ImageMagick Writer-Reader

ImageMagick Writer-Reader (IMWRI) is a plugin that can read and write many image formats.

imwri.Write(clip clip, string imgformat, string filename[, int firstnum=0, int quality=75, bint dither=True, string compression_type, bint overwrite=False, clip alpha])
Supported input formats for writing:

ImageMagick with Quantum Depth 16 and HDRI: 8-16 bit integer, 32 bit float

Write will write each frame to disk as it’s requested. If a frame is never requested it’s also never written to disk.


Input clip. RGB and Gray supported. See list for accepted inputs.


The name of the output format. Examples of supported format strings are “JPEG”, “PNG”, and “DPX”. Visit the ImageMagick website for a full list.


The filename string must have one or more frame number substitutions. The syntax is printf style. For example “image%06d.png” or “/images/%d.jpg” is common usage.


The first image number in the sequence to write.


Quality adjustment for formats where it’s applicable. Range is 0 to 100.


Use Floyd–Steinberg dithering if the input needs to be reduced in depth.


Select the specific compression type for imgformats that have more than one possible compression method. Recognized constants are: Undefined, None, BZip, DXT1, DXT3, DXT5, Fax, Group4, JPEG, JPEG2000, LosslessJPEG, LZW, RLE, Zip, ZipS, Piz, Pxr24, B44, B44A, LZMA, JBIG1, JBIG2


Overwrite already existing files. This option also disables the requirement that output filenames contain a number.


A grayscale clip containing the alpha channel for the image to write. Apart from being grayscale, its properties must be identical to the main clip.

imwri.Read(string[] filename[, int firstnum=0, bint mismatch=False, bint alpha=False, bint float_output = False])

Possible output formats when reading: 8-16 bit integer and 32 bit float

Note that by default 8-16 bit images are returned as integer and 32 bit images as float. When reading half precision float images you have to manually set float_output to have the unmodified floating point range returned.

Read is a simple function for reading single or series of images and returning them as a clip.


The filename argument has two main modes. Either it takes a list of 1 or more files to open in the given order, or it takes a single filename string with one or more frame number substitutions. The syntax is printf style. For example “image%06d.png” or “/images/%d.jpg” is common usage.


The first image number to start reading from when reading a sequence.


Allow reading of multiple images with different resolutions. If required and not set, an error will be generated.


Return the alpha channel from the read images as a separate grayscale clip. Note that an alpha channel clip is always returned when this parameter is set, even for image formats without support for it.


Always return the read image in a float format. Due to the output format guessing this option can be useful when reading half precision float images.