[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #YYJ-216056]: MSG Navigation inverted along Scan-Axis

Hi Ghansham,

I filed this issue: https://github.com/Unidata/thredds/issues/786

Thanks, it is nice to have such an expert user finding these sorts of problems.

Best,

Unidata IDV Support

> Dear Sir
> 
> I am pretty sure about this. Because I am using the same datasets for
> thredds WMS Service (ncwms).
> And there it is coming quite nicely.
> If you see the documentation of MSGNavigation (in comments), it mentions
> that CFAC and LFAC sign reversal will
> take care of the direction of orientation of image. It says
> "The sign of CFAC/LFAC gives orientation of the image.
> Negative sign of CFAC/LFAC give data scanned from south to north as the
> operational
> scanning and positive sign vice versa"
> But seems like the original C code has be restructured and in new code
> these variables are not there.
> 
> Just let me know if I can be of any help.
> 
> And regarding writing a jython code is not at all an issue. I have
> written full fledged customized plugin for all this.
> But I wanted that the default code could have worked for it.
> Even the SSEC guys  (TommyJ Sir) have written ADDE for these hdf5/nc
> files and the MSGNav that is part of addenav seems to
> work pretty nicely.
> 
> 
> 
> 
> regards
> Ghansham
> 
> On 03/20/2017 11:39 PM, Unidata IDV Support wrote:
> > Hi Ghansham,
> >
> > Assuming you are sure the metadata concerning the orientation are described
> > correctly (please verify this) than the problem lies inside the
> > MSGnavigation.java class, specifically
> >
> > geocoord2pixcoord, pixcoord2geocoord methods. In fact, doing a simple sign
> > reversal in these methods addresses the problem.
> >
> > Yes, I believe the netcdf-Java group has done away with Jira. Now we use 
> > github
> > issues, and in fact, I have a github issue concerning this topic ready to 
> > go if
> > you think the metadata are correct.
> >
> > https://github.com/Unidata/thredds/issues
> >
> > I am not sure whatever happened to the original Jira issue you refer to. I
> > quickly searched for it in the github issue, but could not find it.
> >
> > Best,
> >
> > Unidata IDV Support
> >
> >>> Dear Sir
> >>> Thanks for the globe view suggestion. I am aware of that.
> >>> Display seems to be down only in IDV because the way the projection box 
> >>> is built. otherwise if you just see the variable as image
> >>> using toolsUI it quite OK (I mean its NOT upside down).
> >>> AFAIK, John Caron Sir fixed similar issue for geostationary projection 
> >>> (ucar.unidata.geoloc.projection.sat.Geostationary) reported by me.
> >>> Although I never got a chance to test it.
> >>> Seems like ucar.unidata.geoloc.projection.sat.MSGNavigation class needs 
> >>> similar treatment.
> >>> I could have helped you telling exactly the commit that he did for 
> >>> Geostationary class
> >>> but unfortunately the jira bug tracking site 
> >>> (https://bugtracking.unidata.ucar.edu/browse/TDS) seems to be down.
> >>>
> >>> Has it been migrated to some other link.
> >>> Interestingly that commit is not being visible in the history of that 
> >>> class in github. Is it that commit was done some other branch?
> >>> It was done long time back (may be more than a year or two old).
> >>> You can check with him.
> >>> regards
> >>> Ghansham
> >>> On Fri, Mar 17, 2017 at 20:12, Unidata IDV Support  wrote:Dear Sir
> >>>
> >>> I have uploaded "3DIMG_13MAR2017_0530_L1B_STD_4.nc" file.
> >>>
> >>> regards
> >>> Ghansham
> >>>
> >> Hi Ghansham,
> >>
> >> The satellite imagery dataset normally starts from the upper left conner, 
> >> and the gridded dataset starts from lower left. We all know this is why 
> >> you have the upside down result.
> >> The quick and best solution I am thinking is to write an Jython code to 
> >> modify the dataset (it should be 10 lines code if you are better than 
> >> writing "hello world") and create a plugin to define
> >> this special gridded dataset. Waiting for the CDM library to make the 
> >> change will take  a while.
> >>
> >>
> >> Yuan
> >>
> >>> Hi Ghansham,
> >>>
> >>> To summarize, the display seems to be simply upside down. The netCDF-Java
> >>> library is indeed a good place to start looking, perhaps in the classes 
> >>> that
> >>> deal with projections. While I dig into this problem, there is a quick
> >>> workaround. Plot the image on the globe view. See attached.
> >>>
> >>> Best,
> >>>
> >>> Unidata IDV Support
> >>>
> >>> Ticket Details
> >>> ===================
> >>> Ticket ID: YYJ-216056
> >>> Department: Support IDV
> >>> Priority: Normal
> >>> Status: Open
> >>> ===================
> >>> NOTE: All email exchanges with Unidata User Support are recorded in the 
> >>> Unidata inquiry tracking system and then made publicly available through 
> >>> the web.  If you do not want to have your interactions made available in 
> >>> this way, you must let us know in each email you send to us.
> >>>
> >>>
> >
> > Ticket Details
> > ===================
> > Ticket ID: YYJ-216056
> > Department: Support IDV
> > Priority: Normal
> > Status: Open
> > ===================
> > NOTE: All email exchanges with Unidata User Support are recorded in the 
> > Unidata inquiry tracking system and then made publicly available through 
> > the web.  If you do not want to have your interactions made available in 
> > this way, you must let us know in each email you send to us.
> >
> >
> 
> 
> 


Ticket Details
===================
Ticket ID: YYJ-216056
Department: Support IDV
Priority: Normal
Status: Open
===================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata 
inquiry tracking system and then made publicly available through the web.  If 
you do not want to have your interactions made available in this way, you must 
let us know in each email you send to us.