issues in running spiceinit on Chandrayaan2 OHRC image

[Aaranay]

Hi, although this issue is in isis, Would you happen to know any fix for it?

I'm trying to run spiceinit on an OHRC image that I converted to cub using isisimport command. Apparently the translation file in MissionName2DataDir within the latest isis3 env had no mention of chandrayaan 2 mission. However, I was able to find that line in the github repo. After pasting that in the original file, I ran into the error: PVL keyword [Chandrayaan2] does not exist in [Group = DataDirectory].

I believe that this is because there isn't a single translation file for this mission in the environment, which is actually true because all the missions are listed in /appdata/translations/, except chandrayaan2. I tried searching for these files in the github repo too, but they weren't available. For a temporary fix, I tried copying and renaming the chandrayaan1 files to chandrayaan2 files. But that didn't work. 

Would you happen to know where I can find the translation files for chandrayaan2? Or if there's an alternate fix?

for reference, I also checked the files within the latest isis3 env (~8.3.0), but there were no files for this mission.

[Oleg Alexandrov]

Our Chandrayaan 2 doc says in the intro that this workflow is not ready for general use (https://stereopipeline.readthedocs.io/en/latest/examples/chandrayaan2.html#chandrayaan2).

The issue is that ISIS does not yet distribute the needed code or data. 

It is suggested you also file this as an issue with ISIS. ISIS and ASP are part of different organizations, and ISIS has its own planning and scheduling. Last I recall they were still in negotiation with the Indian space agency about how to properly share their data. 

I think you will need to build ISIS from source, likely fetch the latest ALE from coda (which I think is quite new), download the SPICE kernels from the ISRO (which is a big pain), and I could share my own addendum iak data which is not shipped with the kernels.

Sorry about this. This spring it looked as if the ISIS people were close to fully having this orbiter supported, also potentially with an official release, but then I had no contact with their team and when they were planning to finish up.

Filing an issue on their site is still likely the best way of figuring out what they are up to.

--
You received this message because you are subscribed to the Google Groups "Ames Stereo Pipeline Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ames-stereo-pipeline...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ames-stereo-pipeline-support/13d3bf1e-bb7e-42eb-98ac-640513c346c6n%40googlegroups.com.

[Aaranay]

Hi Oleg!
Thank you for the follow up. I have surpassed a few issues. Now I'm just stuck on the spiceinit step, where it is looking for the necessary files. I have all the spice kernels downloaded from ISRO. I would really appreciate it if you could share your IAK data along with the db files (if you have them as well). I'm really keen on seeing if this could work

much obliged,
Aaranay

[Oleg Alexandrov]

Aaranay,

At the link https://drive.google.com/file/d/1347AWxNFHb3nEA8AySc_6Ud5qeT6omMO/view?usp=sharing there should be all my chandrayaan2 files. This should be extracted in the ISIS data directory, and will go in the chandrayaan2 subfolder. This has .db and iak files (more exactly, everything but ck and spk). 

For reference, this doc describes how .db files get created: https://astrogeology.usgs.gov/docs/how-to-guides/environment-setup-and-maintenance/setup-isis-data/

I know this is a pain. The hope is that all needed code and data will be officially out at some point.

Oleg

To view this discussion visit https://groups.google.com/d/msgid/ames-stereo-pipeline-support/17cc14e0-a52a-4010-9858-1f156512fcban%40googlegroups.com.

[Aaranay]

much appreciated! I'll let you know how it progresses

[Aaranay]

Hi Oleg!

After adding all the files given above in the data along with copying some kernels from the base data, I was able to move forward from spice kernel issues. But, now it is looking for a mention of OHRC in one of the ISIS files. I tallied the data I have from the latest ISIS as well. I understand that ISIS is a part of a different organization and they are yet to add full support for chandrayaan2. But, would you happen to know any fix for it? As in which file should I mention OHRC in? 
I have attached a screenshot of the issue

Please let me know if that is possible. If not, I'll raise an issue on their GitHub.

with regards,
Aaranay

[Alexandrov, Oleg (ARC-TI)[KBR Wyle Services, LLC]]

I don't recall seeing this error. You may need to build latest ISIS from source. If still no luck after that, you can file an issue in their repo.

I also vaguely remember ISIS has dev conda builds somewhere, but I don't know if that is documented. That would save the effort of building the code. That is something you may ask them too.

---

From: ames-stereo-pi...@googlegroups.com <ames-stereo-pi...@googlegroups.com> on behalf of Aaranay <rnaa...@gmail.com>
Sent: Wednesday, October 8, 2025 4:19 AM
To: Ames Stereo Pipeline Support <ames-stereo-pi...@googlegroups.com>
Subject: [EXTERNAL] [BULK] Re: issues in running spiceinit on Chandrayaan2 OHRC image

 

CAUTION: This email originated from outside of NASA.  Please take care when clicking links or opening attachments.  Use the "Report Message" button to report suspicious messages to the NASA SOC.

