Jump to content
PDS Geosciences Node Community

Along-track stripping artifact correction and map projection


Wladimir Acevedo

Recommended Posts

Hi all,

as the title says, how can I remove in a simple way (I wish) this artifact from CRISM cubes?. I'm working with ENVI and CAT, but I'm not an IDL user, so if there exist a simple way to did it, would be really great. 

Also, how can I project a CRISM mineralogical map into a CTX image and into a DTM (HiRISE) terrain model?. I'm trying with ArcScene but it seems that it is possible with ENVI too.

Thanks,

Wladimir

Link to comment
Share on other sites

Wladimir: I suggest you use ENVI 5.5.2 interface and upgrade to CAT 7.4. You can use the flatten function in CAT to suppress column dependent striping. And with ENVI 5.5.2 it is easy to co-register CRISM images with CTX or HiRISE images. Let me know if you have these versions of ENVI and CAT. The CAT 7.4 is available from our PDS web site. https://pds-geosciences.wustl.edu/missions/mro/crism.htm

Ray Arvidson

PDS Geosciences Node Manager

CRISM Team Member

Link to comment
Share on other sites

Hi Wladimir ,
You can download non-PDS Compliant CTX RDR data from Mars ODE(https://ode.rsl.wustl.edu/mars/indexProductSearch.aspx) or from ASU Mars Space Flight Facility's Mars Image Explorer (http://viewer.mars.asu.edu/planetview/inst/ctx/#T=0) directly. Those CTX RDR data are map projected and could be overlain with the HiRISE DTM. ArcGIS can project the data on the fly if they have different projections. I can also help if you need more assists on data registration. Thanks,
June 

Link to comment
Share on other sites

On 1/3/2020 at 4:08 PM, Ray Arvidson said:

Wladimir: I suggest you use ENVI 5.5.2 interface and upgrade to CAT 7.4. You can use the flatten function in CAT to suppress column dependent striping. And with ENVI 5.5.2 it is easy to co-register CRISM images with CTX or HiRISE images. Let me know if you have these versions of ENVI and CAT. The CAT 7.4 is available from our PDS web site. https://pds-geosciences.wustl.edu/missions/mro/crism.htm

Ray Arvidson

PDS Geosciences Node Manager

CRISM Team Member

Thank you for your response Mr. Arvidson. Unfortunately, I only have access to the 5.3 version of ENVI software, so I would need a method using that version or Arcmap. Nevertheless, the flatten function seem to work well with the striping problem. 

Another question that I have is how much is necessary apply the "empirical corrections" for a TRDR product?. I mean, It is enough a standard corrected (atmospherical, photometric) TRDR product to analyze or I have to try to convert this on a "TER type product" through geom. normalization and spectral smile correction to analyze properly a scene?.

Any help will be very appreciated. 
Thanks,

Wladimir

Link to comment
Share on other sites

On 1/3/2020 at 4:39 PM, June Wang said:

Hi Wladimir ,
You can download non-PDS Compliant CTX RDR data from Mars ODE(https://ode.rsl.wustl.edu/mars/indexProductSearch.aspx) or from ASU Mars Space Flight Facility's Mars Image Explorer (http://viewer.mars.asu.edu/planetview/inst/ctx/#T=0) directly. Those CTX RDR data are map projected and could be overlain with the HiRISE DTM. ArcGIS can project the data on the fly if they have different projections. I can also help if you need more assists on data registration. Thanks,
June 

Thank you June. Do you refer to the EDR CTX products?. I'm using ArcScene trying to overlain the CTX images on a HiRISE DTM model but I don't know how to do this with a CRISM image. Also, do yow know how to create a mineralogical map from the CRISM cubes?. I'm trying to export the summary parameters custom stretched images, but sincerely, I don't have idea how to this in a right way.

Any help will be very greatful, thanks.

Wladimir

Link to comment
Share on other sites

Hi Wladimir ,
ENVI 5.3 would also work for you. Please have a try.
You can search for CTX EDR and Non-PDS Compliant RDRs through Mars ODE (https://ode.rsl.wustl.edu/mars/indexProductSearch.aspx) as shown in the attached fig1.  
In the data search results page (fig2), you can download the 'Pyramidized GeoTIFF' under 'Processed CTX' panel on the right. The Pyramidized GeoTIFF image has spatial reference information included in the image header.
You can export a map projected CRISM image using the CAT menu function as shown in the fig3. The CRISM output can be overlain with the CTX RDR in ArcGIS. Please let me know if this doesn't work for you. Thanks,
June

Fig1

Capture_ode_fig1.PNG

 

Fig2

Capture_ode_fig2.PNG


Fig3

Fig3.jpg

 

Link to comment
Share on other sites

2 hours ago, June Wang said:

Hi Wladimir ,
ENVI 5.3 would also work for you. Please have a try.
You can search for CTX EDR and Non-PDS Compliant RDRs through Mars ODE (https://ode.rsl.wustl.edu/mars/indexProductSearch.aspx) as shown in the attached fig1.  
In the data search results page (fig2), you can download the 'Pyramidized GeoTIFF' under 'Processed CTX' panel on the right. The Pyramidized GeoTIFF image has spatial reference information included in the image header.
You can export a map projected CRISM image using the CAT menu function as shown in the fig3. The CRISM output can be overlain with the CTX RDR in ArcGIS. Please let me know if this doesn't work for you. Thanks,
June

Fig1

Capture_ode_fig1.PNG

 

Fig2

Capture_ode_fig2.PNG


Fig3

Fig3.jpg

 

Hi June, 

I followed the steps that you indicated to me, and I've got this:

1.thumb.PNG.5b805af18e68fbc1bae603d4ea4d032c.PNG

Is the frt00003bfb_07_if166j_ter3 product charged with the PHY browse product (I've created frt00003bfb_07_if166j_ter3_2014params archive with CAT) over the D17_033826_2027_XN_22N018W CTX image. 

