Jump to content
PDS Geosciences Node Community

Working with PDS data

  • entries
    2
  • comments
    29
  • views
    49641

Contributors to this blog

Assigning Map Information to PDS Images in ENVI


Jennifer Ward

53552 views

In ENVI, you can assign map information to a PDS image based on values from the PDS label.

 

1. Download an image and corresponding PDS label file (.lbl extension) to the same directory.

2. Open the image in ENVI. It will not be map projected.

3. Select File > Edit ENVI Header and select the image. Click OK.

4. From Edit Attributes, choose Map Info...

5. Click Change Proj.

6. Unless you already have the correct lunar projection, create a new one (steps below assume projection does not exist).

7. Click New... on the Projection Selection dialog box.

8. Enter values based on label. For example, sphere radius 1738000 (notice in meters), projection type sinusoidal, 0 false

easting and northing, 74 central meridian longitude. (False easting/northing not in the label.)

9. Save projection and map_proj file. Select newly created projection.

10. In the Edit Map Information dialog, enter values from the label. For example, image coordinate x,y of 1.5,1.5 (the upper left

corner) might be 0,50 with pixel size of 400 m/pixel. Click OK to save map information and OK to save header.

 

Due to a technical issue, the original comments on this blog post were lost. A transcription of the comments can be found below.

28 Comments


Recommended Comments



User pmishra
Sep 20 2013 05:21 AM

Hi...
This was useful information. I am using LRO Lola's roughness data
(LDRM_75N_240M.img). I was trying to open it in ENVI. However problem occurs in
putting values for map projection. Which value should be given for latitude in true scale
etc.? Could you please specify the values to be entered for this data type.
It would be too helpful if you please tell me how to correctly open the LRO roughness data
in ENVI.
Regards-
Pooja

User June Wang

Sep 25 2013 11:35 AM

Update the Projection of LRO LOLA Products in ENVI 5.0 classic

When working with the LRO LOLA Products, the JP2000 products are read in ESRI®
ArcGIS with the correct map projection information, but ENVI cannot read the PDS IMG
and JP2000 products correctly. The following steps will take LOLA GDR product:
"LDRM_75N_240M.img" as an example, and show you how to fix this in ENVI 5.0 classic.

1. Update the map projection files
- ENVI includes three map projection files related to the definition of datum, ellipse
and map projections. First check where ENVI points to the map projection files.
    - Select File > Preference, under User Defined Files Tab, find Map Projection
      File.
    - If the right blank is empty, ENVI will point to the map projection files saved
      under the default system folder where ENVI was first installed,
      e.g.,"C:\Program Files\Exelis\ENVI50\classic\map_proj\" for ENVI 5.0
      classic. This path may vary machine by machine. Otherwise, ENVI will point
      to the map projection files saved in the path as defined in the right blank.
- Go to the folder where ENVI points to the map projection files. Backup the three
files: "datum.txt", "ellipse.txt", and "map_proj.txt" before you make any changes.
- Update file datum.txt
    - Open datum.txt
    - Add string "D_Moon_2000, GCS_Moon_2000, 0, 0, 0" to the end of the file
    - Save (If you have permission problem to save the file, please copy the file
      out to another folder. After changing, copy the saved file back to the original
      folder. If the permission is still denied, please give your system administrator
      to update the files.)
- Update file ellipse.txt
    - Open ellipse.txt
    - Add string "GCS_Moon_2000, 1737400.0, 1737400.0" to the end of the file
    - Save (take care of the permission problem as above)
- Update file map_proj.txt
    - Open map_proj.txt
    - Add string "31, 1737400.0, 1737400.0, 90.000000, 0.000000, 0., 0., Moon
      North Polar Stereographic" to the end of the file
    - Save (take care of the permission problem as above)
- Restarting ENVI will update the changes in the map projections.

