Abstract
Fragile watermarking has been proposed as a means of adding additional security or functionality to biometric systems, particularly for authentication and tamper detection. In this paper we describe an experimental study investigating the effect of watermarking iris images with a particular class of fragile algorithm, reversible watermarking, on the ability to correctly perform iris recognition. We investigate two scenarios, matching watermarked images to unmodified images, and matching watermarked images to watermarked images. We found that the payload capacity and watermarking scheme chosen can have a significant effect on the performance seen. At high embedding rates most algorithms cause significant reduction in recognition perfor-
mance. However, in many cases, for low embedding rates, an increase in recognition performance was found.
mance. However, in many cases, for low embedding rates, an increase in recognition performance was found.
| Original language | English |
|---|---|
| Number of pages | 6 |
| Publication status | Unpublished - 2012 |
| Event | The IEEE Fifth International Conference on Biometrics: Theory, Applications and Systems (BTAS 2012) - Washington, DC, United States Duration: 23 Sept 2012 → 27 Sept 2012 |
Conference
| Conference | The IEEE Fifth International Conference on Biometrics: Theory, Applications and Systems (BTAS 2012) |
|---|---|
| Country/Territory | United States |
| City | Washington, DC |
| Period | 23/09/12 → 27/09/12 |