Binary Matched Filtering with Pre-Processing
Feature Extraction with Matched Filters
by Robert Sparr
Overview of Methods Examined
Our first approach is inspired by the matched filter idea. Simply put, the optimal way to detect a known signal in an input signal with unknown noise is by finding a peak in the cross-correlation of the input signal and a version of the desired signal [Castleman]. This approach is the equivalent of designing a filter with an impulse response matching the desired signal.
We examined several variations on the straightforward use of matched filters to improve extraction:
- Matched filters with pre-processing and maximum-contrast filters
- Normalized matched filtering
- Techniques to improve matched filter feature extraction
jchen@micro.ti.com
tdorney@ti.com
sparr@owlnet.rice.edu
Last updated on May 3, 1997
Copyright © 1997 All rights reserved.