2. Open the file in ENVI and edit the map information in the image header (take
LDRM_75N_240M.IMG as an example)
- Open LDRM_75N_240M.IMG in ENVI 5.0 classic
- Follow the instructions in the original blog post above to edit the map information
    - Select File > Edit ENVI Header and select the image. Click OK.
    - From Edit Attributes, choose Map Info.
    - Click Change Proj.
    - In the pop-up Projection Selection window
      - Choose the projection "Moon North Polar Stereographic " that was
        defined in the first step
      - Choose D_Moon_2000 as the projection datum for the lunar product.
      - Use meter as units
      - Click OK
    - We go back to the Edit Map Information dialog. In this dialog, enter values
      from the label. For example, image coordinate x,y of 1905.6, 1905.6 (the
      projection center of this product) might be 0,0 with pixel size of 240 m/pixel.
      Click OK to save map information and OK to save header.
- You can follow the same operations in step 2 to edit the map information of the
JP2000 products, E.g., LDRM_75N_240M.JP2. This JP2000 product has the same
product ID with LDRM_75N_240M.img. ENVI will treat them with the same *.hdr
file. If both JP2 and IMG file are saved in the same folder, ENVI will have problem
to open both files together. Either save them to different folders or change one of
the file names will solve the conflict.

User pmishra

Sep 25 2013 11:59 PM

Hi June,

In step 2, of Edit Map information dialogue, I have put map coordinates as 90 N and 0 E.
Is it correct?

Thanks
Pooja

User pmishra

Sep 26 2013 12:07 AM

Hi June,

Could you please tell the procedure to open and process this data in ArcGIS also.

Thanks
Pooja

User June Wang

Sep 26 2013 09:15 AM

Hi Pooja,

1) Regarding your question in step 2 -No. Since you already changed the map projection
to the "Moon North Polar Stereographic", and the unit is meter. Even you put 90N and 0 E
there, ENVI will interpolate this number as x = 0 m and y = 90m instead of degrees. Also it
is kind of tricky. The pole actually corresponds to Lat 90, and Lon in [0, 360] or [-180, 180].

2) Opening JP2000 map products in ArcGIS is straight forward. Using File->Open or
directly dragging the file to the map window will work. ArcGIS will read those map
projection parameters automatically.

As for the PDS IMG data, ESRI mentioned in the following link that ArcGIS 10.1 supports
some PDS 16-bit signed integer data, but still no multiband support.
http://resources.arcgis.com/en/help/main/10.1/index.html#//009t0000000q000000.

Trent at USGS also had a post about HiRISE PDS DEMs and ArcMap 10.x in the following
link.
https://isis.astrogeology.usgs.gov/IsisSupport/index.php?topic=3440.0.

My suggestion is if your PDS img is not recognized by ArcGIS, try to convert the PDS IMG
to an ENVI format, then ArcGIS will read the data in ENVI format. Please let me know if
you need more help. 

Thanks,
June
Link to comment

User Shubhra mathur

Aug 02 2014 02:26 AM

I have prepared the mosaic of Lunar poles using the Mini-RF data of LSZ where the pixel
resolution is 14.8 in ISIS software but when I visualize the data in ENVI i did not get the
Latitude and Longitude values how do I can get this help me...

User June Wang

Aug 05 2014 10:19 AM

Hi Shubhra mathur,

I am not sure if you got my e-mail yesterday. Or maybe you have solved your problem.
Would you please provide some more detail about your question if you need further help?
1) Do you process the LRO mini-RF CDR data?
2) Can you provide me the product_ID of the data you are processing, so I may reproduce
your issue?
3) What kind of ISIS output you used to be read in ENVI. Does that file have a ENVI
header *.hdr?

Thanks!
June

User Shubhra mathur

Aug 06 2014 12:52 AM

Dear Dr. June Wang,

I have used the MAPCDR data of the Mini-RF and processed the data strip in the ISIS
software using its different module and prepared the mosaics of the North pole. for
example the product id of Mini-RF datastrip is lsb_00260_2cd_oiu_87s283_v1.img.

When i visualize this strip in ENVI i did not get the latitude and longitute so i try to edit the
map info of this file in the option of edit ENVI header but still i am not able to edit the map
info of this file.

Could you please help me in this regard.

