Wiener filter
Signal estimation technique
A Wiener filter is a signal processing technique that removes noise from a corrupted signal by minimizing the mean square error between the estimated and true signals.
Named after mathematician Norbert Wiener, it works optimally when the statistical properties of both the signal and noise are known.
Wiener filters appear in audio restoration, image deblurring, and communications systems: The old recording was cleaned up using a Wiener filter.
estimation
Joint deviation measure
- autocorrelationSignal's similarity to itself delayed
- signal-to-noise ratioWanted versus background level
- noise reductionDecreasing auditory interference
- denoisingRemoving unwanted signal artifacts
prediction
Extending data beyond bounds
- forecastPrediction of future conditions
- time seriesChronologically ordered observations
- predictive codingBrain anticipates sensory data
deconvolution
Reconstructing degraded visual media
- regularizationPenalty preventing model complexity
- motion blurSmearing from movement
- defocusRemove optical clarity
- microscopyTechnique using powerful lenses
covariance
Joint deviation measure
- autocorrelationSignal's similarity to itself delayed
- signal-to-noise ratioWanted versus background level
- noise reductionDecreasing auditory interference
extrapolation
Extending data beyond bounds
- forecastPrediction of future conditions
- time seriesChronologically ordered observations
- predictive codingBrain anticipates sensory data
image restoration
Reconstructing degraded visual media
- regularizationPenalty preventing model complexity
- motion blurSmearing from movement
- defocusRemove optical clarity
- filtershighpass, mesh screen, tea strainer, sieve, strainer
- adaptive filtersLMS
- Estimationballpark estimate, estimator, educated guess, guesstimate, rough estimate
filter, filtered, filters, filtering, unfiltered, biofilter, microfilter, prefilter, prefiltered