Iris recognition has the potential to be extremely accurate, but it is highly dependent on the quality of the input data. Iris occlusion, off-axis gaze, blurred images, and iris rotation are common problems that can make recognizing individuals more difficult. A fraction of the IREX III dataset consists of poor quality images that reduced overall performance in the evaluation. The purpose of this failure analysis is to identify the causes of poor sample quality in the dataset and to provide best practice recommendations for how to improve the quality of captured samples. This analysis began by placing iris images that were a common source of error for several matching algorithms into a special failure set. A thorough manual inspection was then performed on each image to identify specific cause(s) of failure.
Citation: NIST Interagency/Internal Report (NISTIR) - 7853Report Number:
NIST Pub Series: NIST Interagency/Internal Report (NISTIR)
Pub Type: NIST Pubs
biometrics, iris recognition, sample quality, IREX