To view this discussion visit https://groups.google.com/d/msgid/ames-stereo-pipeline-support/de223ada-c9b4-4b47-a842-9a4576ccaee1n%40googlegroups.com.

[Aaranay]

Hi Oleg!
Building the latest ISIS3 and ALE from source did the job. Parallel stereo ran successfully on OHRC. Really appreciate your help with the kernel files and guidance!

Although now, I want to look into SFS. I'll give it a try first, but I wanted to know if its possible.

Secondly, TMC-2 is still showing issues while making a cub file. Even though its making the file using isisimport, it doesn't seem to be valid. I compared the xml files for OHRC and TMC. There doesn't seem to be any errors in TMC. If anything, the TMC file carries a lot more information. So, then I'm unable to understand why isisimport shows I/O error while running it. I read your open issue #5835 on the ISIS repository. Were you also facing the same error? If not, have you been able to find any fixes?

much obliged,
Aaranay

[Oleg Alexandrov]

Aaranay, your persistence is appreciated.

Although now, I want to look into SFS. I'll give it a try first, but I wanted to know if its possible.

SfS is not easy to set up right, and it needs images with multiple illumination conditions, which TMC likely does not have. Our workflow has been tested with LRO NAC and Kaguya, and parameters need tweaking for these. There are more pointers here: https://stereopipeline.readthedocs.io/en/latest/sfs_usage.html#examples

For the record, we did produce a few large terrains with SfS, but it always requires quite some careful work and time. This is high-risk and high-effort work. CSM cares should be used if going down that path.

 

Secondly, TMC-2 is still showing issues while making a cub file. Even though its making the file using isisimport, it doesn't seem to be valid. I compared the xml files for OHRC and TMC. There doesn't seem to be any errors in TMC. If anything, the TMC file carries a lot more information. So, then I'm unable to understand why isisimport shows I/O error while running it. I read your open issue #5835 on the ISIS repository. Were you also facing the same error? If not, have you been able to find any fixes?

I did not work with TMC-2 much. I think it still needs time to mature. My best guess is that one should create the cub file with isisimport, then right away generate the CSM camera file, and test that one. I think CSM cameras will work even if the cub doesn't, or so I hope.

 

[Oleg Alexandrov]

To add to my previous response:

>  why isisimport shows I/O error while running it.

I did not see this particular error. In my case isisimport ran fine, but later the .cub did not appear right.

I think USGS is under gov shutdown, but if you have a reproducible error you can file an issue.

[Aaranay]

Really appreciate your help and insights.

For sfs, I was trying to find an available DEM of the same region as we created the OHRC DEM on (Region: LAT{-70, -67}, LON(20, 21)). I tried https://data.lroc.im-ldi.com/lroc , but the nearest available DEM was too far off. I also tried https://ode.rsl.wustl.edu/moon/lrololadataPointSearch.aspx , but the result wasn't consolidated. I'd really appreciate it if you can locate me to a webpage, where I can acquire a high res dem of that region. I tried Selene's DEMs of the same region, but they have a very low resolution as compared to the OHRC.

[Alexandrov, Oleg (ARC-TI)[KBR Wyle Services, LLC]]

I think the site  http://imbrium.mit.edu/ has a lot of DEMs, likely at 20 m/pixel or so. Note that, at least for some of them, the values need scaling by 0.5 as I recall (one can see what docs they have). I think you will need the cylindrical GDR product, but I did not look carefully.

Given that OHRC is at 0.25 m / pixel, those could be way too coarse. Maybe you are lucky to find LRO NAC stereo pairs and make a DEM. I recall there was somewhere a catalog of such stereo pairs. 

You are welcome to write to me offline for how sfs is going. The registration and picking good parameters will be the harder parts.

---

From: ames-stereo-pi...@googlegroups.com <ames-stereo-pi...@googlegroups.com> on behalf of Aaranay <rnaa...@gmail.com>

Sent: Thursday, October 23, 2025 4:42 AM

To: Ames Stereo Pipeline Support <ames-stereo-pi...@googlegroups.com>

Subject: Re: [EXTERNAL] [BULK] Re: issues in running spiceinit on Chandrayaan2 OHRC image

 

CAUTION: This email originated from outside of NASA.  Please take care when clicking links or opening attachments.  Use the "Report Message" button to report suspicious messages to the NASA SOC.

--

You received this message because you are subscribed to the Google Groups "Ames Stereo Pipeline Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ames-stereo-pipeline...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ames-stereo-pipeline-support/026c94db-7c5e-4082-9402-44e9a7371399n%40googlegroups.com.