Movatterモバイル変換

Type:	Package
Title:	Advanced Graphics and Image-Processing in R
Version:	2.9.0
Description:	Bindings to 'ImageMagick': the most comprehensive open-source image processing library available. Supports many common formats (png, jpeg, tiff, pdf, etc) and manipulations (rotate, scale, crop, trim, flip, blur, etc). All operations are vectorized via the Magick++ STL meaning they operate either on a single frame or a series of frames for working with layers, collages, or animation. In RStudio images are automatically previewed when printed to the console, resulting in an interactive editing environment. Also includes a graphics device for creating drawing onto images using pixel coordinates.
License:	MIT + file LICENSE
URL:	https://docs.ropensci.org/magick/https://ropensci.r-universe.dev/magick
BugReports:	https://github.com/ropensci/magick/issues
SystemRequirements:	ImageMagick++: ImageMagick-c++-devel (rpm) orlibmagick++-dev (deb)
VignetteBuilder:	knitr
Depends:	R (≥ 4.1.0)
Imports:	Rcpp (≥ 0.12.12), magrittr, curl
LinkingTo:	Rcpp
Suggests:	av, spelling, jsonlite, methods, knitr, rmarkdown, rsvg,webp, pdftools, ggplot2, gapminder, IRdisplay, tesseract,gifski
Encoding:	UTF-8
RoxygenNote:	7.3.2
Language:	en-US
NeedsCompilation:	yes
Packaged:	2025-09-07 21:52:50 UTC; jeroen
Author:	Jeroen Ooms [aut, cre]
Maintainer:	Jeroen Ooms <jeroenooms@gmail.com>
Repository:	CRAN
Date/Publication:	2025-09-08 14:30:02 UTC

Magick Image Processing

Description

Themagick package for graphics and image processing in R. Important resources:

• R introduction vignette: getting started

• Magick++ API andMagick++ STL detailed descriptions of methods and parameters

Details

Documentation is split into the following pages:

• analysis - metrics and calculations:compare,fft

• animation - manipulate or combine multiple frames:animate,morph,mosaic,montage,average,append,apply

• attributes - image properties:comment,info

• color - contrast, brightness, colors:modulate,quantize,map,transparent,background,colorize,contrast,normalize,enhance,equalize,median

• composite - advanced joining:composite,border,frame

• device - creating graphics and drawing on images

• editing - basic image IO:read,write,convert,join,display,brose

• effects - fun effects:despecle,reducenoise,noise,blur,charcoal,edge,oilpaint,emboss,implode,negate

• geometry - specify points, areas and sizes using geometry syntax

• ocr - extract text from image usingtesseract package

• options - list option types and values supported in your version of ImageMagick

• painting - flood fill and annotating text

• transform - shape operations:trim,chop,rotate,resize,scale,samplecrop,flip,flop,deskew,page

See Also

Other image:analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Image Analysis

Description

Functions for image calculations and analysis. This part of the package needs more work.

Usage

image_compare(image, reference_image, metric = "", fuzz = 0)image_compare_dist(image, reference_image, metric = "", fuzz = 0)image_fft(image)

Arguments

Details

For details seeImage++documentation. Short descriptions:

• image_compare calculates a metric by comparing image with a reference image.

• image_fft returns Discrete Fourier Transform (DFT) of the image as amagnitude / phase image pair. I wish I knew what this means.

Hereimage_compare() is vectorized over the first argument and returns the diff imagewith the calculated distortion value as an attribute.

See Also

Other image:_index_,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

out1 <- image_blur(logo, 3)out2 <- image_oilpaint(logo, 3)input <- c(logo, out1, out2, logo)if(magick_config()$version >= "6.8.7"){  diff_img <- image_compare(input, logo, metric = "AE")  attributes(diff_img)}

Image Frames and Animation

Description

Operations to manipulate or combine multiple frames of an image. Details below.

Usage

image_animate(  image,  fps = 10,  delay = NULL,  loop = 0,  dispose = c("background", "previous", "none"),  optimize = FALSE)image_coalesce(image)image_morph(image, frames = 8)image_mosaic(image, operator = NULL)image_flatten(image, operator = NULL)image_average(image)image_append(image, stack = FALSE)image_apply(image, FUN, ...)image_montage(  image,  geometry = NULL,  tile = NULL,  gravity = "Center",  bg = "white",  shadow = FALSE)

Arguments

Details

For details seeMagick++ STLdocumentation. Short descriptions:

• image_animate coalesces frames by playing the sequence and converting togif format.

• image_morph expands number of frames by interpolating intermediate frames to blendinto each other when played as an animation.

• image_mosaic inlays images to form a single coherent picture.

• image_montage creates a composite image by combining frames.

