期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：9
DOI：10.15680/IJIRCCE.2015. 0309097
出版社：S&S Publications
摘要：To ensure the actual presence of a real legitimate trait in contrast to a fake self-manufactured syntheticor reconstructed sample is a significant problem in biometric authentication, which requires the development of newand efficient protection measures. The iris and finger print are among the most promising biometric authentication thatcan precisely identify and analysis a person as their unique textures can be quickly extracted during the recognitionprocess. This biometric detection and authentication often deals with non-ideal scenarios such as blurred images, offangles,reflections, expression changes. In this project, we presented a fake detection method that can be used inmultiple biometric systems to detect different types of fraudulent access attempts. The objective of the proposed systemis to detect the fraudulent biometric samples. The proposed approach presents a very low degree of complexity, whichmakes it suitable for real-time applications, using Gabor wavelet for features extraction. The experimental results,obtained on publicly available data sets of fingerprint and iris, show that the proposed method is highly competitivecompared with other state-of-the-art approaches. For the preprocessing we used low pass filter. For the FeatureExtraction we have used Gabor wavelet. The classification is done by SVM which is an efficient classification.
