NIST logo
*

Video-based Automatic System for Iris Recognition

Description/Summary:

logo_VASIR_with_Text_final_1

Video-based Automatic System for Iris Recognition (VASIR) is a NIST implemented iris recognition algorithm designed to work on both conventional iris images and iris images collected at a distance.

VASIR is fully automated system of the video-based iris recognition for less-constrained videos. All videos were captured while a person walked through a portal at a distance. The system was developed to address the challenge of recognizing a person in less-than-optimal environments, coping with high and low still-image and video-sequence quality.

VASIR_overview_for_website

As shown in figure above, VASIR can be principally categorized into seven parts of components:
  • Part1: Image Acquisition,
  • Part2: Eye Region Detection/Extraction,
  • Part3: Quality Measures and Selecting the Best Quality Image,
  • Part4: Segmentation,
  • Part5: Normalization,
  • Part6: Feature Extraction and Encode, and
  • Part7: Similarity Matching Templates.

VASIR has the capacity to automatically detect the eye region and subsequently to automatically select the best quality iris image from video. After the process, VASIR carries out a comprehensive comparison analysis of biometric samples which in turn yields a state-of-the-art individual verification.

 

VASIR also accommodates multiple scenarios:

  • distant-video to distant-video (VvsV) ,
  • distant-video to classical-still (VvsS), and
  • classical-still to classical-still (SvsS) iris recognition.

In VvsV matching, the extracted iris region of distant-video-sequences is matched to other sequences from the same video sequence or from a different video sequence of the same person. VvsS means that the video-sequence captured at a distance is compared to classical-still-images, captured by a different camera. In SvsS matching, a classical-still-image is matched against other classical-still-images of the same person that were captured by the same device. 

VASIR's performance and practical feasibility have been evaluated using the Multiple Biometric Grand Challenge (MBGC) dataset.

Uses:

 Currently, the VASIR source code beta version 1.0 and its user guide are available. This source code is not the completed version. Note that you may get a known warning or memory related messages.

We plan to update each component of VASIR incrementally, after evaluating its performance.

Details

Type of software: Open source research software for automated iris recognition.

Authors:

Yooyoung Lee, Ross J. Micheals, P. Jonathon Phillips, James J. Filliben

System/Platform:

VASIR was developed using the following software.

- Microsoft Windows XP, Microsoft Windows Server 2008, Windows 7
- Microsoft Visual Studio Professional 2005/2008
- Qt Framework latest version
- OpenCV Library 1.0

Detailed requirements can be found in the user guide.

Licensing info:

This software is released by NIST as a service and is expressly provided "AS IS." NIST MAKES NO WARRANTY OF ANY KIND, EXPRESS, IMPLIED OR STATUTORY, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTY OF MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE, NONINFRINGEMENT AND DATA ACCURACY. NIST DOES NOT REPRESENT OR WARRANT THAT THE OPERATION OF THE SOFTWARE WILL BE UNINTERRUPTED OR ERROR-FREE, OR THAT ANY DEFECTS WILL BE CORRECTED.

Disclaimer

Certain trade names and company products are mentioned in the text or identified. In no case does such identification imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the products are necessarily the best available for the purpose.

Contact

Ross J. Micheals / Yooyoung Lee

vasir@<NOSPAM>nist.gov