• image_flatten merges frames as layers into a single frame using a given operator.

• image_average averages frames into single frame.

• image_append stack images left-to-right (default) or top-to-bottom.

• image_apply applies a function to each frame

Theimage_apply function calls an image function to each frame and joinsresults back into a single image. Because most operations are already vectorizedthis is often not needed. Note thatFUN() should return an image. To apply otherkinds of functions to image frames simply uselapply,vapply, etc.

See Also

Other image:_index_,analysis,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# Combine imageslogo <- image_read("https://jeroen.github.io/images/Rlogo.png")oldlogo <- image_read("https://jeroen.github.io/images/Rlogo-old.png")# Create morphing animationboth <- image_scale(c(oldlogo, logo), "400")image_average(image_crop(both))image_animate(image_morph(both, 10))# Create thumbnails from GIFbanana <- image_read("https://jeroen.github.io/images/banana.gif")length(banana)image_average(banana)image_flatten(banana)image_append(banana)image_append(banana, stack = TRUE)# Append images togetherwizard <- image_read("wizard:")image_append(image_scale(c(image_append(banana[c(1,3)], stack = TRUE), wizard)))image_composite(banana, image_scale(logo, "300"))# Break down and combine framesfront <- image_scale(banana, "300")background <- image_background(image_scale(logo, "400"), 'white')frames <- image_apply(front, function(x){image_composite(background, x, offset = "+70+30")})image_animate(frames, fps = 10)# Simple 4x3 montageinput <- rep(logo, 12)image_montage(input, geometry = 'x100+10+10', tile = '4x3', bg = 'pink', shadow = TRUE)# With varying frame sizeinput <- c(wizard, wizard, logo, logo)image_montage(input, tile = '2x2', bg = 'pink', gravity = 'southwest')

Convert to EBImage

Description

Convert a Magick image toEBImageclass. Note that EBImage only supports multi-frame images in greyscale.

Usage

as_EBImage(image)

Arguments

Image Attributes

Description

Attributes are properties of the image that might be present on some imagesand might affect image manipulation methods.

Usage

image_comment(image, comment = NULL)image_info(image)

Arguments

Details

Each attribute can be get and set with the same function. Theimage_info()function returns a data frame with some commonly used attributes.

See Also

Other image:_index_,analysis,animation,color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

RStudio Graphics AutoViewer

Description

This enables aaddTaskCallback that automatically updates the viewer afterthe state of a magick graphics device has changed. This is enabled by default inRStudio.

Usage

autoviewer_enable()autoviewer_disable()

Examples

# Only has effect in RStudio (or other GUI with a viewer):autoviewer_enable()img <- magick::image_graph()plot(1)abline(0, 1, col = "blue", lwd = 2, lty = "solid")abline(0.1, 1, col = "red", lwd = 3, lty = "dotted")autoviewer_disable()abline(0.2, 1, col = "green", lwd = 4, lty = "twodash")abline(0.3, 1, col = "black", lwd = 5, lty = "dotdash")autoviewer_enable()abline(0.4, 1, col = "purple", lwd = 6, lty = "dashed")abline(0.5, 1, col = "yellow", lwd = 7, lty = "longdash")

Magick Configuration

Description

ImageMagick can be configured to support various additional tool and formatsvia external libraries. These functions show which features ImageMagick supportson your system.

Usage

coder_info(format)magick_config()magick_set_seed(seed)

Arguments

Details

Note thatcoder_info raises an error for unsupported formats.

References

https://imagemagick.org/Magick++/CoderInfo.html

Examples

coder_info("png")coder_info("jpg")coder_info("pdf")coder_info("tiff")coder_info("gif")# Reproduce random imagemagick_set_seed(123)image_blank(200,200, pseudo_image = "plasma:fractal")

Image Color

Description

Functions to adjust contrast, brightness, colors of the image. Details below.

Usage

image_modulate(image, brightness = 100, saturation = 100, hue = 100)image_quantize(  image,  max = 256,  colorspace = "rgb",  dither = TRUE,  treedepth = NULL)image_map(image, map, dither = FALSE)image_ordered_dither(image, threshold_map)image_channel(image, channel = "lightness")image_separate(image, channel = "default")image_combine(image, colorspace = "sRGB", channel = "default")image_transparent(image, color, fuzz = 0)image_background(image, color, flatten = TRUE)image_virtual_pixel(image, virtual_pixel_method)image_colorize(image, opacity, color)image_contrast(image, sharpen = 1)image_normalize(image)image_enhance(image)image_equalize(image)image_median(image, radius = 1)

Arguments

Details

For details seeMagick++ STLdocumentation. Short descriptions:

• image_modulate adjusts brightness, saturation and hue of image relative to current.

• image_quantize reduces number of unique colors in the image.

