Project title:

CRI: Imaging and Non-Imaging Sensor Laboratory for Urban Disaster Management

Funded by:

National Science Foundation, Award #0551719

Investigators:

Ertem Tuncel(Principal Investigator)
Amit Roy-Chowdhury (Co-Principal Investigator)
Srikanth Krishnamurthy (Co-Principal Investigator)
Vana Kalogeraki (Co-Principal Investigator)
Sundararajan V (Co-Principal Investigator)

Students:

Luis Gonzalez-Argueta
Bi Song

Abstract:

This project, instrumenting a laboratory with a range of sensory devices to enable research on urban disaster management, focuses on collecting, integrating, and analyzing data from video and other sensors. The lab consists of imaging and non-imaging sensors that facilitate interdisciplinary research involving sensor networks, video processing, communications, and networking. Such facility enables collaborations leading to the development of algorithms whose novelty lies in the tight integration of the output of the sensors for maximum information exploitation. This integration should lead to a more holistic approach to understanding the disaster environment and enable providing better decision support systems for emergency personnel. The research focuses on four problems:

-Networking protocols,
-Video data processing (mainly motion tracking and behavior modeling),
-Data compression, and
-Data fusion at different stages.

Broader Impact: The project develops a framework that integrates existing methods of disaster management with new sensors and communication technology to provide a more current and up-to-date information of the disaster environment. The work addresses the prevalent problem of disaster management by enhancing the type and processing of information to the emergency personnel, hence the expectation of high societal impact.

 Publications:

• Multi-Target Tracking Through Opportunistic Camera Control In A Resource Constrained Multimodal Sensor Network, J. Nayak, L. Gonzalez-Argueta, B. Song, E. Tuncel, A. Roy-Chowdhury, IEEE/ACM Intl. Conf. on Distributed Smart Cameras, 2008.
• A Multi-Terminal Model-Based Video Compression Algorithm, B. Song, A. Roy-Chowdhury, E. Tuncel, IEEE Intl. Conf. on Image Processing, 2006.
• Towards A Multi-Terminal Video Compression Algorithm Using Epipolar Geometry, B. Song, O. Bursalioglu, E. Tuncel, A. Roy-Chowdhury, IEEE Intl. Conf. on Acoustics, Speech and Signal Processing, 2006. (Best Student Paper Award)

Project Results:

Multi-Target Tracking Through Opportunistic Camera Control In A Resource Constrained Multimodal Sensor Network

50

 Sensor Modules (IRIS MOTES)

• 2.4 GHz IEEE 802.15.4, Wireless Measurement System capable of 250 kbps, High Data Rate Radio
• Flash 512K bytes
• Transmission range  > 50 m

50

MTS300 Sensor Boards

• Light, Temperature, Acoustic and Sounder Sensor Board

5

Base Station (Gateway MIB510)

• Communication to Sensor Modules

1

Dell Laptop Computer

• Intel Celeron 1.6GHz, 1GB RAM

4

Wireless Pan/Tilt/Zoom Cameras