This paper proposes an iris image quality and segmentation accuracy evaluation method for video-based iris recognition systems, operating in unconstrained environment. Proposed approach consists of two stages of video quality evaluation that allows improving the iris recognition rates in non-ideal or no cooperative situations; where the first stage discards the low quality eye frames, while the second stage discards frames with low quality iris segmentation providing properly segmented iris frames to carry out the unconstrained iris recognition task. Although proposed scheme was evaluated using the Daugman algorithm, it may be used with several other iris recognition systems operating in constrained environments.Evaluation results show that the performance of conventional iris recognition system, using the proposed scheme, reduces the equal error rate (EER) value in about 12.2%.
Key words: Biometric systems, video-based iris recognition, image quality measure, non-ideal iris frames, segmentation accuracy evaluation, unconstrained environments.
Copyright © 2023 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0