• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

Photo Scenery process

K

Kack911

Messages
97
Country
unitedstates
Hello all,

I've looked through the wiki, and searched the forums, but unfortunately, I'm not finding the information I need. Here's my problem...

I obtained 1ft imagery from the state of Ohio, which I trimmed down to cover the airport boundary. The result is a .TIF file with the dimensions of 17,248 x 13,200. I broke that down into 2048 x 2048 chunks, converted to DXT3 with mips, and applied them to plane objects in 3ds Max.

I exported, tweaked the .asm and compiled into the .bgl, which loads in FSX just fine. Performance is fantastic, but I have these gray lines around each of the photo tiles that totally ruin the scenery.

I've read about the technique where you're supposed to apply a "Planar" UVW map to the tiles, and I did that to no avail (though it shows up correctly in Max using that method).

I've also read about copying the outside rows of pixels and then resizing the image. I can't do that because messing with the dimensions of the tiles causes a noticeable misalignment given the high resolution.

So despite my apparent success of getting the imagery into FSX, the result is unacceptable due to the grid lines.

So next I decided to try and create "photoscenery" using tools like SBuilder, Resample, and FWTools. Unfortunately, the TIF I downloaded from the state's website does not seem to be Geo-referenced. It has a bunch of files in addition to the actual image, but the data appears to be in pixel offsets, not Lat/Longs.

To complicate matters, even if it had been properly geo-referenced, the original image is of the entire county. I exported only a small portion of the whole image.

Without reliable data on the projection type, datum, and lat/longs of my edited TIF file, how in the world am I supposed to use the SDK tools for photoscenery?

Can someone help me either get rid of the seams in my poly-based scenery, or help me determine the necessary information in order to use resample, or SBuilderX, etc...

Thanks!

Nick
 
Hello Nick,

Reading your description it seems that you know the process of creating the photo scenery quite well. There is one thing, however, that was not clear to me. Do you actually know the 2 lat values (north and south) and the 2 lon values (west and east) that define the borders of your 17,248 x 13,200 picture? or do you know these borders for the smaller chunks? If you know, then it is very simple to use the resample tool. You could also use SBuilderX and work on those shunks (I can not say if the whole picture would be tratable in a single job by SBuilderX due to memory) just to get the INF files done automatically.

On the contrary, if you do not know the borders (or 2 distinct points in each picture) or if the picture is not in the correct projection (geographic, WGS84 datum) you need to sort that out before attempting to use them in FSX.

Finally: if you were able to paint planar models with the chunks and place them in FSX, that implies that you know the borders or similar placment information. May be I misunderstood your post!

Kind Regards,

Luis
 
Hi Luis,

Unfortunately, even though the original TIF image was derived from a GIS database, it does not appear to be georeferenced. I don't know the lat/longs of the corners.

Even if I had known the original image's geo info, that file is 130,000 x 135,000 pixel image (52.6Gb). I extracted only a small portion of that file which surrounds the airport (17,248 x 13,200 and 680mb). I further broke that image down into 2048 x 2048 tiles (about 50 of them), which means 1px = 1ft in FSX.

By this point, the thought of preserving and/or determining the precise lat/longs for each of these chunks is daunting to say the least. :(

I was able to accurately place the 3ds Max derived poly's because I know the "spot" lat/longs of each of the runway thresholds. So I simply moved the planes in Max so that I aligned one of the runway thresholds with the scene's origin.

So when I place the .bgl at the threshold's known lat/long, it appears at the correct spot in FSX.

So from my research, it appears that I have two avenues to address my issues:

1) Correct the "seams" that are appearing between adjacent texture tiles, which would allow me to continue to use my prefered method of creation.

or;

2) Abandon the Max creation pipeline, and instead adopt the "photoscenery" approach using the SBuilderX/Resample pipeline.

Of course, I'm having issues with both of those pipelines. I don't know the required geo information to properly follow the "photoscenery" route, nor do I know of a way to fix the seam issue in the "Max" route...

Sorry for the verbose reply, I just want to cover all the bases here. Are there any other methods that I've missed?

Regards,

Nick
 
I imagine your source image from Ohio is distributed in the State Plane Coordinate System for Ohio. Ohio has two projections defined, Ohio North and Ohio South. The unit of measure is the U.S. Survey Foot. Typically the datum used is NAD83.

