Southeastern Section - 64th Annual Meeting (19–20 March 2015)

Paper No. 3
Presentation Time: 1:00 PM-5:00 PM

INITIAL GEOMATERIALS INVESTIGATION FOR DEVELOPING HYPERSPECTRAL REMOTE SENSING LIBRARIES FOR PERSONNEL RECOVERY: PRELIMINARY WORK ON THE LOST HUNTER SCENARIO


CRUMBAKER, Joahua, Geology & Environmental Earth Science, Miami University, 114 Shideler Hall, 250 S. Patterson Avenue, Oxford, OH 45056, BARRETT, Heather, Geology & Environmental Earth Science, Miami University, 114 Shideler Hall, Oxford, OH 45056 and KREKELER, Mark P.S., Department of Geology & Environmental Earth Science, Miami University-Hamilton, Hamilton, OH 45011, crumbaje@miamioh.edu

Hyperspectral remote sensing is increasingly becoming important in both military and law enfrorcement activities. In both law enforcement and military operations personnel recovery (PR) events are of top priority. Hyperspectral technology has the capacity to enable search area refinement and can be fused with other types of intelligence or information that will ultimately allow for additional situational awareness and enable those searching for personnel and strategic decision makers to develop a more informed and timely response. A critical gap in exploiting this technology is developing reference libraries for hyperspectral imaging with detailed geomaterial metadata. Without such data poor image interpretation and material misidentification in hyperspectral images can occur. Here we present preliminary results of a proof of concept investigation using a “lost hunter scenario”. This initial study is a geomaterials investigation of a polyester hunting vest, representing human material, along with confuser-geomaterials consisting of organic material and geo-materials with the intention of developing representative reflective spectra for the human-material and confuser-material combinations. Supporting data include scanning electron microscopy, Fourier transform infrared spectroscopy and reflective spectroscopy. This work demonstrates the potential and challenges of producing a first generation application of a human-material spectral library to hyperspectral imaging and its potential for remote sensing. The proposed approach shows some promise for future development of software plug-in modules for PR utilization. Challenges for such development include data collection and numerous integration issues.