Spatial Scan Density Estimates

Abstract

The detection and identification of regions of interest in spatial or temporal data is a common concern in automatic target recognition. One approach to region-of-interest identification involves the use of spatial scan statistics. A difficulty arises due to competing concerns: Small scan windows are required for potentially small targets, but larger scan windows are necessary to improve the accuracy of the detector. When the scan statistics are mixture-model density estimates, a borrowed strength profile likelihood approach is shown to be superior to conventional likelihood estimators. We investigate these spatial scan density estimates on example imagery from an unmanned aerial vehicle.

KEY WORDS :

• Automatic target recognition
• Borrowed strength
• Density estimation mixture
• Models
• Nonhomogeneity detection
• Profile likelihood
• Region-of-interest identification
• Spatial scan analysis
• Texture features
• Unmanned aerial vehicle imagery