Analytic performance model for grayscale quantization in the presence of additive noise

Abstract

ABSTRACT Synthetic aperture radar (SAR) exploitation algorithms typically rely on the use of derived features to representthe target. These features are chosen to discriminate between target classes while exhibiting robustness tonoise and calibration artifacts. One of the challenges in working with such features, is understanding when thisassumption of robustness is no longer valid. In this paper, we focus on characterizing the performance of thegray scale quantization feature in the presence of additive noise. We derive an approximation for the varianceof the intraclass distance by treating the additive noise as an independently identically distributed (iid) process.The analytic model is contrasted with em pirical results for a two class problem.Keywords: Sensor Exploitation, Gray Scale Quantization, Synthetic Aperture Radar, Performance Model 1. INTRODUCTION The imageformedbysynthetic apertureradar(SAR), hasmanybene“ts for useinasensorexploitationalgorithm.These include robustnessto time of day, cloud cover, and distance to target. However, there areseveraldrawbacksto working with SAR imagery; including noise, (both additive thermal noise and multiplicative speckle noise)