yumorishita / LiCSBAS

LiCSBAS: InSAR time series analysis package using LiCSAR products
https://doi.org/10.3390/RS12030424
GNU General Public License v3.0
231 stars 110 forks source link

Visualization Result #197

Closed araditiya closed 1 year ago

araditiya commented 1 year ago

Hi Morishita,

A similar problem appears in the new machine (ubuntu 20.04). I still can't plot using the left click. It is totally similar to previous post, https://github.com/yumorishita/LiCSBAS/issues/184#issuecomment-1305494485 Hfff... it is strange.

araditiya commented 1 year ago

May I know, how you can visualize annual LOS in GIS format (including color bar) for each orbit as shown in your paper (2021) ?

yumorishita commented 1 year ago

The --sin option of LiCSBAS_cum2vel.py can output the amplitude and time offset of the annual component.

araditiya commented 1 year ago

Thank you for your great assistance. I completed getting the results. However, I didn't see the raster legend for each geotiff file. Could you help us with this case?

yumorishita commented 1 year ago

amp is in mm and dt is in day (time difference wrt Jan 1).

araditiya commented 1 year ago

Sorry for my unclear question. I meant there is no range value or raster legend when geotiff is imported to arcgis/qgis. It looks like DEM data if we import to gis.

yumorishita commented 1 year ago

I cannot understand your situation. Can you show me the screenshot or video?

araditiya commented 1 year ago

Please find the picture file as attached.

Screen Shot 2023-02-01 at 21 02 51
yumorishita commented 1 year ago

Did you use LiCSBAS_color_geotiff.py? The output GeoTIFF from it only contain color information, not numerical values. You should use the output of LiCSBAS_flt2geotiff.py to get the numerical values.

araditiya commented 1 year ago

Ok. It has been resolved. Anyway, how to remove vertical lines (GAP) when we plot using LiCSBAS_plot_ts command ?

yumorishita commented 1 year ago

There is no option to remove the vertical line, but I guess you can do it by comment out the line below. https://github.com/yumorishita/LiCSBAS/blob/dfdb8fadf323c821f386ed60eb0e871288961a72/bin/LiCSBAS_plot_ts.py#L869

araditiya commented 1 year ago

Is there any standard or template of the colour ramp as you did in gis apps ?

yumorishita commented 1 year ago

Please see the usage of LiCSBAS_disp_img.py and the links there: https://github.com/yumorishita/LiCSBAS/blob/dfdb8fadf323c821f386ed60eb0e871288961a72/bin/LiCSBAS_disp_img.py#L17

https://matplotlib.org/tutorials/colors/colormaps.html http://www.fabiocrameri.ch/colourmaps.php https://docs.generic-mapping-tools.org/dev/cookbook/cpts.html https://matplotlib.org/cmocean/ https://colorcet.holoviz.org/

araditiya commented 1 year ago

Almost perfect, however, I can't save to kmz file after get following error

saka@xps8930:~/work$ LiCSBAS_disp_img.py -i 20160109_20221227.vel -p EQA.dem_par --kmz 20160109_20221227

LiCSBAS_disp_img.py ver1.11.0 20210205 Y. Morishita LiCSBAS_disp_img.py -i 20160109_20221227.vel -p EQA.dem_par --kmz 20160109_20221227 /usr/lib/python3.8/zipfile.py:1517: UserWarning: Duplicate name: '20160109_20221227' return self._open_to_write(zinfo, force_zip64=force_zip64) Traceback (most recent call last): File "/home/saka/LiCSBAS/bin/LiCSBAS_disp_img.py", line 317, in os.remove(pngnametmp) FileNotFoundError: [Errno 2] No such file or directory: '20160109_20221227

yumorishita commented 1 year ago

--kmz 20160109_20221227 should be like --kmz 20160109_20221227.vel.kmz

araditiya commented 1 year ago

Thanks, it works. I expect it should be add "mm" on the color bar. Anyway, will you plan to make licsbas possible process palsar-2, at least for slc format ?

yumorishita commented 1 year ago

LiCSBAS is the time series analysis software using unwrapped images and can already process ALOS-2 data (or any other satellite's data) if unwrapped images are ready.

