Top Percentile Filter

Local Algorithm - One-Dimensional Algorithm

Top Percentile Filter algorithm replaces the original data values with a chosen percentile of a certain number of neighbor points (e.g. their left $M$ and right $M$ data points). Note that the choice of filter width $L = 2M + 1$ has a great impact on the processed results. It is equivalent to the Centered Moving Median algorithm if the 50th percentile is chosen.

Input Parameters

Parameter	Type	Constraint	Description	Remarks
$Y[n]$	$Y[n] \in \mathbb R^N$	$N \in \mathbb{N}$	Input data sequence of length $N$
$L$	$L \in \mathbb N$	$L = 2M + 1, \quad M \in \mathbb{N}$
$p$	$p \in \mathbb{R}$	$0 \leq p \leq 1$	A specified percentile of a certain number of data points considered

Output Parameters

Parameter	Type	Constraint	Description	Remarks
$\hat{Y}[n]$	$\hat{Y}[n] \in \mathbb R^N$	$N \in \mathbb{N}$	Output data sequence of length $N$

Tool Support

• OriginLab

Single Steps using the Algorithm

• Data Denoising with Percentile Filter