Filtering Interpolators
Abstract
Interpolation error contributes to clutter leakage through a frame differencing signal processor, especially for highly structured scenes. This error can be reduced by an order of magnitude by applying a spectral filter to the unshifted frame that matches the filtering effect of applying the interpolator to the frame that is shifted. The signature reduction penalty for point targets is rarely as large as a factor of two, leading to substantial improvements in signal-to-clutter ratios when interpolation error is the dominant source of clutter. Parameterized families of local convolutional interpolators (polynomial and trigonometric) that can be adjusted to the particular target/clutter/noise combination of interest are presented. For spline interpolation applications, the trigonometric family yields an alternative to the cubic B-spline kernel.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 20, 1992
- Accession Number
- ADA253871
Entities
People
- Alan D. Stocker
- Robert L. Lucke
Organizations
- United States Naval Research Laboratory