• image_ordered_dither reduces number of unique colors using a dithering threshold map.

• image_map replaces colors of image with the closest color from a reference image.

• image_channel extracts a single channel from an image and returns as grayscale.

• image_transparent sets pixels approximately matching given color to transparent.

• image_background sets background color. When image is flattened, transparent pixels get background color.

• image_colorize overlays a solid color frame using specified opacity.

• image_contrast enhances intensity differences in image

• image_normalize increases contrast by normalizing the pixel values to span the full range of colors

• image_enhance tries to minimize noise

• image_equalize equalizes using histogram equalization

• image_median replaces each pixel with the median color in a circular neighborhood

• image_virtual_pixel sets thevirtual pixel filling method to use when a raw distortion transformation (e.g.image_distort) introduces new pixels in the image.Some high-level transformations (e.g.image_rotate andimage_shear) will override the virtual pixel value with a default one.

Note thatcolors are also determined by image propertiesimagetype andcolorspacewhich can be modified viaimage_convert().

See Also

Other image:_index_,analysis,animation,attributes(),composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# manually adjust colorslogo <- image_read("logo:")image_modulate(logo, brightness = 200)image_modulate(logo, saturation = 150)image_modulate(logo, hue = 200)# Reduce image to 10 different colors using various spacesimage_quantize(logo, max = 10, colorspace = 'gray')image_quantize(logo, max = 10, colorspace = 'rgb')image_quantize(logo, max = 10, colorspace = 'cmyk')image_ordered_dither(logo, 'o8x8')# Change background colortranslogo <- image_transparent(logo, 'white')image_background(translogo, "pink", flatten = TRUE)# Compare to flood-fill method:image_fill(logo, "pink", fuzz = 20)# Black virtual pixel on a 45° rotationlogo |> image_virtual_pixel("Black") |>  image_distort("AffineProjection", sqrt(0.5) * c(1,1,-1,1,0,0), bestfit = TRUE)# Tile virtual pixel on a 45° rotationlogo |> image_virtual_pixel("Tile") |> image_distort("AffineProjection", sqrt(0.5) * c(1,1,-1,1,0,0), bestfit = TRUE)# Other color tweaksimage_colorize(logo, 50, "red")image_contrast(logo)image_normalize(logo)image_enhance(logo)image_equalize(logo)image_median(logo)# Alternate way to convert into black-whiteimage_convert(logo, type = 'grayscale')

Image Composite

Description

Similar to the ImageMagickcomposite utility: compose an image on top of another one using aCompositeOperator.

Usage

image_composite(  image,  composite_image,  operator = "atop",  offset = "+0+0",  gravity = "northwest",  compose_args = "")image_border(image, color = "lightgray", geometry = "10x10", operator = "copy")image_frame(image, color = "lightgray", geometry = "25x25+6+6")image_shadow_mask(image, geometry = "50x10+30+30")image_shadow(  image,  color = "black",  bg = "none",  geometry = "50x10+30+30",  operator = "copy",  offset = "+20+20")image_shade(image, azimuth = 30, elevation = 30, color = FALSE)

Arguments

Details

Theimage_composite function is vectorized over both image arguments: if the first image hasn frames and the secondm frames, the output image will containn *m frames.

Theimage_border function creates a slightly larger solid color frame and then composes theoriginal frame on top. Theimage_frame function is similar but has an additional feature tocreate a shadow effect on the border (which is really ugly).

See Also

Other image:_index_,analysis,animation,attributes(),color,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# Compose images using one of many operatorsimlogo <- image_scale(image_read("logo:"), "x275")rlogo <- image_read("https://jeroen.github.io/images/Rlogo-old.png")# Standard is atopimage_composite(imlogo, rlogo)# Same as 'blend 50' in the command lineimage_composite(imlogo, rlogo, operator = "blend", compose_args="50")# Offset can be geometry or gravityimage_composite(logo, rose, offset = "+100+100")image_composite(logo, rose, gravity = "East")# Add a border frame around the imageimage_border(imlogo, "red", "10x10")image_frame(imlogo)image_shadow(imlogo)image_shade(imlogo)

Set encoder defines

Description

So called 'defines' are properties that are passed along to externalfilters and libraries. Usually defines are used inimage_read orimage_write to control the image encoder/decoder, but you can alsoset these manually on the image object.

Usage

image_set_defines(image, defines)

Arguments

Details

The defines values must be a character string, where the names containthe defines keys. Each name must be of the format "enc:key" where thefirst part is the encoder or filter to which the key is passed. Forexample"png:...." defines can control the encoding and decoding ofpng images.

