DiamondBack: Sensor Fusion and Feature-Based Human/Animal Classification for UGS
Small Business Information
Scientific Systems Company, Inc.
500 West Cummings Park, Suite 3000, Woburn, MA, 01801-6562
AbstractThe Department of Homeland Security (DHS) needs a low cost reliable automated system to detect illegal border crossings. Current seismic UGS systems use cadence-based intrusion detection algorithms and are easily confused between humans and animals. The ensuing false alarms reduce the trustworthiness of the system and lead to unnecessary actions which may be costly. Scientific Systems Company Inc. (SSCI) through its Phase I research has developed and tested novel signal processing and classification algorithms to robustly discriminate between humans, animals and vehicles. SSCI identified that foot contact characteristics have a significant impact on the time-frequency characterization of recorded seismic signals. SCCI is teaming with Crane Wireless Systems (Crane WMS) and Applied Research Associates (ARA) for the Phase II work. SSCI will provide the signal processing and classification expertise, and Crane Wireless Monitoring Solutions and Applied Research Associates, both leading provides of Seismic Unattended Ground Sensors, will provide the hardware platform and systems support for porting the classification algorithms onto the UGS sensors. The UGS sensors will be field tested with input from the DHS for performance evaluation. SSCI intends to transition and commercialize the UGS technology using licensing arrangements for the signal processing, classification and power management algorithms.
* information listed above is at the time of submission.