LAVI Iris Database DB1
Laboratory of Computer Vision
Department of Electrical and Computer Engineering, University
of São Paulo (USP)
São
Carlos, Brazil
Based on the “consensual reflex,” we developed a device for the acquisition
of iris images, as shown in Fig. 1. This device performs different and
independent tasks on each eye. The right eye receives visible light stimuli
(white light) at computer-controlled specific time intervals, while the left eye
image is captured and digitized into a video sequence under NIR illumination.
The two sides of the device related to the left and right eyes are isolated from
each other. The video sequence obtained under NIR illumination is synchronized
with the “visible light” pulses (a white light LED was employed) applied to the
other eye.
-
Were captured 5 videos from 124 different individuals, totaling 620 videos.
-
The videos are named as: PN_nv.mpeg where PN is the
Person Number (video/eye class) and nv is the video number (class
element).
- Download of the Iris Video DB1 Database: (each
.rar file size is from 70 MB to 190 MB)
Please, cite the following paper when using this
database:
DA COSTA, R. M.; GONZAGA, A.;
Dynamic Features for Iris Recognition,
IEEE Transactions on Systems, Man, and
Cybernetics, Part B: Cybernetics,
Aug. 2012, Vol. 42 , Issue: 4,
pp. 1072 – 1082., doi:
10.1109/TSMCB.2012.2186125