Python scripts to replicate human and spacecraft color vision from CRISM images.
- requirements.txt
This Python package converts a Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) data cube to various color spaces, including human perceptual color and emulated spacecraft filters. It integrates across a given wavelength range using the CIE color matching functions (for human perceptual color) or spacecraft filter functions (for spacecraft color) before conversion into sRGB color space. Currently, this package is only capable of processing Map-projected Targed Reduced Data Records (MTRDR), which represent the highest level of processing by the CRISM team. MTRDR images are map-projected, have the instruments 3 detectors joined into a single image, and are processed to reduce signal from atmospheric features (water ice/dust) and instrumental artifacts.
This code was developed to aid visualization of hyperspectral imaging data. It is free for personal use and academic presentations and publications. Please provide an acknowledgement in your visualization/presentation/publication when using this work.
Install the dependent packages with pip. Copy all files to a new directory. This program uses the Fire package to implement a command-line interface. To use it, open a terminal in the directory where you copied the files. Type python crism.py [function] --arg1=[$1] --arg2=[$2] to run a command.
The mtrdr_to_color() function uses the CIE color matching functions as a close approximation to standard human vision. Additional keyword arguments can be used to apply the matching function to a user-specified wavelength range, emulating the standard human visual response for that wavelength range.
User input: python crism.py mtrdr_to_color --file="[cube_name].lbl" --name="[output_name]". The standard outputs from this function include the following:
- 'VIS' - CIE human visual response, covering the wavelength range from 380 - 780 nm
- 'FAL' - Human visual response shifted to 1.1 - 2.2 microns, covering range of CRISM FAL parameter.
- 'FEM' - Human visual response shifted to 750 - 1.2 microns, capturing color variability in iron oxidation and Fe-rich mineralogy.
- 'MAF' - Human visual response shifted to 800 - 2.0 microns, capturing color variability in unweathered basaltic minerals.
- 'PHY' - Human visual response shifted to 1.8 - 2.2 microns, with color strongly dependent on presence of H20 and OH vibrational bands (hydrated minerals).
- 'FAR' - Human visual response shifted to 2.8 - 3.9 microns, spanning range of CRISM longwave sensor. Sensor prone to light leaks, not all images are usable.
- 'CAR' - Human visual response shifted to 2.9 - 3.4 microns, color somewhat dependent on presence of carbonate minerals. Same caveat as "FAR" applies.
If you would not like the above standard outputs, add --standard_params=False to the mtrdr_to_color inputs.
If you would like to specify a custom wavelength range, add --new_params=[[wave1, wave2], [wave1, wave2]...] to the mtrdr_to_color inputs. To find wavelength ranges which may highlight interesting mineralogy, see Vivano-Beck (2014), which discusses the spectral signals of various common Mars minerals.
Work in Progress Several functions are provided to calculate color data returned from various spacecraft. As of the current version, color information can be calculated for the following spacecraft instruments.
- Trace Gas Orbiter CaSSIS
- Mars Reconnaissance Orbiter HiRISE
- Mars Express HRSC
- Curiosity Mastcam
- Perseverance Mastcam-Z
- Mars Exploration Rover Pancam
This functionality is still experimental, and may not be radiometrically accurate. In addition, function control flow is still being worked out. Suggestions welcome!
To calculate CaSSIS images, use mtrdr_to_cassis --file="[cube_name].lbl" --fname="output_name". Standard output returns an IPG (NIR-PAN-BLU) color image. To change color combinations, use --color=[kwarg]. Currently supported combinations are "IRB" (RED-PAN-BLU) and "ENH" (a band-ratio image with the following ratios: R - RED/PAN; G - PAN/BLU; B - PAN/NIR)
To calculate HRSC images, use mtrdr_to_hrsc --file="[cube_name].lbl" --fname="[output_name]". Standard output returns an IGB (NIR-GRN-BLU) color image and grayscale images through the spacecraft's Nadir, NIR, RED, GRN, BLU, P1, and S1 filters.
To change color image combinations, add --color=[kwarg] to the command. Kwargs are "IGB" (NIR-GRN-BLU), "IRB" (NIR-RED-BLU), and "RGB" (RED-GRN-BLU)
If grayscale images are not desired, add --lumin=False to the command.
To calculate HiRISE images, use mtrdr_to_hirise --file="[cube_name].lbl" --fname="[output_name]". Standard output returns an IRB (NIR-RED-BGR) color image.
Additional color output options can be accessed with the --color=[kwarg] option. Kwargs are "RGB" (RED-BGR-Synthetic Blue, with synthetic blue image calculated using the HiRISE team formulation) and "ENH" (a band-ratio image with the following ratios: R - NIR/RED; G - NIR/BGR, B - RED/BGR).
To calculate Mastcam images, use mtrdr_to_mastcam --file="[cube_name].lbl" --fname="output_name". Standard output returns an RGB image equivalent to standard bayer-filtered Mastcam color images and grayscale images equivalent to images taken through narrowband filters L1-L6 and R1-R6.
If narrowband filter images are not desired, add --narrowband=False to the mtrdr_to_mastcam inputs.
To calculate Mastcam-Z images, use mtrdr_to_mastcamz --file="[cube_name].lbl" --fname="output_name". Standard output returns an RGB image equivalent to standard bayer-filtered Mastcam color images and grayscale images equivalent to images taken through narrowband filters L1-L6 and R2-R6.
If narrowband filter images are not desired, add --narrowband=False to the mtrdr_to_mastcamz inputs.
To calculate PanCam images, use mtrdr_to_pancam --file="[cube_name].lbl" --fname="output_name". Standard output returns a natural-color RGB image using the L3-5-7 combination and grayscale images for filters L1-L7 and R1-R7.
--color="IRB" can be used to access the L2-5-7 combination commonly used by the rover team for geological documentation imagery.
If narrowband filter images are not desired, add narrowband=False to the mtrdr_to_pancam inputs.
- Continue implementing color-matching functions for various spacecraft filters, double-check to make sure method is radiometrically accurate.
- Condense code for spacecraft filter combination if possible.
- Refine spacecraft filter outputs to be more user-friendly.
This code makes heavy use of the 'ColourSystem' class described on the SciPython blog. It also uses the SpectRes package to convert CIE matching functions to different wavelength ranges.