Sakina-A commented 1 year ago

Hey Morishita I was trying to draw contours but I keep getting this error although the file exist. Also, without downsampling the resolution is ~100 m but how to find the resolution of the resulted LOS displacement and velocity if downsampling was used? Finally, how to know the stable reference point used in step1-3 and its coordinates? can it be displayed in a .txt file?

contour2

yumorishita commented 1 year ago

The resolution can be found in info/EQA.dem_par and the reference point information can be found in info/13ref.txt and info/13ref.kml.

Sakina-A commented 1 year ago

Thank you. what about the contour error?

yumorishita commented 1 year ago

I see the folder name results/Today, not results.

Sakina-A commented 1 year ago

No the word today is a filter I used it to make it easier to show the files I want to contour it because the folder has a large number of files.

yumorishita commented 1 year ago

I think the suffix .tif is not shown in the explore. Check the full name by ls and try e.g., vel.mskd.geo.tif.

Sakina-A commented 1 year ago

It worked! If I want to produce a txt file of time series of displacement using LiCSBAS_cum2tstxt.py but for the whole area (or clipped area) instead of a point will this be possible?

yumorishita commented 1 year ago

You can iterate LiCSBAS_cum2tstxt.py for all the points you want though it would not be a realistic solution if the number of points is huge. Currently, LiCSBAS has no function to export all the time series to the text format at a time. Exporting all the epochs to GeoTIFF and converting the GeoTIFF to txt would be the easiest way.

vtsironi commented 1 year ago

Hello! I have a question about your previous comment. How can I export all the epochs to Geotiff from LicSBAS? With the LiCSBAS_flt2geotiff.py? and where I can found all the epochs? Thank you in advance!

yumorishita commented 1 year ago

LiCSBAS_cum2flt.py and LiCSBAS_flt2geotiff.py can do it. All the epochs can be found in info/13used_image.txt.

ditafaith commented 1 year ago

Hi, How I can get result in geotiff format with GMT.polar option from float file ?

yumorishita commented 1 year ago

Sorry but I don't know. Maybe there is a function in GMT to convert geotiff.

ditafaith commented 1 year ago

Sorry but I don't know. Maybe there is a function in GMT to convert geotiff.

How you can make vertical result with colour (blue to red) as shown your paper for each pictures ?

yumorishita commented 1 year ago

You mean the color map! I thought GMT.polar meant a kind of map projection. I used QGIS where various color maps are available to make figures. You can also create colored GeoTIFF using LiCSBAS_color_geotiff.py with -c GMT.polar after creating data GeoTIFF from a float file by LiCSBAS_flt2geotiff.py.

ditafaith commented 1 year ago

You mean the color map! I thought GMT.polar meant a kind of map projection. I used QGIS where various color maps are available to make figures. You can also create colored GeoTIFF using LiCSBAS_color_geotiff.py with -c GMT.polar after creating data GeoTIFF from a float file by LiCSBAS_flt2geotiff.py.

Thank you for making licsbas more enjoyable.

ditafaith commented 1 year ago

I guess you used varidis for annual LOS of each frame. What about the vertical deformation velocity ? I'd glad if you can share the .txt template as you used.

yumorishita commented 1 year ago

I used SCM.vik in Scientific colour maps. https://github.com/yumorishita/LiCSBAS/blob/master/LiCSBAS_lib/SCM/vik/vik.txt

ditafaith commented 1 year ago

I used SCM.vik in Scientific colour maps. https://github.com/yumorishita/LiCSBAS/blob/master/LiCSBAS_lib/SCM/vik/vik.txt

I can't load colour map use file vik.txt at qgis. It looks error or unrecognized format.

yumorishita commented 1 year ago

The text file is not in QGIS format. You can download the XML for QGIS in the Scientific color maps. Please see p12 in https://www.fabiocrameri.ch/colourmaps-userguide/

ditafaith commented 1 year ago

The text file is not in QGIS format. You can download the XML for QGIS in the Scientific color maps. Please see p12 in https://www.fabiocrameri.ch/colourmaps-userguide/

I did, thanks your support as always. Furthermore, do you know on how this works (https://doi.org/10.1029/2020GL087376) generate LOS without same reference point for whole continental scale ? On p4, it mentioned each frame has its own independent reference point for velocity determination. It looks as different work-flow as yours (2021).

yumorishita commented 1 year ago

It used GNSS data in the post-processing as written in the paper. Please read the paper and related references.