Looks very raw, but finally I'm moving forward, thank you. Nevertheless, what I'm looking to produce is something like this (images from Carter, 2015)

carter.thumb.png.35ba1e6f451b674de6c0507154fdaa8a.png:

 

What procedures do I need to follow to obtain  similar results?

Also, I've notice that the HiRiSE DTM are very limited in quantity and spatially. If I need a 3D model of the surface where a DTM isn't available should I use a HRSC or MOLA product?. Do they work similar to a DTM?

Thank you so much,

Wladimir

Link to comment
Share on other sites

I am not quite understand your question. Do you like to know how to give a 3D view? I saw your overlain of the CRISM onto a CTX. If you overlain all the images onto a DTM, and change the terrain exaggeration factors, you will see something like above figures from different points of views.

HiRiSE DTM has much higher resolution than the DTMs from HRSC and MOLA.

Map resolution for a HiRISE image is about 25cm/pixel, and the HiRISE DTM is about 1m/pixel.
HRSC DTMs have various resolution from 50m/pixel to 400m/pixel. Most of the products are at the resolution of ~50m/pixel, 75m/pixel, and 100m/pixel.
MOLA DTM is about 500m/pixel.
CTX image is about 6m/pixel. And FRT CRISM data are about 18m/pixel.

Depending on your needs, the higher resolution of a DTM, the more detailed features the terrain can represent.

Link to comment
Share on other sites

15 hours ago, June Wang said:

I am not quite understand your question. Do you like to know how to give a 3D view? I saw your overlain of the CRISM onto a CTX. If you overlain all the images onto a DTM, and change the terrain exaggeration factors, you will see something like above figures from different points of views.

HiRiSE DTM has much higher resolution than the DTMs from HRSC and MOLA.

Map resolution for a HiRISE image is about 25cm/pixel, and the HiRISE DTM is about 1m/pixel.
HRSC DTMs have various resolution from 50m/pixel to 400m/pixel. Most of the products are at the resolution of ~50m/pixel, 75m/pixel, and 100m/pixel.
MOLA DTM is about 500m/pixel.
CTX image is about 6m/pixel. And FRT CRISM data are about 18m/pixel.

Depending on your needs, the higher resolution of a DTM, the more detailed features the terrain can represent.

Thank you so much for your support June, now I can picture more clearly the general panorama. A remaining question would be about the process of create a "layer" from the CRISM image (using a browse product?) with only the "painted" pixels which represents the main mineralogy from a particular scene, as the maps who I've attached on the previous answer. Is there an automatized method to do is or maybe I should use a photo edition software to clean the images from the "black pixels" and create a layer with only the "colorized pixels"?

Thanks,

Wladimir

Link to comment
Share on other sites

Do you mean the black pixels along the edge of the image, which represent no data? If that is the "black pixels" you want to clean, in ArcGIS, when you overlain the CRISM output onto a DTM,  right click the image, and choose properties. Below is the instruction to remove the background area of a raster layer using the Unique Values renderer in ArcMap. Arcscene has similar function to set the background color as transparent. Or you can google an example. I cannot use ArcGIS desktop these days due to a university license issue. People are still working on that. If you still have problem, I can send you a pic example as soon as I can access my ArcGIS desktop again. Thanks, 
https://desktop.arcgis.com/en/arcmap/10.3/manage-data/raster-and-images/removing-the-background-using-the-unique-values-renderer.htm
June

Link to comment
Share on other sites

22 minutes ago, June Wang said:

Do you mean the black pixels along the edge of the image, which represent no data? If that is the "black pixels" you want to clean, in ArcGIS, when you overlain the CRISM output onto a DTM,  right click the image, and choose properties. Below is the instruction to remove the background area of a raster layer using the Unique Values renderer in ArcMap. Arcscene has similar function to set the background color as transparent. Or you can google an example. I cannot use ArcGIS desktop these days due to a university license issue. People are still working on that. If you still have problem, I can send you a pic example as soon as I can access my ArcGIS desktop again. Thanks, 
https://desktop.arcgis.com/en/arcmap/10.3/manage-data/raster-and-images/removing-the-background-using-the-unique-values-renderer.htm
June

I mean the black pixels which ENVI don't process for certain band selections custom stretched:

2.PNG.e2937a6db3ec3cb4d9cb0bf5ffe96da6.PNG

 

I want to "substract" only the painted shapes to put that over a CTX image, so the black part of this image would be the CTX image detailed surface or a DTM.

Thank you,

Wladimir

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...