We examine metrics for measuring clutter effectiveness on model-based autom
atic target recognition (ATR) systems with forward-looking infrared (FLIR)
sensors. The measure for clutter effectiveness proposed is the difference o
f two Kullback-Leibler distances between the idealized approximate probabil
istic models without clutter and the real model containing clutter. We esta
blish that occluding objects and clutter, when manipulated, do not present
a fundamental challenge to model-based ATR systems if the model manipulated
is an accurate representation of the obscuring clutter. However, if the ob
scurer is not manipulated, performance degrades in cases where the obscurer
is an "effective clutterer." To quantify the effect of clutter in ATR, est
imation and detection problems are considered for rigid ground-based target
s. For estimating the orientation of a vehicle, the Hilbert-Schmidt distanc
e is employed. (C) 1999 Society of Photo-Optical Instrumentation Engineers.
[S0091-3286(99)01112-5].