Iris and face image sequences collected at varying distances and quality. Collected at Clarkson University. Funded by the Department of Homeland Security (DHS) Science and Technology (S&T) Directorate in cooperation with the National Science Foundation. To be used in IREX II. (Link to IREX II: http://iris.nist.gov/irexII/)
Note: Dataset will be released to public upon completion of IREX II.
This biometric dataset collection contains images of two biometric modalities:
- Iris
- Face
The purpose of this data collection protocol is to develop a multi-biometric database which include iris and face quality images for varying distances and a range of quality. The experimental setup has been designed for acquiring images for the second phase of ‘DHS/NSF Dynamic Decisional Fusion Face/Iris Database’ grant.
The main factors studied are the impact on quality and performance in face and iris images over five distances: baseline, 5’, 7’, 11’, 15’, and 25’. Subjects were asked to walk through a ‘portal’ at 7’ and 15’ to introduce motion blur. Subjects also enrolled in a baseline iris/face system (MBARK system with OKI and Olympus cameras). To produce non-uniformity in the dataset, variability in the data was introduced through six factors (resolution, ambient (or external) illumination, out-of-focus blur, multiple faces, gaze/pose angles, motion blur). Additionally, an occlusion factor was introduced for iris images. For each factor, the data acquisition equipment (Dalsa and Canon cameras) was controlled to achieve high, medium, and low quality iris and face images.
Over 175 subjects are included with at least two visits each. The dataset also includes soft biometrics such as height and weight, for subjects of different age groups, ethnicity and gender with variable number of sessions/subject.
| Biometric Modality | Device | Nr. of subjects | Nr. of files/subject |
|---|---|---|---|
| Iris | OKI IRISPASS EQ5016A | ~175 | 8 |
| Face | Olympus C-5060 camera | ~175 | 20 |
| Iris | Dalsa 4M30 infrared camera | ~175 | ~3000 |
| Face | Canon VIXIA HFS100, HD Camcorder | ~175 | %7e3000 |
When using biometric collection, please cite:
Stephanie Schuckers, Paulo Meyer Lopez, Peter Johnson, Nadezhda Sazonova, Fang Hua, Rick Lazarick, Chris Miles, Elham Talbassi, Edward Sazonov, Arun Ross, Lawrence Hornak, Quality–Face / Iris Research Ensemble (Q-FIRE) Dataset Overview, Technical Report, Clarkson University, Dept of Electrical and Computer Engineering, 2010.
Stephanie Schuckers, Paulo Meyer Lopez, Peter Johnson, Nadezhda Sazonova, Fang Hua, Rick Lazarick, Chris Miles, Elham Talbassi, Edward Sazonov, Arun Ross, Lawrence Hornak, Quality–Face / Iris Research Ensemble (Q-FIRE) Data Collection Steps, Technical Report, Clarkson University, Dept of Electrical and Computer Engineering, 2010.