Theimage_set_defines function does not make a copy of the image, sothe defined values remain in the image object until they are overwrittenor unset.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# Write an imagex <- image_read("https://jeroen.github.io/images/frink.png")image_write(x, "frink.png")# Pass some properties to PNG encoderdefines <- c("png:compression-filter" = "1", "png:compression-level" = "0")image_set_defines(x, defines)image_write(x, "frink-uncompressed.png")# Unset propertiesdefines[1:2] = NAimage_set_defines(x, defines)image_write(x, "frink-final.png")# Compare size and cleanupfile.info(c("frink.png", "frink-uncompressed.png", "frink-final.png"))unlink(c("frink.png", "frink-uncompressed.png", "frink-final.png"))

Magick Graphics Device

Description

Graphics device that produces a Magick image. Can either be used like a regulardevice for making plots, or alternatively viaimage_draw to open a devicewhich draws onto an existing image using pixel coordinates. The latter is vectorized,i.e. drawing operations are applied to each frame in the image.

Usage

image_graph(  width = 800,  height = 600,  bg = "white",  pointsize = 12,  res = 72,  clip = TRUE,  antialias = TRUE)image_draw(image, pointsize = 12, res = 72, antialias = TRUE, ...)image_capture()

Arguments

Details

The device is a relatively recent feature of the package. It should support alloperations but there might still be small inaccuracies. Also it is a bit slower thansome of the other devices, in particular for rendering text and clipping. Hopefullythis can be optimized in the next version.

By defaultimage_draw sets all margins to 0 and uses graphics coordinates tomatch image size in pixels (width x height) where(0,0) is the top left corner.Note that this means the y axis increases from top to bottom which is the oppositeof typical graphics coordinates. You can override all this by passing customxlim,ylim ormar values toimage_draw.

Theimage_capture function returns the current device as an image. This onlyworks if the current device is a magick device or supportsdev.capture.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# Regular imagefrink <- image_read("https://jeroen.github.io/images/frink.png")# Produce image using graphics devicefig <- image_graph(res = 96)ggplot2::qplot(mpg, wt, data = mtcars, colour = cyl)dev.off()# Combineout <- image_composite(fig, frink, offset = "+70+30")print(out)# Or paint over an existing imageimg <- image_draw(frink)rect(20, 20, 200, 100, border = "red", lty = "dashed", lwd = 5)abline(h = 300, col = 'blue', lwd = '10', lty = "dotted")text(10, 250, "Hoiven-Glaven", family = "monospace", cex = 4, srt = 90)palette(rainbow(11, end = 0.9))symbols(rep(200, 11), seq(0, 400, 40), circles = runif(11, 5, 35),  bg = 1:11, inches = FALSE, add = TRUE)dev.off()print(img)# Vectorized example with custom coordinatesearth <- image_read("https://jeroen.github.io/images/earth.gif")img <- image_draw(earth, xlim = c(0,1), ylim = c(0,1))rect(.1, .1, .9, .9, border = "red", lty = "dashed", lwd = 5)text(.5, .9, "Our planet", cex = 3, col = "white")dev.off()print(img)

Edge / Line Detection

Description

Best results are obtained by finding edges withimage_canny() andthen performing Hough-line detection on the edge image.

Usage

image_edge(image, radius = 1)image_canny(image, geometry = "0x1+10%+30%")image_hough_draw(  image,  geometry = NULL,  color = "red",  bg = "transparent",  size = 3,  overlay = FALSE)image_hough_txt(image, geometry = NULL, format = c("mvg", "svg"))

Arguments

Details

For Hough-line detection, the geometry format is⁠{W}x{H}+{threshold}⁠defining the size and threshold of the filter used to find 'peaks' inthe intermediate search image. For canny edge detection the format is⁠{radius}x{sigma}+{lower%}+{upper%}⁠. More details and examples areavailable at theimagemagick website.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

if(magick_config()$version > "6.8.9"){shape <- demo_image("shape_rectangle.gif")rectangle <- image_canny(shape)rectangle |> image_hough_draw('5x5+20')rectangle |> image_hough_txt(format = 'svg') |> cat()}

Image Editing

Description

Read, write and join or combine images. All image functions are vectorized, meaningthey operate either on a single frame or a series of frames (e.g. a collage, video,or animation). Besides paths and URLs,image_read() supports commonly used bitmapand raster object types.