User June Wang

Aug 07 2014 10:39 AM

Hi Shubhra mathur,

I also had hard time to change the map projection parameters in ENVI for this product. If
you read Trent Thare's post at the below link, you will find ISIS is currently the only
package working with Oblique Cylindrical projection data. You would need to reproject
MAPCDR data to a different projection in ISIS for ENVI to understand the map projection.
Then you can view it in ENVI in X,Y meters. Please let me know if you need further help.
Thanks.

https://isis.astrogeology.usgs.gov/IsisSupport/index.php/topic,2698.msg11049.html#msg11049
Link to comment

User Aeronstorm

Dec 31 2016 05:00 AM

Hi,

I am trying to georeference, project and mosaic chandrayaan - 1, HYSI PDS 3 format files.
It has a .qub and a label fle .lbl containing metadata. Do i need a separate grid file to
georeference the raster ? What are the alternative softwares that can be used to finish
the job ?
It is an old post... But ,hey if anyone is hanging around .... help me.

Thanks.

User June Wang

Jan 03 2017 03:05 PM

Hi Aeronstorm,

I haven't worked with Chandrayaan HYSI data before, but I would recommend ENVI for
your work. ENVI works with multi-spectral and hyper spectral PDS data. It also includes
geometric correction/registration, raster re-projection, and mosaicking tools. As for the
grid, you can use it or not, depending on your own application. Feel free to let me know if I
am wrong or if you have any difficulties with ENVI. 

Thanks,
June

User Aeronstorm

Jan 03 2017 08:33 PM

Thank you June. I did try the data with ENVI and it works fine. I need to know that if the
polar stereo graphic projection technique mentioned above by you for LRO LOLA produts
can be used with other Lunar satellite images.

User June Wang

Jan 04 2017 10:17 AM

Hi Aeronstorm,

Glad that ENVI works for you.

The polar stereo graphic projection mentioned above should work with other Lunar data in
ENVI. But please pay attention to below two points.
1) Three map projection files related to the definition of datum, ellipse and map projections
were saved under the default system folder where ENVI was first installed. It varies on
different machines and different versions of ENVI being installed. E.g., for ENVI 5.3 in my
machine, it was at "C:\Program Files\Exelis\ENVI53\classic\map_proj"
2) you may need to change the parameters in the string in file map_proj.txt according to
your own projection being used.
"31, 1737400.0, 1737400.0, 90.000000, 0.000000, 0., 0., Moon North Polar
Stereographic" is for Moon North Polar Stereographic projection
"31, 1737400.0, 1737400.0, -90.000000, 0.000000, 0., 0., Moon South Polar
Stereographic" is for Moon South Polar Stereographic projection

In the string, 31 means Polar Stereographic projection in ENVI, 1737400.0 and 1737400.0
are the a and b value of an ellipse, (90.000000, 0.000000) degrees in (lat, lon)
corresponds to (0, 0)m in a projected map. You may modified those numbers to apply for
different projections you would like to use in ENVI. And it is not limited to Moon.
Hope this will answer your questions. 

Thanks,
June

User Aeronstorm

Jan 05 2017 02:16 AM

Thanks June. Can i use the datum created as D_moon_2000 for other projections such as
sinusoidal or should i create new datum file for other projections?

User June Wang

Jan 05 2017 09:42 PM

In the string "D_Moon_2000, GCS_Moon_2000, 0, 0, 0", added to file datum.txt,
'GCS_Moon_2000' is the name of an ellipse defined in file ellipse.txt
If you use the same ellipsoid for other lunar projections, you can use the same lunar
datum, otherwise, change the datum accordingly.

User Aeronstorm

Jan 06 2017 02:42 AM

Thanks June. I am trying to use sinusoidal projection for my satellite image. Should i
change the sphere radius equal to one mentioned on the ellipse.txt?

User June Wang

Jan 06 2017 11:26 AM

Would you please give me more information about the satellite image you are working on?
e.g., instrument, product type, product ID etc. And if you can post your current projection
string for your data, I can take a look. Thanks.

