Detection of Camouflaged Targets in Cluttered Backgrounds Using Fusion of Near Simultaneous Spectral and Polarimetric Imaging

Abstract

The detection of low signature or camouflaged targets in cluttered backgrounds is a crucial problem in tactical reconnaissance. In the past few years, imaging spectral and polarimetric sensors have been evaluated for this application. Although these sensors have separately generated promising results, each imaging modality alone appears to have not achieved the desired level of target detection. Fusion of data from multiple sensing modalities may potentially improve performance to acceptable levels. In the case of a key issue is the correlation of the spatial location of the false detection within spectral and polarmetric imaging. This paper presents a study of a data set consisting of near simultaneous spectral and polarimetric images recorded from sensors colocated on North Oscura Peak in the White Sands test range. The sensors overlooked a scene composed of natural background, military vehicles, and camouflage material. The sensors operated in the visible band with nearly equal, simultaneous field of view. The RX anomaly detection algorithm was separately applied to each data set to obtain a two dimensional map of target and false detection. The paper will analyze the correlation of false detection for image fusion. Background segmentation of the hyperspectral and polarization data sets was also examined.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 09, 2000
Accession Number
ADA392956

Entities

People

  • Alan Schaum
  • Christopher Steliman
  • Geoffrey Hazel
  • Richard Priest
  • Rulon Mayer

Organizations

  • United States Naval Research Laboratory

Tags

Communities of Interest

  • Sensors
  • Weapons Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Anomaly Detection
  • Change Detection
  • Data Analysis
  • Data Fusion
  • Data Sets
  • Detection
  • Detectors
  • Distortion
  • False Alarms
  • Field Tests
  • Hyperspectral Imagery
  • Information Science
  • Measurement
  • Operating Systems
  • Polarization
  • Target Detection

Readers

  • Image Processing and Computer Vision.
  • Sensor Fusion and Tracking Systems.