Journal
OPTICS AND LASERS IN ENGINEERING
Volume 98, Issue -, Pages 168-175Publisher
ELSEVIER SCI LTD
DOI: 10.1016/j.optlaseng.2017.06.022
Keywords
Hyperspectral imaging; Banknote; Principal component analysis; Classification
Categories
Funding
- COLE-EDB
- NTU-PhotoniTech RCA
- AcRF Tier 1, Ministry of Education (MOE), Singapore [RG162/15]
Ask authors/readers for more resources
The use of counterfeit banknotes increases crime rates and cripples the economy. New countermeasures are required to stop counterfeiters who use advancing technologies with criminal intent. Many countries started adopting polymer banknotes to replace paper notes, as polymer notes are more durable and have better quality. The research on authenticating such banknotes is of much interest to the forensic investigators. Hyperspectral imaging can be employed to build a spectral library of polymer notes, which can then be used for classification to authenticate these notes. This is however not widely reported and has become a research interest in forensic identification. This paper focuses on the use of hyperspectral imaging on polymer notes to build spectral libraries, using a pushbroom hyperspectral imager which has been previously reported. As an initial study, a spectral library will be built from three arbitrarily chosen regions of interest of five circulated genuine polymer notes. Principal component analysis is used for dimension reduction and to convert the information in the spectral library to principal components. A 99% confidence ellipse is formed around the cluster of principal component scores of each class and then used as classification criteria. The potential of the adopted methodology is demonstrated by the classification of the imaged regions as training samples. (C) 2017 Elsevier Ltd. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available