User Aeronstorm

Jan 07 2017 01:38 AM

Thanks June. I am working on Chandrayaan M3 data.
Product ID- M3G20090421T082045_V03_RDN.
map_proj.txt - 16, 1734000.0, 276.331769, 0.0, 0.0, moonsinusoidal

User June Wang

Jan 09 2017 10:26 AM

Hi Aeronstorm,
Is '1734000.0' a typo? The radius should be 1737400.0.
I am not sure ENVI works with longitude in the range of [0, 360] or [-180, 180]. If
276.331769 doesn't work for you, you can try to change it in the range of[-180, 180].
Otherwise, your map projection string looks fine for me. And you can use the same datum
as above mentioned. thanks
Link to comment

User Shashwat Shukla

Apr 12 2017 05:07 AM

Hi June,
I am working on Chandrayaan-1 M3 Data.
Product ID: m3g20090205t071411_v03_rdn
I am trying to geo reference the data as the method you suggested in your earlier
comments, but I am unable to find what should I enter in Central Longitude option in
customized map projection settings.
I changed the projection type to Sinusoidal, Sphere Radius: 1737400 Easting and
Northing as 0 and still couldnt find what to enter for 'Longitude for Central Meridian'. For
now, I entered 74 as you mentioned.
Then in edit map information, Datum chosen was D_Moon_2000 as you suggested. What
should I enter in X, Y Coordinates ?

Pixel Size I am using is 140 m as Spatial Resolution of M3 in Global Mode of OP1B data
is 140 m/pixel. Is this right?

Kindly guide me in the above!

PS: I am not able find any detail in label file. The label file I am using is :

m3g20090205t071411_v03_l1b.lbl

Thanks a lot for your time!

User June Wang

Apr 12 2017 02:23 PM

Hi Shashwat Shukla,

It looks other users may have similar confusion as you when working with Chandrayaan-1
M3 data in ENVI, so I put some tips here.
A spectral radiance image cube label (*_L1B.LBL) is detached and points to the following
L1B data products (Please see data product SIS file pages 20, 81 in 
https://pds-imaging.jpl.nasa.gov/data/m3/CH1M3_0003/DOCUMENT/DPSIS.PDF):
"¢ the single multi-band image (*_RDN.IMG) and its respective detached header file
(*_RDN.HDR),
"¢ the pixel location data (*_LOC.IMG) and its respective detached header file
(*_LOC.HDR),
"¢ the observation geometry data (*_OBS.IMG) and its respective detached header file
(*_OBS.HDR),
"¢ the UTC timing data (*_TIM.TAB)

Download all the *.loc, *.obs, *.rdn cubes together with the ENVI headers from ODE to
your local drive. You can access the location and geometry information in the *.LOC
and *.OBS files and relate it to the spectra of the *.RDN files using ENVI. Below link gives
the instruction for basic viewing of an M3 *.img cube file using ENVI 4.3. Read those M3 
cube file with high version of ENVI would have similar operations.
https://pds-imaging.jpl.nasa.gov/documentation/Appendix_E_DPSIS_ENVI_Users_Guide.pdf

note: In 'M3G20090205T071411_V03_LOC.HDR', it looks to me all the location data are
in geographic lat and lons in degrees. There is no projection defined in the *.hdr. You can
re-project the data to the Sinusoidal projection you defined. Central Meridian is the place
where there is no distortion. Feel free to let me know if you need more help. 

Thanks,
June

User Shashwat Shukla

Apr 13 2017 12:55 AM

Hi June,
I had linked all the display to access the location and geometry information in the *.LOC
and *.OBS files and relate it to the spectra of the *.RDN file.

I had come across another method of building GLT files and geo referencing through
them.

The process I followed in that is by choosing X as Longitude and Y as Latitude, and then
keeping the input projection as Arbitrary with Unknown Datum and output projection as
you suggested , i.e., Sinusoidal with Sphere Radius 1737400 m , False Easting and False
Northing as 0 and Central Meridian as 74 (Unsure whether right or not).

As you said that Central Meridian is the place where there is no distortion, how to know
the exact value of that. Because I am using the value which you suggested but it might
vary with datasets (Not sure). Kindly tell whether 74 which I am using is right or there is
some file in which it is written ?

And one more doubt, how to reproject the data when the input file is not geo referenced ?
PS: Since the *_RDN.IMG file is not geo referenced, so reprojecting it to Moon Sinusoidal
projection can come up with error.

