NEWSENSE: New Generation Information Processing Techniques for Imaging Sensors and Wireless Sensor Networks
Gokhan Uzunbas (alumni)
Erkin Tekeli (alumni)
EU Marie Curie Program
This project is funded by an EU Marie Curie International Reintegration Grant (IRG) to support Dr. Müjdat Çetin's early career development at Sabancı University. The research topics involved in this project have been motivated by recent developments in sensing systems, which have started inspiring revolutionary new applications. For example, environmental remote sensing using systems such as unmanned air vehicles enable the collection of rich data sets about a spatial region. Similarly, modern biomedical science has seen an explosion of exciting three-dimensional imaging modalities, including e.g. magnetic resonance imaging, and X-ray tomography. On the other hand, developments in micro electro-mechanical systems and wireless communications have led to the concept of wireless sensor networks. It is envisioned that these technologies will increasingly play a significant role in our lives, through applications such as noninvasive health monitoring, earthquake detection, infrastructure monitoring, precision agriculture, inventory monitoring, and smart homes. The promise of the rich data sets collected by these types of systems leads to the envisioning of a variety of applications, however successful operation of these systems requires the extraction of information from the chaos generated by such huge data sets. The objective of this research effort is to develop principled and practical new generation signal and information processing methods for such data-rich sensing systems. Research performed in this project is composed of three inter-related thematic components:
1) New, robust image formation algorithms based on data collected by imaging sensors, such as ultrasound and radar
2) Automatic image analysis techniques, with a focus on biomedical imaging and vehicle driver assistance
3) Scalable, distributed algorithms for data fusion and statistical inference based on wireless sensor network data.
The project supports a number of graduate students whose thesis topics are aligned with various aspects of the research themes mentioned above.
In the framework of the "New Generation Information Processing Techniques for Imaging Sensors and Wireless Sensor Networks" project, the following papers are published:
 - Marian Bartlett, Gwen Littlewort, Javier Movellan, Esra Vural, Kang Lee, Müjdat Çetin, and Aytül Erçil, "Datamining Spontaneous Facial Behavior with Automatic Expression Coding," Verbal and Nonverbal Features of Human-Human and Human-Machine Interaction, A. Esposito, N.G. Bourbakis, N. Avouris, and I. Hatzilygeroudis (eds.), Springer, Lecture Notes in Computer Science 5042, October 2008 (ISBN: 978-3540708711).