Fake biometric detection using image quality assessment: Application to iris, fingerprint recognition
Authors:
- 8 September 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 98-103
Abstract
Image Quality Assessment (IQA) is one of the statistical techniques used in image processing to determine whether the biometric sample is real or fake. The objective of the system is to enrich the biometric recognition security. This paper deals with two distinct measures of IQA. The first measure is Full-Reference(FR) IQA consists of a 2D image extracting different image quality features using a reference image which is filtered by a technique called Gaussian filtering. The second measure is No-Reference (NR) IQA used to estimate the quality level of an image. Eventually, 26 image quality features are exacted to minimize the degree of complexity. Quality of test sample implies to results of the following process of classification based on IQA. Presented paper briefly introduces the IQA theory and its measures. Results are documented for the selected real and fake pictures.Keywords
This publication has 9 references indexed in Scilit:
- Counter-measures to photo attacks in face recognition: A public database and a baseline2011
- RRED Indices: Reduced Reference Entropic Differencing for Image Quality AssessmentIEEE Transactions on Image Processing, 2011
- On the vulnerability of face verification systems to hill-climbing attacksPattern Recognition, 2009
- A no-reference sharpness metric sensitive to blur and noise2009
- First International Fingerprint Liveness Detection Competition—LivDet 20092009
- Biometric Template SecurityEURASIP Journal on Advances in Signal Processing, 2008
- Image manipulation detectionJournal of Electronic Imaging, 2006
- Feature selection: evaluation, application, and small sample performanceIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Floating search methods in feature selectionPattern Recognition Letters, 1994