Title :
On the Extraction of Features from Slowly Wandering Patterns
fDate :
5/1/1981 12:00:00 AM
Abstract :
The problem of feature extraction with slowly wandering patterns is considered. Models are proposed to characterize the moving patterns. Using the perturbation theory of symmetric matrices, recursive relationships between the features at successive time intervals are obtained.
Keywords :
Communication systems; Covariance matrix; Data compression; Data mining; Feature extraction; Pattern recognition; Probability; Random processes; Signal representations; Symmetric matrices;
Journal_Title :
Aerospace and Electronic Systems, IEEE Transactions on
DOI :
10.1109/TAES.1981.309074
