Abstract:
Performance of biometric systems is highly dependent on the quality of the input samples captured by the sensing device. Although measures are taken for capturing high qu...Show MoreMetadata
Abstract:
Performance of biometric systems is highly dependent on the quality of the input samples captured by the sensing device. Although measures are taken for capturing high quality images, but the authentication system mandates the analysis of captured images for selection of precise data. The benefit of such an analysis are two-fold; it helps to identify the best sample, and is useful for improving the sensor design, user interface for sample collection and providing data interchange standards. In this work, we propose to analyse the quality of the sample data by using a two-stage fuzzy quality evaluation system. The proposed work has been demonstrated on the iris images using CASIA - 3.0 Interval, CASIA 4.0 Interval and IIT Delhi iris database. We evaluate the -quality of the images by classifying them into classes. The experimental results verify the efficacy of the proposed method.
Published in: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Date of Conference: 12-17 May 2019
Date Added to IEEE Xplore: 17 April 2019
ISBN Information: