Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/45498
Title: The use of hyperspectral analysis for ink identification in handwritten documents
Authors: Morales Moreno,Aythami 
Ferrer, Miguel A. 
Diaz-Cabrera, Moises 
Carmona, Cristina 
Thomas, Gordon L.
UNESCO Clasification: 3307 Tecnología electrónica
220990 Tratamiento digital. Imágenes
Keywords: Hyperspectral imaging
Handwritten document analysis
Ink identification
Issue Date: 2014
Journal: Proceedings - International Carnahan Conference on Security Technology 
Conference: 48th Annual IEEE International Conference Carnahan on Security Technology (ICCST) 
Abstract: Hyperspectral analysis is employed in many different areas, such as medicine, environmental studies, security and forensics. Focusing on law enforcement, ink discrimination has become an important factor for the detection of fraudulent documents. This paper proposes an approach for ink analysis in handwritten documents and pen verification using hyperspectral analysis and Least Square SVM classification. The proposed method obtains immediate results in a non-contact way from the document or test sample. The first step is to determine the best possible lighting conditions. Then a detailed study is made of components and properties of the ink and pens used. This paper proposes a classification method based on the hyperspectral characteristics of the ink derived from its physical properties. Furthermore, a database of hyperspectral curves of several types of inks is made, which is used to obtain the characteristics of different inks. The proposed method for automated ink type identification is tested using 25 different pens and more than 1000 samples. The achieved discrimination between types of ink was 87.5%. The experimental protocol includes three different scenarios.
URI: http://hdl.handle.net/10553/45498
ISSN: 1071-6572
DOI: 10.1109/CCST.2014.6986980
Source: Proceedings - International Carnahan Conference on Security Technology [ISSN 1071-6572],v. 2014-October (6986980)
Appears in Collections:Actas de congresos
Show full item record

SCOPUSTM   
Citations

10
checked on Apr 18, 2021

WEB OF SCIENCETM
Citations

2
checked on Apr 18, 2021

Page view(s)

44
checked on Apr 18, 2021

Google ScholarTM

Check

Altmetric


Share



Export metadata



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