Usage

image_read(  path,  density = NULL,  depth = NULL,  strip = FALSE,  coalesce = TRUE,  defines = NULL)image_read_svg(path, width = NULL, height = NULL)image_read_pdf(path, pages = NULL, density = 300, password = "")image_read_video(path, fps = 1, format = "png")image_write(  image,  path = NULL,  format = NULL,  quality = NULL,  depth = NULL,  density = NULL,  comment = NULL,  flatten = FALSE,  defines = NULL,  compression = NULL)image_convert(  image,  format = NULL,  type = NULL,  colorspace = NULL,  depth = NULL,  antialias = NULL,  matte = NULL,  interlace = NULL,  profile = NULL)image_data(image, channels = NULL, frame = 1)image_raster(image, frame = 1, tidy = TRUE)image_display(image, animate = TRUE)image_browse(image, browser = getOption("browser"))image_strip(image)image_blank(width, height, color = "none", pseudo_image = "", defines = NULL)image_destroy(image)image_join(...)image_attributes(image)image_get_artifact(image, artifact = "")demo_image(path)

Arguments

Details

All standard base vector methods such as[,[[,c(),as.list(),as.raster(),rev(),length(), andprint() can be used to work with magickimage objects. Use the standardimg[i] syntax to extract a subset of the framesfrom an image. Theimg[[i]] method is an alias forimage_data() which extractsa single frame as a raw bitmap matrix with pixel values.

For reading svg or pdf it is recommended to useimage_read_svg() andimage_read_pdf()if thersvg andpdftools R packages are available.These functions provide more rendering options (including rendering of literal svg) andbetter quality than built-in svg/pdf rendering delegates from imagemagick itself.

X11 is required forimage_display() which is only works on some platforms. A moreportable method isimage_browse() which opens the image in a browser. RStudio hasan embedded viewer that does this automatically which is quite nice.

Image objects are automatically released by the garbage collector when they are no longerreachable. Because the GC only runs once in a while, you can also callimage_destroy()explicitly to release the memory immediately. This is usually only needed if you createa lot of images in a short period of time, and you might run out of memory.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# Download image from the webfrink <- image_read("https://jeroen.github.io/images/frink.png")worldcup_frink <- image_fill(frink, "orange", "+100+200", 20)image_write(worldcup_frink, "output.png")# extract raw bitmap arraybitmap <- frink[[1]]# replace pixels with #FF69B4 ('hot pink') and convert back to imagebitmap[,50:100, 50:100] <- as.raw(c(0xff, 0x69, 0xb4, 0xff))image_read(bitmap)# Plot to graphics device via legacy raster formatraster <- as.raster(frink)par(ask=FALSE)plot(raster)# Read bitmap arrays from other image packagesdownload.file("https://jeroen.github.io/images/example.webp", "example.webp", mode = 'wb')if(require(webp)) image_read(webp::read_webp("example.webp"))unlink(c("example.webp", "output.png"))if(require(rsvg)){tiger <- image_read_svg("http://jeroen.github.io/images/tiger.svg")svgtxt <- '<?xml version="1.0" encoding="UTF-8"?><svg width="400" height="400" viewBox="0 0 400 400" fill="none"> <circle fill="steelblue" cx="200" cy="200" r="100" /> <circle fill="yellow" cx="200" cy="200" r="90" /></svg>'circles <- image_read_svg(svgtxt)}if(require(pdftools))image_read_pdf(file.path(R.home('doc'), 'NEWS.pdf'), pages = 1, density = 100)# create a solid canvasimage_blank(600, 400, "green")image_blank(600, 400, pseudo_image = "radial-gradient:purple-yellow")image_blank(200, 200, pseudo_image = "gradient:#3498db-#db3a34",  defines = c('gradient:direction' = 'east'))

Image Effects

Description

High level effects applied to an entire image.These are mostly just for fun.

Usage

image_despeckle(image, times = 1L)image_reducenoise(image, radius = 1L)image_noise(image, noisetype = "gaussian")image_blur(image, radius = 1, sigma = 0.5)image_motion_blur(image, radius = 1, sigma = 0.5, angle = 0)image_charcoal(image, radius = 1, sigma = 0.5)image_oilpaint(image, radius = 1)image_emboss(image, radius = 1, sigma = 0.5)image_implode(image, factor = 0.5)image_negate(image)

Arguments

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,fx,geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

logo <- image_read("logo:")image_despeckle(logo)image_reducenoise(logo)image_noise(logo)image_blur(logo, 10, 10)image_motion_blur(logo, 10, 10, 45)image_charcoal(logo)image_oilpaint(logo, radius = 3)image_emboss(logo)image_implode(logo)image_negate(logo)

Image FX

Description

Apply a custom anfx expression to the image.

