Multispectral Data Fusion for Moving Target Detection

Period of Performance: 12/18/1998 - 06/18/1999

$70K

Phase 1 SBIR

Recipient Firm

Physical Optics Corp.
1845 West 205th Street Array
Torrance, CA 90501
Principal Investigator

Research Topics

Abstract

Day/night reconnaissance and surveillance are critical for military operations. Conventional airborne and ground sensor systems, use spatial information to detect and recognize targets. However, current automatic target detection techniques cannot readily find camouflaged or hidden targets. Furthermore, active Doppler detection cannot handle targets with zero or near-zero velocity. Multispectral and hyperspectral sensors can enhance target detection and recognition by increasing probability of detection and lowering false alarm rates for low signature moving targets. Physical Optics Corporation (POC) proposes to develop an innovative multispectral moving target detection (M~D) system, which applies both spectral and spatial detection filters, and performs spatial-spectral data fusion, multispectral, hyperspectral, and spatial detection, and correlates movement measurement to precisely detect and locate targets. The goal of this Phase I project is to optimally combine results from spatially based detection with high-resolution fast spectral imaging and spectral-based discrimination techniques. The proposed spatial-spectral detection will combine POC's multi-sensor data fusion, correlation target tracking, and efficient linear and nonlinear spatial filtering for clutter reduction, detection, identification, and tracking of military targets. Phase I will identify and develop a system concept, and will select moving target detection and clutter reduction algorithms for application to current military surveillance systems. BENEFITS: By greatly improving probability of detection and clutter reduction, and reducing the false alarm rate, multispectrallhyperspectral data fusion will open a broad market for spatial-spectral automatic target recognition. Multi-sensor remote sensing will aid agriculture, geophysical surveying, and prospecting, pollution monitoring, reconnaissance, surveillance, and mine and explosives detection.