There should be included a metadata.xml or metadata.html file which indicates the projection and datum used.

Assuming I am correct, using resample on this data will require that it first be reprojected into a geographic projection with units of arc degrees and WGS84 datum. you will need an appropriate GIS application to do this.

scott s.
.
 
Hi Scott,

I had a sneaking suspicion that I might have mistakenly deleted some files that were associated with the TIFF. I re-unzipped the archive and sure enough, there IS an xml file associated with it, and it does specify the NAD83 datum. While I see a reference to "State Plane Ohio South" in the PROJCS section, the PROJECTION section indicates "Lambert Conformal Conic". :confused:

With these files in place, I re-exported the portion of the TIFF that was applicable to my project, but the export process only resulted in the cropped image, a .tab file, and a .tfw file. Neither of which contain lat/long references...only x/y positions in feet.

Did I do something wrong?

I originally used ArcGIS Explorer to connect directly to the OSIP database at:

http://gis1.oit.ohio.gov)

But I didn't find a way to export the data using that software. So instead I found a link to download products by county here:

http://gis1.oit.ohio.gov/geodatadownload/osip.aspx

I downloaded the Montgomery county file which contained the .TIF, .AUX, .SDW, .SID, and .SID.AUX files. The TIF image is viewable using the MrSID viewer, which is what I used to export the area around the airport.

Which brings me back to the point where there doesn't appear to be any georeference attributes associated with the cropped image I exported.

I'm still optimistic that there's a way out of this. I just need the help of somebody smarter than me. :D

Regards,

Nick
 
Last edited:
Hi,

From the information you provide I would go for the resample route and forget the 3DS option. I think with all the files you have all the information is there, you just need to combine it :).

Personally I would try to do this in FWTools, but then you have to be a bit handy with the command prompt. Once you have made the TIF georeferenced with the information provide you can reproject it to WGS84 and clip it to the area you want. You can do all this with FWTools, but you need to figure out which options to apply.
 
Kack911 said:
Hi Scott,

I had a sneaking suspicion that I might have mistakenly deleted some files that were associated with the TIFF. I re-unzipped the archive and sure enough, there IS an xml file associated with it, and it does specify the NAD83 datum. While I see a reference to "State Plane Ohio South" in the PROJCS section, the PROJECTION section indicates "Lambert Conformal Conic". :confused:

With these files in place, I re-exported the portion of the TIFF that was applicable to my project, but the export process only resulted in the cropped image, a .tab file, and a .tfw file. Neither of which contain lat/long references...only x/y positions in feet.

Did I do something wrong?

I originally used ArcGIS Explorer to connect directly to the OSIP database at:

http://gis1.oit.ohio.gov)

But I didn't find a way to export the data using that software. So instead I found a link to download products by county here:

http://gis1.oit.ohio.gov/geodatadownload/osip.aspx

I downloaded the Montgomery county file which contained the .TIF, .AUX, .SDW, .SID, and .SID.AUX files. The TIF image is viewable using the MrSID viewer, which is what I used to export the area around the airport.

Which brings me back to the point where there doesn't appear to be any georeference attributes associated with the cropped image I exported.

I'm still optimistic that there's a way out of this. I just need the help of somebody smarter than me. :D

Regards,

Nick
Click to expand...

Took a quick look at the link. I think you can get the identical orthoimage by using the USGS seamless server. An advantage is that you can create a custom area to download and the metadata from USGS should include the lat/long of the corners of the data download. So even though it is not in the right projection, for purposes of loading into SBX you could probably use the corner data to calibrate into SBX and avoid having to reproject the data.

SPCS Ohio South projection uses the Lambert Conformal Conic with the following parameters:

Lambert_Conformal_Conic:

Standard_Parallel: 40.433333
Standard_Parallel: 41.700000
Longitude_of_Central_Meridian: -82.500000
Latitude_of_Projection_Origin: 39.666667
False_Easting: 1968500.000000
False_Northing: 0.000000

Depending on the shape of a state, the SPCS projections will either be based on the LCC or transerse mercator (TM), differing in the parameters. Most GIS software will already have the parameters from NADCON or other packages so all you need do is select SPCS Ohio South.

But if you are having trouble getting your source data lat/long, give seamless a try:
http://seamless.usgs.gov/index.php

scott s.
.
 
You must log in or register to reply here.
Back
Top