Skip to contents

Estimate an image clip level

Source: R/all_generic.R, R/filebacked_neurovec.R, R/mapped_neurovec.R, and 2 more
clip_level-methods.Rd

Computes an AFNI-inspired clip threshold for separating foreground from low-intensity background. For 4D images, the threshold is computed from a representative 3D volume, typically the voxelwise median across time.

Usage

clip_level(x, mfrac = 0.5, gradual = FALSE, representative = "median")

# S4 method for class 'FileBackedNeuroVec'
clip_level(x, mfrac = 0.5, gradual = FALSE, representative = "median")

# S4 method for class 'MappedNeuroVec'
clip_level(x, mfrac = 0.5, gradual = FALSE, representative = "median")

# S4 method for class 'DenseNeuroVec'
clip_level(x, mfrac = 0.5, gradual = FALSE, representative = "median")

# S4 method for class 'AbstractSparseNeuroVec'
clip_level(x, mfrac = 0.5, gradual = FALSE, representative = "median")

# S4 method for class 'NeuroVol'
clip_level(x, mfrac = 0.5, gradual = FALSE, representative = "median")

Arguments

x

A neuroimaging object.

mfrac

Fraction used in the median-update step. Values outside (0, 0.99) fall back to 0.5.

gradual

If FALSE, return a scalar clip level. If TRUE, return a 3D clip map interpolated across image octants.

representative

For multi-volume inputs, the 3D summary image used for thresholding. Supported values are "median" and "mean_abs".

Value

If gradual = FALSE, a numeric scalar clip level. If gradual = TRUE, a DenseNeuroVol with voxelwise clip levels.

Examples

sp <- NeuroSpace(c(8, 8, 8), spacing = c(1, 1, 1))
vol <- NeuroVol(array(abs(rnorm(512)), c(8, 8, 8)), sp)
clip_level(vol)
#> [1] 0.5240544