Loading...
Please wait, while we are loading the content...
Similar Documents
Spoofing in iris recognition
| Content Provider | Indraprastha Institute of Information Technology, Delhi |
|---|---|
| Author | Gupta, Priyanshu Behera, Shipra |
| Abstract | A human iris contains rich textural information, and this serves as key information for biometric identifications. It is very unique and hence one of the most accurate modalities. Though iris patterns are very unique, ”liveness” and ”spoofing” techniques can be used to disguise or steal identities. Different studies have shown that eye surgeries, prescription lens, consumption of alcohol, affect the iris and hence performance of iris recognition. Similarly, pupil dilation using eye drops as well as interventions during cataract surgical procedures affect the iris patterns. The experiments performed on the databases collected as a part of this research, show how the accuracies reduce when matching between pre and post images.This leads to the risk of false acceptance or false rejection. We introduce alcohol and eye drop influence as one of the covariates of iris recognition and use the Iris Under Alcoholic Influence(IUAI) database and also under the influence of eye drops.In this research, we review the types of spoofing attacks that can be performed on an iris recognition system and how these deceptions can affect the recognition accuracies.We observe that for the iris to be identified and matched correctly, it does not even have to be live. One can easily fool a system even by using the printout of an iris image and even unintentional spoofing can be done using diseased iris images or images which have been affected due to presence of external factors like alcohol or eye drops. We also illustrate how contact lens plays a role in the spoofing. As a part of this research, we present the IIIT-Delhi Iris Spoofing Database (IISD), which contains printouts of the iris images. These printouts have been then scanned and captured using an iris sensor. This is a large dataset with 4848 iris images. This research also proposes different algorithms to measure useful area of iris images that can be helpful in pupil dilation and constriction cases. We then compare our results with an existing algorithm. |
| File Format | |
| Language | English |
| Publisher | IIIT Delhi |
| Access Restriction | Authorized |
| Subject Keyword | Iris recognition Spoofing Contact lens |
| Content Type | Text |
| Educational Degree | Bachelor of Technology (B.Tech.) |
| Resource Type | Thesis |
| Subject | Data processing & computer science |
