Multidimensional Digital Filters for Point-Target Detection in Cluttered Infrared Scenes

TL;DR

This paper presents a multidimensional digital filtering approach using 3-D prediction-error filters and optical flow to detect point targets in cluttered infrared scenes, enhancing contrast and discriminating targets based on velocity.

Contribution

It introduces a novel velocity-tuned 3-D prediction-error filtering method with analytical frequency response expressions for improved point-target detection in infrared imagery.

Findings

Enhanced foreground/background contrast in infrared sequences

Effective velocity-based target extraction in cluttered scenes

Analytical filter design for velocity-tuned FIR filters

Abstract

A 3-D spatiotemporal prediction-error filter (PEF), is used to enhance foreground/background contrast in (real and simulated) sensor image sequences. Relative velocity is utilized to extract point-targets that would otherwise be indistinguishable on spatial frequency alone. An optical-flow field is generated using local estimates of the 3-D autocorrelation function via the application of the fast Fourier transform (FFT) and inverse FFT. Velocity estimates are then used to tune in a background-whitening PEF that is matched to the motion and texture of the local background. Finite-impulse-response (FIR) filters are designed and implemented in the frequency domain. An analytical expression for the frequency response of velocity-tuned FIR filters, of odd or even dimension, with an arbitrary delay in each dimension, is derived.