Face Biometrics for Security: Long-Range and Surveillance

  • Online

This tutorial introduces issues in long-range facial image acquisition  and measures for image quality and their usage, as well as subsequent  challenges for face recognition. The first several modules on image  acquisition for face recognition discuss concerns related to lighting,  sensors, and lenses, which impact short-range biometrics, but are more  pronounced in long-range biometrics. We then go on to introduce the  design of controlled experiments for long-range face recognition and why  they are needed. With our experiments, we go on to show some of the  weather and atmospheric effects that occur for long-range imaging, with  numerous examples. Next, we address measurements of "system quality"  including image-quality measures and their use in the prediction of face  recognition algorithm performance. That module introduces the concept  of post-recognition score analysis and techniques for analyzing  different "equality" measures. The last two modules of this tutorial  explore long-range face recognition directly. Facial feature detection  is an important prerequisite for face recognition, and we look at two  different approaches for accurately accomplishing this for long-range  scenarios. Finally, we address the very difficult problem of blur "both  motion and atmospheric" including common sources in acquisition and  algorithms to mitigate its effects.

What you will learn:

  • Review motivating concerns over surveillance with biometrics in difficult environments
  • Consider lighting, sensor and weather and atmospheric impacts

Related courses:

Who should attend: Electrical engineer, Systems engineer, Hardware engineer, Design engineer, Product engineer,  Communication engineer

Instructors

Walter J. Scheirer

Walter J. Scheirer Photo

Dr. Walter J. Scheirer received his Ph.D. from the University of  Colorado and his M.S. and B.A. degrees from Lehigh University, focused  in security. He is currently the Director of Research and Development at  Securics, Inc. He has published and lectured widely on a variety of  security related topics, including face detection for surveillance,  privacy enhancing technologies, biometric fusion, secure cryptographic  channels, network intrusion detection and Unix system security

Terrance Boult

Terrance Boult Photo

Dr. Boult is a Professor of Computer Science at the University of  Colorado where his role includes working with faculty and local  companies to develop and transfer technology in the Springs area. He  studied at Columbia University, earning his MS and PhD degrees in  computer science. Dr. Boult has published over 150 papers and holds 7  patents, with 9 patents pending. He publishes in Computer  Vision/Robotics as well as in Networking and Mobile Computing.

Publication Year: 2010

ISBN: 978-1-4244-6214-8


Face Biometrics for Security: Long-Range and Surveillance
  • Course Provider: Educational Activities
  • Course Number: EDP197
  • Duration (Hours): 1
  • Credits: 0.1 CEU/ 1 PDH