Skip to contents

Laplacian Enhancement Filter for Volumetric Images

Source: R/spat_filter.R
laplace_enhance.Rd

This function applies a multi-layer Laplacian enhancement filter to a volumetric image (3D brain MRI data). The filter enhances details while preserving edges using a non-local means approach with multiple scales.

Usage

laplace_enhance(
  vol,
  mask,
  k = 2,
  patch_size = 3,
  search_radius = 2,
  h = 0.7,
  mapping_params = NULL,
  use_normalization_free = TRUE
)

Arguments

vol

A NeuroVol object representing the image volume to be enhanced.

mask

A LogicalNeuroVol object specifying the region to process. If not provided, the entire volume will be processed.

k

An integer specifying the number of layers in the decomposition (default is 2).

patch_size

An integer specifying the size of patches for non-local means. Must be odd (default is 3).

search_radius

An integer specifying the radius of the search window (default is 2).

h

A numeric value controlling the filtering strength. Higher values mean more smoothing (default is 0.7).

mapping_params

An optional list of parameters for the enhancement mappings.

use_normalization_free

Logical indicating whether to use normalization-free weights (default is TRUE).

Value

A NeuroVol object representing the enhanced image.