Image Processing Group

Introduction

The basic theme of establishing Image Processing group is to explore different areas of Image processing and computer vision. The research in this group mainly focuses on Biometric Recognition Systems, that can be helpful to overcome current security problems.

All the group members have research publications on various reputed forums.

Group members

  • Muhammad Aurangzeb Khan, Head
  • Tariq Bashir
  • Tariq Mahmood
  • Syed Muzahir Abbas

Current Research

  • Recently our research group is working on iris recognition system. Most of the research is based on Segmentation and Normalization. In an iris recognition system Segmentation mean extracting the inner circular boundary (pupil) and outer circular boundary from an eye image. We proposed some new methods for both pupil segmentation and iris as well.
  • Iris normalization is a process to transform the circular reign to rectangular so that the matching algorithms can easily be implemented regardless pupil dilation and the different iris size caused by the different distance between the eye and video zoom factor.
  • Figure 4 and Figure 7 show some of the results of our proposed iris and pupil segmentation methods.
  • In face recognition, currently we are on making facial images illumination invariant, which is one of the most fundamental preprocessing step of any face recognition system. Since, achieving illumination invariance in the presence of large pose changes remains one of the most challenging aspects of automatic face recognition from low resolution imagery.

References

[1]: “Specifications of CASIA iris image database,” in Chinese Academy of Sciences, Available at: www.nlpr.ia.ac.cn/english/irds/irisdatabase.htm

[2]: “MMU iris image database,” in 2007 Available at: www.persona.mmu.edu.my.ccteo

Future aims

Our objective is to design a complete Multimodal Biometric Security System. Multimodal Biometric Technology uses more than one biometric identifier to compare the identity of the person. Therefore in the case of a system using more than one technology, if one fails to identify, the system can still use other to accurately identify the person.

Contact

Muhammad Aurangzeb Khan
Assistant Professor
Dept. of Electrical Engineering
Email: aurangzeb_niazi@comsats.eu.pk
 Phone: 0333-5268456