation of handwriting strokes on character images was developed. The method was designed to extract the skeleton which 7 ATTENTION: The Singapore Copyright Act applies to the use of this document. Nanyang Technological University Library Some interactive tools were developed as a by-product of this study to facilitate visual comparison of handwriting samples and aid in development of the skeletonisation method. Study of document examiner features in forged and disguised handwriting is a necessary next step to provide a scientific support to forensic analysis of handwriting, and is suggested as a possible direction for further research.

show abstract

“…tion. An overview of the complete system is shown in Figure 2.5 (Greening et al, 1995a(Greening et al, , 1995b. Feature extraction blocks are emphasised.…”

Section: Computational Featuresmentioning

confidence: 99%

Discriminative power of features used by forensic document examiners in the analysis of handwriting

Pervouchine¹

View full text Add to dashboard Cite

show abstract

An image processing and pattern recognition framework for forensic document analysis

Greening

Sagar

Proceedings the Institute of Electrical and Electronics Engineers. 29th Annual 1995 International Carnahan Conference on Securi

Self Cite

View full text Add to dashboard Cite

This paper presents solme research work that has been undertaken in the area of software tools for forensic document analysis. The purpose of research into this area is to study the methods used by human experts in the forensic analysis and examination of questioned hand-written documents and to formulate computer tools to assist in this task. The computer tools needed will require image processing andl pattern recognition techniques.

show abstract

Slant manipulation and character segmentation for forensic document examination

Sagar

Chong²,

Leedham³

et al.

Proceedings of Digital Processing Applications (TENCON '96)

View full text Add to dashboard Cite

Handwriting identification using global and local features for forensic purposes

Cited by 3 publications

References 0 publications

Discriminative power of features used by forensic document examiners in the analysis of handwriting

Discriminative power of features used by forensic document examiners in the analysis of handwriting

An image processing and pattern recognition framework for forensic document analysis

Slant manipulation and character segmentation for forensic document examination

Contact Info

Product

Resources

About