An Overview of ARL's Multimodal Signatures Database and Web Interface

Report No. ARL-TR-4337
Authors: Kelly Bennett, Sam Wen, James Robertson
Date/Pages: December 2007; 30 pages
Abstract: The U.S. Army Research Laboratory (ARL) Multimodal Signatures Database (MMSDB) is a centralized collection of sensor data of various modalities that are co-located and co-registered. The signatures include ground and air vehicles, personnel, mortar, artillery, small arms gunfire from potential sniper weapons, explosives, and many other high value targets. This data is made available to Department of Defense (DoD) and DoD contractors, Intel Agencies, other government agencies (OGA) and academia for use in developing target detection, tracking, and classification algorithms and systems to protect our Soldiers. A platform independent Web interface disseminates the signatures to researchers and engineers within the scientific community. Hierarchical Data Format 5 (HDF5) signature models provide an excellent solution for the sharing of complex multimodal signature data for algorithmic development and database requirements. Many open source tools for viewing and plotting HDF5 signatures are available over the Web. Future development includes extending the Web interface into a portal system for accessing ARL algorithms and signatures and High Performance Computing (HPC) resources.
Distribution: Approved for public release
  Download Report ( 1.065 MBytes )
If you are visually impaired or need a physical copy of this report, please visit and contact DTIC.

Last Update / Reviewed: December 1, 2007