Usage

image_fx(image, expression = "p", channel = NULL)image_fx_sequence(image, expression = "p")

Arguments

Details

There are two different interfaces. Theimage_fx function simply appliesthe same fx to each frame in the input image. Theimage_fx_sequence functionon the other hand treats the entire input vector as a sequence, allowing youto apply an expression with multiple input images. See examples.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),geometry,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# Show image_fx() expressionimg <- image_convert(logo, colorspace = "Gray")gradient_x <- image_convolve(img, kernel = "Prewitt")gradient_y <- image_convolve(img, kernel = "Prewitt:90")gradient <- c(image_fx(gradient_x, expression = "p^2"),                image_fx(gradient_y, expression = "p^2"))gradient <- image_flatten(gradient, operator = "Plus")#gradient <- image_fx(gradient, expression = "sqrt(p)")gradientimage_fx(img, expression = "pow(p, 0.5)")image_fx(img, expression = "rand()")# Use multiple source imagesinput <- c(logo, image_flop(logo))image_fx_sequence(input, "(u+v)/2")

Geometry Helpers

Description

ImageMagick uses a handy geometry syntax to specify coordinates and shapesfor use in image transformations. You can either specify these manually asstrings or use the helper functions below.

Usage

geometry_point(x, y)geometry_area(width = NULL, height = NULL, x_off = 0, y_off = 0)geometry_size_pixels(width = NULL, height = NULL, preserve_aspect = TRUE)geometry_size_percent(width = 100, height = NULL)

Arguments

Details

SeeImageMagick Manualfor details about the syntax specification.Examples ofgeometry strings:

• "500x300" –Resize image keeping aspect ratio, such that width does not exceed 500 and the height does not exceed 300.

• "500x300!" –Resize image to 500 by 300, ignoring aspect ratio

• "500x" –Resize width to 500 keep aspect ratio

• "x300" –Resize height to 300 keep aspect ratio

• "50%x20%" –Resize width to 50 percent and height to 20 percent of original

• "500x300+10+20" –Crop image to 500 by 300 at position 10,20

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,morphology,ocr,options(),painting,segmentation,transform(),video

Examples

# Specify a pointlogo <- image_read("logo:")image_annotate(logo, "Some text", location = geometry_point(100, 200), size = 24)# Specify image areaimage_crop(logo, geometry_area(300, 300), repage = FALSE)image_crop(logo, geometry_area(300, 300, 100, 100), repage = FALSE)# Specify image sizeimage_resize(logo, geometry_size_pixels(300))image_resize(logo, geometry_size_pixels(height = 300))image_resize(logo, geometry_size_pixels(300, 300, preserve_aspect = FALSE))# resize relative to current sizeimage_resize(logo, geometry_size_percent(50))image_resize(logo, geometry_size_percent(50, 20))

Image to ggplot

Description

Create a ggplot with axes set to pixel coordinates and plot the raster imageon it usingggplot2::annotation_raster. See examples for how to plot an imageonto an existing ggplot.

Usage

image_ggplot(image, interpolate = FALSE)

Arguments

Examples

# Plot with base Rplot(logo)# Plot image with ggplot2library(ggplot2)myplot <- image_ggplot(logo)myplot + ggtitle("Test plot")# Show that coordinates are reversed:myplot + theme_classic()# Or add to plot as annotationimage <- image_fill(logo, 'none')raster <- as.raster(image)myplot <- qplot(mpg, wt, data = mtcars)myplot + annotation_raster(raster, 25, 35, 3, 5)# Or overplot image using gridlibrary(grid)qplot(speed, dist, data = cars, geom = c("point", "smooth"))grid.raster(image)

Morphology

Description

Apply a morphology method. This is a very flexible function whichcan be used to apply any morphology method with custom parameters.Seeimagemagick websitefor examples.

Usage

image_morphology(  image,  method = "convolve",  kernel = "Gaussian",  iterations = 1,  opts = list())image_convolve(  image,  kernel = "Gaussian",  iterations = 1,  scaling = NULL,  bias = NULL)

Arguments

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,ocr,options(),painting,segmentation,transform(),video

Examples

#example from IM website:if(magick_config()$version > "6.8.8"){pixel <- image_blank(1, 1, 'white') |> image_border('black', '5x5')# See the effect of Dilate methodpixel |> image_scale('800%')pixel |> image_morphology('Dilate', "Diamond") |> image_scale('800%')# These produce the same output:pixel |> image_morphology('Dilate', "Diamond", iter = 3) |> image_scale('800%')pixel |> image_morphology('Dilate', "Diamond:3") |> image_scale('800%')# Plus examplepixel |> image_morphology('Dilate', "Plus", iterations = 2) |> image_scale('800%')# Rose examplesrose |> image_morphology('ErodeI', 'Octagon', iter = 3)rose |> image_morphology('DilateI', 'Octagon', iter = 3)rose |> image_morphology('OpenI', 'Octagon', iter = 3)rose |> image_morphology('CloseI', 'Octagon', iter = 3)# Edge detectionman <- demo_image('man.gif')man |> image_morphology('EdgeIn', 'Octagon')man |> image_morphology('EdgeOut', 'Octagon')man |> image_morphology('Edge', 'Octagon')# Octagonal Convex Hull man |>   image_morphology('Close', 'Diamond') |>   image_morphology('Thicken', 'ConvexHull', iterations = 1)# Thinning down to a Skeletonman |> image_morphology('Thinning', 'Skeleton', iterations = 1)# Specify custom kernel matrix usingn a string:img <- demo_image("test_mag.gif")i <- image_convolve(img, kernel = '4x5:       0 -1  0  0      -1 +1 -1  0      -1 +1 -1  0      -1 +1 +1 -1       0 -1 -1  0 ', bias = "50%")}

Image Text OCR

Description

Extract text from an image using thetesseract package.