Thanks a ton for your timely assistance.

Regards,
Shashwat

User June Wang

Apr 13 2017 03:59 PM

Hi Shashwat,
It is not right to set 74 as the Central Meridian for the Sinusoidal projection. I didn't
suggest this value before. I think it is just an example case that Jenn put together for users
years ago. The Central Meridian of a Sinusoidal projection mainly depends on your own
application. It is the place where you like it to have the least distortions.

When you build the GLT in ENVI, you could choose Geographic Lat/Lon and
D_moon_2000 for both the input and output as below figures.

There are no data that could be free of errors. Uncertainties come from different sources
such as instruments, measurements, processing methods and others. You could still reproject
the data even it is not georeferenced, just with more uncertainties in the output.
And the kind of projection depends on your own application and needs.

Hope this will answer your questions. 

thanks,
June

User June Wang

Apr 13 2017 04:02 PM

Hi Shashwat,

It seems I could not attach my screen shots in the above post. Feel free to let me know if
you like me to send 'Geometry Projection Information' dialogue from ENVI to you by
e-mail. 

Thanks,
June

User Shashwat Shukla

Apr 15 2017 04:57 AM

Hi June,

I did by assigning Geographic Lat/Long and D_moon_2000, but still I m not getting the
right output.
It would be better if you share your 'Geometry Projection Information' dialogue with me.
My email: sshashwat93@gmail.com
And I could also share my results with you.

Thanks a ton June!

Regards,
Shashwat

User June Wang

Apr 17 2017 06:51 AM

Hi Shashwat,

Please check your e-mail and let me know more details. 

Thanks,
June
Link to comment

hello everyone,

I have used the MAPCDR data of the Mini-RF , and able to open .img in ENVI, 

strip i m using is:lsz_04490_2s1_oku_77n214_v1

After foolowing the steps as mentioned in the earlier post, i m getting the following error

datum: D_MOON_2000

ellipsoid:GCS_Moon_2000

unable to include this datum because the ellipsoid was not found in the ''ellipse.txt'file

thanks !!

waiting for a response.

 

Link to comment

Hi Tarun,
It looks to me you didn't edit the 'ellipse.txt'file as mentioned in the previous post. I have copied the steps below for your reference. Feel free to let me know if you still have problems. Thanks,
June

1. Update the map projection files

- ENVI includes three map projection files related to the definition of datum, ellipse and map projections. First check where ENVI points to the map projection files.
    - Select File > Preference, under User Defined Files Tab, find Map Projection File.
    - If the right blank is empty, ENVI will point to the map projection files saved under the default system folder where ENVI was first installed, e.g.,"C:\Program Files\Exelis\ENVI50\classic\map_proj\" for ENVI 5.0 classic. This path may vary machine by machine. Otherwise, ENVI will point to the map projection files saved in the path as defined in the right blank.
    - Go to the folder where ENVI points to the map projection files. Backup the three files: "datum.txt", "ellipse.txt", and "map_proj.txt" before you make any changes.
    - Update file datum.txt
         - Open datum.txt
         - Add string "D_Moon_2000, GCS_Moon_2000, 0, 0, 0" to the end of the file
         - Save (If you have permission problem to save the file, please copy the file out to another folder. After changing, copy the saved file back to the original folder. If the permission is  still denied, please let your system administrator to update the files.)
    - Update file ellipse.txt
        - Open ellipse.txt
        - Add string "GCS_Moon_2000, 1737400.0, 1737400.0" to the end of the file
        - Save (take care of the permission problem as above)
    - Update file map_proj.txt
        - Open map_proj.txt
        - Add string "31, 1737400.0, 1737400.0, 90.000000, 0.000000, 0., 0., Moon North Polar Stereographic" to the end of the file [This string needs to be changed to the actual projection you needed for your data)
        - Save (take care of the permission problem as above)
    - Restarting ENVI will update the changes in the map projections.

