fastpng (version 0.1.7)

read_png: Read a PNG

Description

Read a PNG

Usage

read_png(
  src,
  type = c("array", "raster", "nativeraster", "indexed", "raw"),
  rgba = FALSE,
  flags = 1L,
  avoid_transpose = FALSE,
  array_type = c("dbl", "int")
)

Value

R image object of the specified type

Arguments

src

PNG image provided as either a file path, or a raw vector containing encoded PNG data

type

type of R object in which to store image data. Valid types are 'array', 'raster', "nativeraster", 'indexed' and 'raw'. Note that indexed image objects can only be loaded from indexed PNGs.

rgba

Should the result be forced into RGBA? Default: FALSE means to use the most appropriate format of the given R image type to store the data. If TRUE, then the image will be forced into RGBA color mode.

flags

Flags to apply when reading PNG. Default: 1 (always decode transparency from tRNS chunks). See ?spng_decode_flags for other options. Must be an integer.

avoid_transpose

Default: FALSE. If TRUE, then transposing the image from row-major (in the PNG), into column-major (in R) will be avoided if possible. This option only applies when reading grayscale or indexed images. Since the transposition is avoided, the decode step can be faster, but the image will not be in the correct orientation.

array_type

'dbl' or 'int'. Default: dbl. When reading PNG into an array, should the data be stored as a double (i.e. real) in the range [0, 1] or an integer in the range [0,255] (for 8 bit images) or [0,65535] (for 16 bit images).

Examples

# create a small greyscale matrix, and write it to a PNG file
ras <- matrix(c('#880000', '#000088'), 3, 4) 
ras <- grDevices::as.raster(ras)
pngfile <- tempfile()
write_png(ras, file = pngfile)
ras2 <- read_png(pngfile, type = 'raster')
plot(ras2, interpolate = FALSE)