Usage

image_ocr(image, language = "eng", HOCR = FALSE, ...)image_ocr_data(image, language = "eng", ...)

Arguments

Details

To use this function you need to tesseract first:

  install.packages("tesseract")

Best results are obtained if you set the correct language intesseract.To install additional languages see instructions intesseract_download().

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,options(),painting,segmentation,transform(),video

Examples

if(require("tesseract")){img <- image_read("http://jeroen.github.io/images/testocr.png")image_ocr(img)image_ocr_data(img)}

Magick Options

Description

List option types and values supported in your version of ImageMagick. Fordescriptions seeImageMagick Enumerations.

Usage

magick_options()magick_fonts()option_types()filter_types()metric_types()dispose_types()compose_types()colorspace_types()channel_types()image_types()kernel_types()noise_types()gravity_types()orientation_types()morphology_types()style_types()decoration_types()compress_types()distort_types()virtual_pixel_methods()dump_option_info(option = "font")

Arguments

Details

Thedump_option_info function is equivalent to callingconvert -list [option] onthe command line. It does not return anything, it only makes ImageMagick printstuff to the console, use only for debugging.

References

ImageMagick Manual:Enumerations

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,painting,segmentation,transform(),video

Image Painting

Description

Theimage_fill() function performs flood-fill by painting starting point and allneighboring pixels of approximately the same color. Annotate prints some text onthe image.

Usage

image_fill(image, color, point = "+1+1", fuzz = 0, refcolor = NULL)image_annotate(  image,  text,  gravity = "northwest",  location = "+0+0",  degrees = 0,  size = 10,  font = "",  style = "normal",  weight = 400,  kerning = 0,  decoration = NULL,  color = NULL,  strokecolor = NULL,  strokewidth = NULL,  boxcolor = NULL)

Arguments

Details

Note that more sophisticated drawing mechanisms are available via the graphicsdevice usingimage_draw.

Setting a font, weight, style only works if your imagemagick is compiledwith fontconfig support.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),segmentation,transform(),video

Examples

logo <- image_read("logo:")logo <- image_background(logo, 'white')image_fill(logo, "pink", point = "+450+400")image_fill(logo, "pink", point = "+450+400", fuzz = 25)# Add some text to an imageimage_annotate(logo, "This is a test")image_annotate(logo, "CONFIDENTIAL", size = 50, color = "red", boxcolor = "pink", degrees = 30, location = "+100+100")# Setting fonts requires fontconfig support (and that you have the font)image_annotate(logo, "The quick brown fox", font = "monospace", size = 50)

Objects exported from other packages

Description

These objects are imported from other packages. Follow the linksbelow to see their documentation.

magrittr

%>%

Image Segmentation

Description

Basic image segmentation like connected components labelling, blob extraction and fuzzy c-means

Usage

image_connect(image, connectivity = 4)image_split(image, keep_color = TRUE)image_fuzzycmeans(image, min_pixels = 1, smoothing = 1.5)

Arguments

Details

• image_connect Connect adjacent pixels with the same pixel intensities to do blob extraction

• image_split Splits the image according to pixel intensities

• image_fuzzycmeans Fuzzy c-means segmentation of the histogram of color components

image_connect performs blob extraction by scanning the image, pixel-by-pixel from top-leftto bottom-right where regions of adjacent pixels which share the same set of intensity valuesget combined.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,transform(),video

Examples

# Split an image by colorimg <- image_quantize(logo, 4)layers <- image_split(img)layers# This returns the original imageimage_flatten(layers)# From the IM websiteobjects <- image_convert(demo_image("objects.gif"), colorspace = "Gray")objects# Split image in blobs of connected pixel levelsif(magick_config()$version > "6.9.0"){objects |>  image_connect(connectivity = 4) |>  image_split()# Fuzzy c-meansimage_fuzzycmeans(logo)logo |>  image_convert(colorspace = "HCL") |>  image_fuzzycmeans(smoothing = 5)}

Image thresholding

Description

Thresholding an image can be used for simple and straightforward image segmentation.The functionimage_threshold() allows to do black and white thresholding whereasimage_lat() performs local adaptive thresholding.

