Image File Channel Map


A file channel map is associated with a Draft.Image and is used to specify on write and retrieve on read the bit depth and type of the image channels.

Implementation Details

A file channel map is represented using a dictionary. Each dictionary entry is made of two strings separated by a colon. The first string corresponds to the channel name, typically ‘R’, ‘G’, ‘B’ and ‘A’, and the second string corresponds to the channel bit depth. An ‘ui’ is appended to the bit depth string to specify that the channel represents an index and an ‘f’ to specify that the channel is of type float. Such a string represents a channel data type.

Here are some common valid channel data types: ‘8’, ‘16’, ‘16f’, ‘32f’ and ‘32ui’. When Draft creates an image or add a new channel to an existing image, a default file channel data type of ‘8’ is assigned to the new channel(s).

A dictionary representing a file channel map can be passed to the function SetFileChannelMap() to set the file channel map of a Draft.Image. Note that the dictionary must contain one entry for each channel present in the Draft.Image and for those channels only. An error will be thrown otherwise.

A dictionary is returned by the function GetFileChannelMap() to retrieve the current file channel map of a Draft.Image.

Internally, Draft is still storing image channel data using 32 bit floats.

Valid Channel Data Types

It is important to understand that each file format only supports specific bit depth and type. Here’s a list of the valid channel data types supported by Draft for common file formats:

File Format Channel Data Type
CIN 10
DPX 8, 10, 12, 16
EXR 16f, 32f, 32ui
HDR 16
PNG 8, 16
TGA 5*, 8
TIF/TIFF 8, 16

* Writing a TGA image file with a channel data type of ‘5’ corresponds to ImageMagick’s 16-bit TGA – 5*3 + 1.

While Draft supports specifying the file channel map for other file formats, correct results are not guaranteed.