deepFeatures: extract deep features from 2D or 3D image

In stnava/patchMatchR: Local patch matching with flexible transformation models in n-dimensional images

extract deep features from 2D or 3D image

Description

High-level function for extracting features based on a pretrained network.

Usage

deepFeatures(
  x,
  mask,
  patchSize = 64,
  featureSubset,
  block_name = "block2_conv2",
  vggmodel,
  subtractor = 127.5,
  patchVarEx = 0.95,
  meanCenter = FALSE
)

Arguments

x	input input image
mask	defines the object of interest in the fixedImage
patchSize	vector or scalar defining patch dimensions
featureSubset	a vector that selects a subset of features
block_name	name of vgg feature block, either block2_conv2 or integer. use the former for smaller patch sizes. Or try ripmmarc.
vggmodel	prebuilt feature model
subtractor	value to subtract when scaling image intensity; should be chosen to match training paradigm eg 127.5 for vgg and 0.5 for resnet like.
patchVarEx	patch variance explained for ripmmarc
meanCenter	boolean mean center the patch for ripmmarc

Value

feature array, patches and patch coordinates

Author(s)

Avants BB

Examples

library(ANTsR)
img <- ri( 1 ) %>% iMath( "Normalize" )
mask = randomMask( getMask( img ), 20 )
features = deepFeatures( img, mask, patchSize = 32 )

stnava/patchMatchR documentation built on March 23, 2022, 6:47 a.m.