Usage

image_threshold(  image,  type = c("black", "white"),  threshold = "50%",  channel = NULL)image_level(  image,  black_point = 0,  white_point = 100,  mid_point = 1,  channel = NULL)image_lat(image, geometry = "10x10+5%")

Arguments

Details

• image_threshold(type = "black"): Forces all pixels below the threshold into black while leaving all pixelsat or above the threshold unchanged

• image_threshold(type = "white"): Forces all pixels above the threshold into white while leaving all pixelsat or below the threshold unchanged

• image_lat(): Local Adaptive Thresholding. Looks in a box (width x height) around thepixel neighborhood if the pixel value is bigger than the average minus an offset.

Examples

test <- image_convert(logo, colorspace = "Gray")image_threshold(test, type = "black", threshold = "50%")image_threshold(test, type = "white", threshold = "50%")# Turn image into BWtest |>  image_threshold(type = "white", threshold = "50%") |>  image_threshold(type = "black", threshold = "50%")# adaptive thresholdingimage_lat(test, geometry = '10x10+5%')

Image Transform

Description

Basic transformations like rotate, resize, crop and flip. Thegeometry syntaxis used to specify sizes and areas.

Usage

image_trim(image, fuzz = 0)image_chop(image, geometry)image_rotate(image, degrees)image_resize(image, geometry = NULL, filter = NULL)image_scale(image, geometry = NULL)image_sample(image, geometry = NULL)image_crop(image, geometry = NULL, gravity = NULL, repage = TRUE)image_extent(image, geometry, gravity = "center", color = "none")image_flip(image)image_flop(image)image_deskew(image, threshold = 40)image_deskew_angle(image, threshold = 40)image_page(image, pagesize = NULL, density = NULL)image_repage(image)image_orient(image, orientation = NULL)image_shear(image, geometry = "10x10", color = "none")image_distort(image, distortion = "perspective", coordinates, bestfit = FALSE)

Arguments

Details

For details seeMagick++ STLdocumentation. Short descriptions:

• image_trim removes edges that are the background color from the image.

• image_chop removes vertical or horizontal subregion of image.

• image_crop cuts out a subregion of original image

• image_rotate rotates and increases size of canvas to fit rotated image.

• image_deskew auto rotate to correct skewed images

• image_resize resizes using customfilterType

• image_scale andimage_sample resize using simple ratio and pixel sampling algorithm.

• image_flip andimage_flop invert image vertically and horizontally

The most powerful resize function isimage_resize which allows for settinga custom resize filter. Output ofimage_scale is similar toimage_resize(img, filter = "point").

For resize operations it holds that if nogeometry is specified, all framesare rescaled to match the top frame.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,video

Examples

logo <- image_read("logo:")logo <- image_scale(logo, "400")image_trim(logo)image_chop(logo, "100x20")image_rotate(logo, 45)# Small imagerose <- image_convert(image_read("rose:"), "png")# Resize to 400 width or height:image_resize(rose, "400x")image_resize(rose, "x400")# Resize keeping ratioimage_resize(rose, "400x400")# Resize, force size losing ratioimage_resize(rose, "400x400!")# Different filtersimage_resize(rose, "400x", filter = "Triangle")image_resize(rose, "400x", filter = "Point")# simple pixel resizeimage_scale(rose, "400x")image_sample(rose, "400x")image_crop(logo, "400x400+200+200")image_extent(rose, '200x200', color = 'pink')image_flip(logo)image_flop(logo)skewed <- image_rotate(logo, 5)deskewed <- image_deskew(skewed)attr(deskewed, 'angle')if(magick_config()$version > "6.8.6")  image_orient(logo)image_shear(logo, "10x10")building <- demo_image('building.jpg')image_distort(building, 'perspective', c(7,40,4,30,4,124,4,123,85,122,100,123,85,2,100,30))

Write Video

Description

High quality video / gif exporter based on external packagesgifskiandav.

Usage

image_write_video(image, path = NULL, framerate = 10, ...)image_write_gif(image, path = NULL, delay = 1/10, ...)

Arguments

Details

This requires an image with multiple frames. The GIF exporter accomplishes the samething asimage_animate but much faster and with better quality.

See Also

Other image:_index_,analysis,animation,attributes(),color,composite,defines,device,edges,editing,effects(),fx,geometry,morphology,ocr,options(),painting,segmentation,transform()

Example Images

Description

Example images included with ImageMagick:

Usage

logo

Format

An object of classmagick-image of length 1.

Details

• logo: ImageMagick Logo, 640x480

• wizard: ImageMagick Wizard, 480x640

• rose : Picture of a rose, 70x46

• granite : Granite texture pattern, 128x128