When you load the ICOR file in SNAP, does it have geolocation?
You can use Band Maths to copy bands from one file to another, but only if the navigation of both files is identical. Try loading both files in SNAP, and check the metadata for navigation details. The two ground footprints World View should match. If the ICOR file has been mapped, you may not need lon,lat bands as GIS software can compute pixel locations.
S3 OLCI L1 data file have geolocation… but after ICOR run geolocation are missing…
And also I checked the worldview footprint of both file.
L1 data have -Footprint
ICOR output file -No foot print in worldview
I have some doubt It is possible to add lat long by using Before atmospheric correction S3 L1 Data file?
It the arrays in ICOR results are the same size as the input bands, you may be able to copy the geolocation using one of the methods from How to Attach Geo Coding Information from Another File. Try using gpt Merge first (it wasn’t available at the time of the original post, but that was several versions ago).
Thanks for your reply …In my case ICOR results are not the same size as the input bands
S3 L1-21
Icor output-19 ((here Size variation I mean number of band is different))
Also I seen your link there I could able to see command merging product
Since I never used command line process step ,
Also Tried to attach geo-coding by SNAP, when click that option the new box show 19 band
Here this asking select two band to attach geocoding
which band i need choose ?-Becuase There is no lat long on 19 band
And Using merge product command line which option i wil choose ?How to start this process ?
What I meant by size is the dimension of the data raster. You should see this in the snap GUI as (NNNNN x NNNNN) when you move the mouse pointer over the band name, or in Band Properties, as in this 8640x4320 example:
This needs to be exactly the same as the size of the band from the input file whose navigation you want to use (for some sensors not all bands have the same resolution). In your case, there is good chance the ICOR results are consistent with the navigation from the input file. I have a workflow that uses products from the same sensor, but from different NASA groups. One is projected and “upside-down” in SNAP, the other is swath data and provides lon.lat. Some NOAA products I have used are given as [0,…360] longitude. [-180,…180] is commonly used in
NASA products. You should be able to compare image views for the same band in input and ICOR results to verify that the mapping is the same.
It is unlikely that the mapping changed. I have used Band Maths to add data from PNG files (no geolocation) to products that have navigation. If Band Maths doesn’t allow you to copy the lon,lat bands to the output file, you could try saving a subset of the input file (use BEAM DIMAP format) with just navigation (you may need to keep one data band), then add the output bands to the smaller product with Band Maths or use gpt merge to add them. If you had to keep one input band you can use subset to remove it after the output bands have been added. You should use BEAM DIMAP for intermediate files (those that are saved from SNAP/gpt and then loaded back with SNAP/gpt) and only convert to GeoTIFF or other formats when ready to load in a 3rd party tool that doesn’t handle BEAM DIMAP.
Sorry I retried to merge L1 laittude and long file but thats Not working …
you could try saving a subset of the input file -----Why need subset I want lat long for full images
gpt merge—Which file I want to give for source file
Further process step u suggested is confusing…
More details would help us understand what went wrong. What files did you use as the master and source? What error did you get?
You can use subset to select bands while leaving the image size unchanged. I’m not sure if Merge will work for a file with only the lon,lat bands – you may need one actual product band.
For merge the master has to be a file that is navigated.
I have one doubt , You clear my error GDAL lib missing by adding command in .py file
try:
import gdal
except ImportError:
from osgeo import gdal
Like this there may be some error in missing command for projection
python3 /opt/vito/icor/src/icor.py --keep_intermediate false --cloud_average_threshold 0.23 --cloud_low_band B02 --cloud_low_threshold 0.2 --aot true --aerosol_type RURAL --aot_window_size 100 --simec false --bg_window 1 --aot_override 0.1 --ozone true --aot_override 0.1 --ozone_override 0.33 --watervapor true --wv_override 2.0 --water_band B18 --water_threshold 0.06 --output_file /tmp/S3A_OL_1_EFR____20211205T042306_20211205T042606_20211205T060845_0179_079_204_2700_LN1_O_NR_002_processed.tif --sensor S3 --apply_gains false --glint_cor false --inlandwater false --productwater false --keep_land false --keep_water true –project false --working_folder /tmp/ /home/hello/E/Sentinel3/S3_CHL/S3A_OL_1_EFR____20211205T042306_20211205T042606_20211205T060845_0179_079_204_2700_LN1_O_NR_002/S3A_OL_1_EFR____20211205T042306_20211205T042606_20211205T060845_0179_079_204_2700_LN1_O_NR_002.SEN3/xfdumanifest.xml
I highlighted that project false ? if any chance to change .py for projection…
There may be a reason the “project” option was removed, and there doesn’t seem to be any description of the projection that might be used. I don’t think you have exhausted the possibilities for adding geolocation to you ICOR output bands.
Thanks, Still I facing that issue… Can u explain step by step In segmentation—generic region merging processor --which file i need choose for source
S3 OLCI L1 data or Icor output file …
You seem to be using “Generic Region Merging” from the Raster/Segmentation menu. This is a completely different type of merge. You should use the GPT Merge operator. You can do this using Graph Builder or with gpt in a terminal.
Thanks… I never use this before …You mean this graph B… if it is kindly tell how to use this what option i need to choose …I currently load L1C data which have geocordinate
The Help documents describe the use of graph builder. Each of the nodes generally has a page in the Help system. There are several youtube videos for SNAP Graph Builder.
I use Graph Builder for first attempt at some workflow, but if I need to repeat a workflow many times I use one of the batch processing tools.
Thanks for sharing I have seen that link GPT…Some are used S1 data … Currently I am in Graph builder ,What are all the step need for merge lat long
Since we don’t know if your data will work, you should first try to reproduce some basic tutorial examples so your are confident that your configuration is working and you are using the tool correctly. Then you can try experimenting with your data and the Merge operator.
[edit] For my data, the gpt Merge operator seems less fussy about what products it accepts than the BandMerge operator in graph builder.
When you are doing something new in complex software like SNAP Graph Builder, you should start with trivial examples (e.g., Read + Write using your own data). Break down your desired workflow into the simplest possible components (e.g, Read + Operation + Write) and make sure you understand what each each step does and can verify that it is working properly.
hai Thanks for you help, Now my output merged image is open… Previously due to low ram I could not…
Glad you found success.
When I’m experimenting with a new workflow I usually have a window that shows memory usage open so I can see if I’m trying to do more than the hardware can handle. Linux has an “Out of Memory Killer” that will abruptly kill user jobs that are using so much memory that it jeopardizes basic system functions. Years ago when student users shared multi-user unix systems some jokers would try to bring the system to a crawl or crash by finding ways to allocate all the memory.
Hey marpet,
I am confused with the installation of python version, in the manual it says that the python version should be 3.6 0r > so I have installed python version 3.9, however while setting up the python path for snap_v_8 - it says that the python version should be 2.7 or 3.4.
Which version shall I download and later how can I download these packages? Do I have to download via pip in cmd?
Thank you.
