Multispectral imagery refers to images that are captured using multiple bands of the electromagnetic spectrum, beyond just the visible light spectrum. This means that they capture data across a wider range of wavelengths, from ultraviolet to infrared.
By capturing data across multiple spectral bands, multispectral imagery can provide additional information about the objects or scenes being imaged, such as their composition, temperature, and vegetation health. This information can be used for a wide range of applications, including agriculture, environmental monitoring, disaster response, and military surveillance.
Multispectral imagery is often captured using specialized sensors on satellites, drones, or aircraft. The data is then processed using specialized software to create images that display the information captured in each spectral band.
Overall, multispectral imagery provides a powerful tool for analyzing and understanding the world around us, allowing us to gain insights that would be impossible to obtain using visible light imagery alone.
Multispectral Support in ImageMagick
ImageMagick supports multispectral images where all channels have the same dimensions and number of pixels as the original image. However, not all image formats support multispectral images. PSD, TIFF, MIFF, MPC, and FTXT have full support for multispectral images up to 31 bands with 21 meta channels. If you build ImageMagick with the configure script --enable-64bit-channel-masks option, you can process 62 band multispectral images with up to 52 meta channels.
We call image channels beyond the standard RGB(A) or CMYK(A) channels, "meta" channels. We label them as "meta", "meta1", "meta2", etc.. Meta channels are treated the same as any other channel meaning they can be read, written, and operated upon-- e.g., resized, gamma adjusted, etc.
Create a Multispectral Image
Begin with a multispectral image, typically in the TIFF or PSD format. If you don't have one, you can create your own. In our example, we create a single CMYK pixel with two meta channels. Stuff this in a file titled multiMeta.txt:
0,0: 10,20,30,40,50,60
Let's convert that to the TIFF image format:
$ magick -size 1x1 -colorspace CMYK -define ftxt:format="\x,\y:\c\n" -define ftxt:hasalpha=false -define ftxt:nummeta=2 ftxt:multiMeta.txt multiMeta.tif
Let confirm that worked as expected:
$ identify -verbose multiMeta.tif
Image:
Filename: multiMeta.tif
Permissions: rw-------
Format: TIFF (Tagged Image File Format)
Mime type: image/tiff
Class: DirectClass
Geometry: 1x1+0+0
Units: PixelsPerInch
Colorspace: CMYK
Type: ColorSeparation
Endianness: LSB
Depth: 16-bit
Channels: 6.2
Channel depth:
Cyan: 16-bit
Magenta: 16-bit
Yellow: 16-bit
Black: 16-bit
Meta channel[0]: 16-bit
Meta channel[1]: 16-bit
Channel statistics:
Pixels: 1
Cyan:
min: 10 (0.00015259)
max: 10 (0.00015259)
mean: 10 (0.00015259)
median: 10 (0.00015259)
standard deviation: 0 (0)
kurtosis: -3
skewness: 0
entropy: 0
Magenta:
min: 20 (0.00030518)
max: 20 (0.00030518)
mean: 20 (0.00030518)
median: 20 (0.00030518)
standard deviation: 0 (0)
kurtosis: -3
skewness: 0
entropy: 0
Yellow:
min: 30 (0.000457771)
max: 30 (0.000457771)
mean: 30 (0.000457771)
median: 30 (0.000457771)
standard deviation: 0 (0)
kurtosis: -3
skewness: 0
entropy: 0
Black:
min: 40 (0.000610361)
max: 40 (0.000610361)
mean: 40 (0.000610361)
median: 40 (0.000610361)
standard deviation: 0 (0)
kurtosis: -3
skewness: 0
entropy: 0
Meta channel[0]:
min: 50 (0.000762951)
max: 50 (0.000762951)
mean: 50 (0.000762951)
median: 50 (0.000762951)
standard deviation: 0 (0)
kurtosis: -3
skewness: 0
entropy: 0
Meta channel[1]:
min: 60 (0.000915541)
max: 60 (0.000915541)
mean: 60 (0.000915541)
median: 60 (0.000915541)
standard deviation: 0 (0)
kurtosis: -3
skewness: 0
entropy: 0
Image statistics:
Overall:
min: 10 (0.00015259)
max: 60 (0.000915541)
mean: 35 (0.000534066)
median: 35 (0.000534066)
standard deviation: 0 (0)
kurtosis: -1.79762
skewness: 4.44476e-15
entropy: 0
Total ink density: 0.15259%
Colors: 1
Histogram:
1: (10,20,30,40) #000A0014001E0028 cmyk(0.0389105,0.077821,0.116732,0.155642)
Rendering intent: Perceptual
Gamma: 0.454545
Chromaticity:
red primary: (0.64,0.33)
green primary: (0.3,0.6)
blue primary: (0.15,0.06)
white point: (0.3127,0.329)
Matte color: grey74
Background color: white
Border color: srgb(223,223,223)
Transparent color: black
Interlace: None
Intensity: Undefined
Compose: Over
Page geometry: 1x1+0+0
Dispose: Undefined
Iterations: 0
Compression: None
Orientation: TopLeft
Properties:
date:create: 2023-03-12T11:38:42+00:00
date:modify: 2023-03-12T11:38:42+00:00
date:timestamp: 2023-03-12T11:38:56+00:00
signature: c51848f16d296f7c95be14404cc582f49db6fd33f31b544a2ce10e440b3747b3
tiff:alpha: unspecified
tiff:endian: lsb
tiff:photometric: separated
tiff:rows-per-strip: 1
Artifacts:
verbose: true
Tainted: False
Filesize: 306B
Number pixels: 1
Notice the channel depth and statistics associated with the two meta channels.
Working with Multispectral Images
Meta channels are treated the same as any other channel meaning they can be read, written, and operated upon-- e.g., resized, gamma adjusted, etc.. Here, we replace the first channel with the contents of the meta channel:
magick multiMeta.tif -channel-fx 'meta1=>cyan' multiMeta-1.tif
The original cyan channel has an intensity of 10. It's now, per the channel FX instructions, 50.
And in this example, we create two meta channels:
magick lena.png \( mandril3.jpg -colorspace gray \) -channel-fx '| gray=>meta' \( zelda1.jpg -colorspace gray \) -channel-fx '| gray=>meta1' meta.tif
Writing Multispectral Images
You can preserve multispectral images by writing them to a format that supports meta channels such as TIFF or PSD. If you write to other image formats that do not support multisprectal images, the channels are not preserved and instead lost-- e.g., PNG. In this example, we read, resize, and write a multispectral image:
magick multiMeta.tif -resize 50% resizedMeta.tif