Link to comment

Hi Tarun,
I just noticed that you are using the Mini-RF MAPCDR data product. And this product has an Oblique Cylindrical projection. I doubt if ENVI can handle the Oblique Cylindrical projection. USGS ISIS3 support the Oblique Cylindrical projection. You can use ISIS's cam2map or map2map functions to reproject the MAPCDR data to other projections, before you further process the data in other GIS software. More information about USGS ISIS tools could be found in the below link. thanks,
https://isis.astrogeology.usgs.gov/ 
June

 

Link to comment

31, 1737400.0, 1737400.0, 90.000000, 0.000000, 0., 0., moon north polar stereographic
31, 1737400.0, 1737400.0, -90.000000, 0.000000, 0., 0., moon south polar stereographic

in ENVI

 

but problem is 

An error has occurred during processing

Error: "Array dimensions must be grater than 0" The result may be invalid

i am stucking with this please help me.

Link to comment

Hi,

Your projection string looks good to me, but would you please post more information about your ENVI process, e.g., what data you are working on, and in which command this error message pop out? I can look into more details of this problem.

Thanks,

June 

Link to comment

Hello,

 
Why do different subset of M3 data gives different spatial resolution while georeferencing through IGM method from ENVI using Lat and Long bands.?
 
Kindly clarify this. 
Link to comment

M3 operates in different spatial modes depending on the orbit.

Would you please supply the actual M3 data set ID that produced different spatial pixels for different wavelengths? That way we can replicated the issue using ENVI's map projection code.

Ray Arvidson

PDS Geoscience Node Manager

Link to comment

Please give me the M3 product ID that you would like to coregister to the LOLA data quoted above. That way I can try it myself with ENVI before I reply to you with instructions.

Ray Arvidson

PDS Geosciences Node Manager

Link to comment

Hi,Ray

M3 Product ID: M3G20090609T095022_V01_RFL

Lola ldem: LDEM :  LDEM_1024_45S_30S_330_360

 LRO LOLA Gridded Data Record Shape Map (GDRDEM) (Derived Data

Plz do the same with this lola product id :

ldem_45s_100m.jp2

Thankyou

Link to comment

Hello Ray, 

please Reply as soon as possible. I have already send details of the data that i am using in my study.

 

 

Link to comment

Jue Wang and I have downloaded the LOLA and M3 data. We used ENVI 5.4 interface (not Classic) to map project the M3 data and then coregistered the mapped data to the LOLA map. Jue will be sending our instructions for you to do this, along with posting on our FTP site the final products.

Ray Arvidson

Link to comment
Thankyou June and Ray
 please specify the txt definition for Moon equirectangular C0 which i used in Map_Proj.txt file 
Link to comment

You can use below string in Map_Proj.txt file for the Moon Equirectangular projection center at 0 longitude.

17, 1737400.0, 0.0, 0.0, 0., 0., Moon Equirectangular C0

Link to comment

Hi june,

This previous data that i had used is from south pole .

if i used data from North pole ,which map projection i  used Moon Equirectangular C0 or Moon Equirectangular 180??

 

 

 

 

 

Link to comment

You can still use the Moon Equirectangular C0 for the North pole data. You can also use the below Moon North Polar Stereographic projection string when projecting the data.

31, 1737400.0, 1737400.0, 90.000000, 0.000000, 0., 0., Moon North Polar Stereographic

Link to comment

Hi Everyone,

I am Working with chandrayaan-1 M3 data and i want to calculate band depth for identify mineral. Plz help me how to calculate band depth and what is the value of n in the formula that is discussed by Mustard et al., 2011.

Link to comment

Hi Arpita,

Just to confirm, you're looking at the formulas on page 4 of this paper, right? https://pubs.er.usgs.gov/publication/70034937

It looks like those are calculating the band depth (1 - R/Rc) at a number of adjacent bands, then adding them together to get the integrated band depth over a range of wavelengths; the variable 'n' is just used to indicate which bands are used in the calculation. So for the 1 micron integrated band depth, you'd calculate band depth at each of 789 nm, 809 nm, 829 nm, and so on, up to 1309 nm, and then add the results (with a similar calculation for 2 micron integrated band depth, just using different M3 bands in the sum).

If you're using ENVI for your analysis, the Band Math tool should be able to help you calculate the band depth for each band that you're looking at, then sum up all of the results.

John

Link to comment

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...