Help from LRO WAC Mosaic Processing
Hello everyone, it has definitely been a while since I have posted. I was taking a Lunar Geology short course and it took up more of my time than I realized. However, now the course is over (with the exception of one assignment due on the 29th of June) so I can now get back to the thick of my research. I wanted to start with a short blog on an USGS ISIS3 processing issue I have been trying to solve for a couple of weeks. Before I start I want to point out that I do not consider myself very experienced with ISIS3, or coding in general for that matter. I can process LRO NAC, Mini-RF, and MRO images with no issues but I am having some trouble with LRO WAC mosaics.
This started when I was making edits to my research paper on radar interpretation of lava flow surface roughness. I wanted to add a figure justifying the planetary science applicability of using radar backscatter. The figure I wanted was a Mini-RF CPR dataset highlighting the rough surface of lunar impact melt flows from Hayn Crater. Similar figures have been used by Neish et al (2014, 2017). I chose Hayn Crater since it has a pristine melt flow over its southern rim and ejecta blanket. I processed the Mini-RF data with minimal issues and wanted to overlay the data on a LRO WAC mosaic of the crater. This is when I ran into issues. I followed the basic steps to convert the .IMG files to .CUB files, assign orbital information, and calibrate the data (https://isis.astrogeology.usgs.gov/Application/index.html):
lrowac2isis - converted .IMG files from the Planetary Database System to .CUB files.
spiceinit - adding orbiter information to the .CUB file
lrowaccal - calibrate the LRO WAC .CUB file
After I finished these three steps I then proceeded to combine the LRO WAC datasets. For those who are not familiar with USGS ISIS3 processing of LRO WAC, after you convert the .IMG files to .CUB files an odd and even file are produced. Both have to be run through the instructions above. To combine the LRO WAC odd and even files I first needed to map project them.
I used the cam2map command to project both the odd and even files to a sinusoidal map projection as it was one of the projections available on ArcGIS. After both files were projected I used the automos command to mosaic the files together. This is when the problem arose. After they were mosaicked I viewed the dataset using qview. The dataset was really pixelated and had horizontal striping errors. At first I thought this stripe error was similar to the vertical stripe error taught in the USGS ISIS3 tutorials (https://isis.astrogeology.usgs.gov/fixit/projects/isis/wiki/Removing_Striping_Noise_from_Image_Data). So I went back to the tutorial to see if the solution for vertical stripe errors would fix horizontal stripes.
The tutorial instructed the use of lowpass and highpass filters to adjust the pixel average in the files. I tried this but no matter what values I assigned to the filter parameters nothing fixed the problem. I thought to myself maybe I am missing something crucial here or that I had forgotten a simple step. Remember, I am still learning image processing so I may have forgotten a simple step in the process but I want to improve and figure the problem out.
I looked online to see if someone else had experienced the same issue as me. I did not come across anything similar to my issue but I did find a paper that had a processing script for LRO WAC mosaics. The paper was written by Ohman (2015) - Procedure for processing LRO WAC monochromatic images with ISIS3 for photogeologic purposes. It looked perfect! I only need the images for photogeologic purposes, I do not need to perform any analysis to obtain quantitative results. The script in the paper had these steps (steps are simplified here. For details I recommend reading the paper. See the reference at the end of the blog):
Copy all WAC images to a single folder.
Execute the lrowac1 script provided in the paper.
converts .IMG to .CUB
performs spiceinit and calibrates the data.
Run photomet for photometric calibration.
Map project the data (select desired map projection).
Merge the odd and even files together using a list file (created using a text file of the path and file names).
Mosaic the images together using the noseam command.
Now I followed these instruction to the point where it asked for a LIST of the odd and even files. I had never come across this in USGS ISIS3 processing before and had no idea how to properly write it out. The tutorials have little information about writing out LIST and most other papers assume you know how to.
Reaching Out for Help: I wanted to ask if anyone who reads this knows how to process LRO WAC images and mosaic them together, and if they know how to write out LIST files. At first I wanted to try and figure this out on my own but after a lot of attempts to fix this I realized I needed to reach out and get help from more experienced people.
This is a short blog but I wanted to get this out first. More blogs are to follow this coming week so I can catch everyone up on what I was up to in May and my research plans for the rest of the summer term.
See you guys next time!
Reference: Öhman, T., 2015. Procedure for processing LRO WAC monochromatic images with ISIS 3 for photogeologic purposes.