with_kernel: Apply a gaussian blur to your layer

In ggfx: Pixel Filters for 'ggplot2' and 'grid'

Apply a gaussian blur to your layer

Description

This filter allows you to apply a custom kernel to your layer, thus giving you more control than e.g. with_blur() which is also applying a kernel.

Usage

with_kernel(
  x,
  kernel = "Gaussian:0x2",
  iterations = 1,
  scaling = NULL,
  bias = NULL,
  stack = FALSE,
  ...
)

Arguments

x	A ggplot2 layer object, a ggplot, a grob, or a character string naming a filter
kernel	either a square matrix or a string. The string can either be a parameterized kerneltype such as: "DoG:0,0,2" or "Diamond" or it can contain a custom matrix (see examples)
iterations	number of iterations
scaling	string with kernel scaling. The special flag "!" automatically scales to full dynamic range, for example: "50%!"
bias	output bias string, for example "50%"
stack	Should the original layer be placed on top?
...	Arguments to be passed on to methods. See the documentation of supported object for a description of object specific arguments.

Value

Depending on the input, either a grob, Layer, list of Layers, guide, or element object. Assume the output can be used in the same context as the input.

Examples

library(ggplot2)
# Add directional blur using the comet kernel
ggplot(mtcars, aes(mpg, disp)) +
  with_kernel(geom_point(size = 3), 'Comet:0,10')

ggfx documentation built on Aug. 8, 2025, 7:18 p.m.