GUI based Photorealistic Condor Scenery Tutorial - draft

Everything related to creation of new sceneries for Condor...

Moderators: Uros, OXO

Post Reply
User avatar
Andy1248
Posts: 1423
Joined: Wed Nov 29, 2006 3:06 pm
Location: Biggleswade, Bedfordshire

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by Andy1248 » Sat Apr 29, 2017 10:07 pm

Wow! fast reply. Thanks Tom, I will try that.

That worked Ok and I now have calibration points data.

One more question;

How do I save the SS so that I can access it with the scenery data again?
Condor CN = E20

luisbriones
Posts: 40
Joined: Fri Nov 14, 2014 4:56 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by luisbriones » Tue May 02, 2017 2:15 am

Hi Tom,

I tried to follow your tutorial but I not understand someting.

When you use 3DEM you don't convert the file to UTM at this moment. I didn't find that you made that conversion at all. Maybe it my mistake because is not easy reading English to me.

With no convertion and using your calibration data from your spreashet, the distance between wp is incorrect

I don't know if I lost something when I read or have another mistake.

BTW, I try to use the SAS.Planet program to obtain satellites images

Regards,

Luis

User avatar
Jan Oorthuijsen
Posts: 491
Joined: Sat Jul 02, 2005 4:27 pm
Location: Utrecht (Terwijde) - Holland

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by Jan Oorthuijsen » Tue May 02, 2017 8:34 am

Hi luis,

Read my questions page 3 and 4.

Grts, Jan
PH-722
WW
It’s Difficult to Soar Like An Eagle
When You’re surrounded By Turkeys

tberry
Posts: 91
Joined: Thu Aug 09, 2012 4:10 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by tberry » Wed May 03, 2017 1:04 am

"How do I save the SS so that I can access it with the scenery data again?"

The spreadsheet is a template. When you open it, you will get a Google Sheets file that is yours. Here is more info: https://support.google.com/docs/answer/ ... ktop&hl=en

tberry
Posts: 91
Joined: Thu Aug 09, 2012 4:10 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by tberry » Wed May 03, 2017 1:04 am

"With no convertion and using your calibration data from your spreashet, the distance between wp is incorrect"

Luis, the SRTM data is already in UTM format. As Jan notes we discussed this on pages 3/4 of this thread. Are you at a high latitude?

After much investigation, I have concluded that this distortion is the result of Condor's method for creating scenery. The scenery is represented as a grid of pixels and those pixels are assigned a certain size. Because the grid is square and latitude lines are NOT square, this distortion is inherent in the method for scenery creation and it becomes more apparent the closer one gets to either pole.

I believe the source of this is the SRTM data and it is the reason that NASA does not supply SRTM data near the poles - the distortion is too great.

luisbriones
Posts: 40
Joined: Fri Nov 14, 2014 4:56 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by luisbriones » Wed May 03, 2017 2:29 am

Jan Oorthuijsen wrote:Hi luis,

Read my questions page 3 and 4.

Grts, Jan
Thanks Jan for your anwser. I read it. The problem is complex. I'm lost!!! :-)

luisbriones
Posts: 40
Joined: Fri Nov 14, 2014 4:56 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by luisbriones » Wed May 03, 2017 2:48 am

tberry wrote:"With no convertion and using your calibration data from your spreashet, the distance between wp is incorrect"

Luis, the SRTM data is already in UTM format. As Jan notes we discussed this on pages 3/4 of this thread. Are you at a high latitude?

After much investigation, I have concluded that this distortion is the result of Condor's method for creating scenery. The scenery is represented as a grid of pixels and those pixels are assigned a certain size. Because the grid is square and latitude lines are NOT square, this distortion is inherent in the method for scenery creation and it becomes more apparent the closer one gets to either pole.

I believe the source of this is the SRTM data and it is the reason that NASA does not supply SRTM data near the poles - the distortion is too great.
My airport is at -39º, -67.62º. I'm closer to equator than you :-)

That are my data in the sreadsheet:

Code: Select all

Input data		
		
						Longitude (X)	Latitude (Y)
Desired scenery center pont	-67.62000000	-39.00000000
						Width (X)	Height (Y)
Desired scenery size in terragens	10	14

Scenery clipping details	Point 1 (upper left)	Point 2 (lower right)
				X	-68.68662400	-66.55337600
				Y	-37.50672640	-40.49327360
My reference wp is S38°59'21", W067°17'25". It's 50.8km away to my airport to the east. All my wp and airport look good in map, but east-west distance are incorrect
task.jpg
You do not have the required permissions to view the files attached to this post.

luisbriones
Posts: 40
Joined: Fri Nov 14, 2014 4:56 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by luisbriones » Wed May 03, 2017 2:45 pm

I was thinking about the problem and made the following test.
I used excel calibration tools from Lurik and made a new calibration point file. For clarity I only used 4 values (Cells L43 to L45)

0,0,-40.56582186,-66.18171986
230400,0,-40.60023145,-68.90325931
230400,322560,-37.6936068,-68.90715888
0,322560,-37.66256638,-66.29515128

Now the distance is correct as you can see here:
task-1.jpg
Is not a good calibration, all point are displace to north of it real position

I put both calibration files in QGIS, with one of my geotiff images tha I made with SAS.Planet program
QGIS.jpg
I don't know how follow from here. I'll do more test, but any help will be apreciated.
You do not have the required permissions to view the files attached to this post.

luisbriones
Posts: 40
Joined: Fri Nov 14, 2014 4:56 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by luisbriones » Wed May 03, 2017 3:42 pm

Jan Oorthuijsen wrote:Hi Luis

Where is the conversion to the correct UTM projection ?,was my question a few years ago.
Now I know that in Qgis you need to do the conversion through, Raster> Projections> Warp (reproject).
 And than Cut with ,Raster> Extraction> Correct Clipper (Multiple of23040m ) .
Did not get a correct answer and therefore I stopped follow that tutorial.

Do not forget to correct the size (look in hdr file) by -45 m and + 50 m , If you follow the old method
Search the forum for this.

Grts, WW
Jan, I move my answer here, because I think that this discussion is better make in this post.

In what part of the tutorial do you have to do the conversion and the cut?

Sorry about that silly question, but I tested many thing and I a little (big!!!) confused now.

I tried to follow the another method to make a photorealistic scenery, but I have troubles too. I'll write about it in the proper post.

tberry
Posts: 91
Joined: Thu Aug 09, 2012 4:10 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by tberry » Wed May 03, 2017 8:03 pm

Now I see. It does look like the calibration points tab is incorrect. I'll take a look at that and see if it is obvious what is wrong.

I suspect it has something to do with the southern hemisphere location. I originally built the spreadsheet for U.S. locations, then added southern hemisphere adjustments to it. As far as I know, it hasn't been used for that yet and I may not have thought it through completely. Sorry about that.

tberry
Posts: 91
Joined: Thu Aug 09, 2012 4:10 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by tberry » Wed May 03, 2017 8:22 pm

@luisbriones:
In what part of the tutorial do you have to do the conversion and the cut?
The reprojection is done at the end of the tutorial. The tutorial uses images in EPSG:3857 projection and reprojects them to EPSG:4326 (Condor) in the step prior to cutting the full mosaic.

luisbriones
Posts: 40
Joined: Fri Nov 14, 2014 4:56 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by luisbriones » Wed May 03, 2017 9:27 pm

tberry wrote:@luisbriones:
In what part of the tutorial do you have to do the conversion and the cut?
The reprojection is done at the end of the tutorial. The tutorial uses images in EPSG:3857 projection and reprojects them to EPSG:4326 (Condor) in the step prior to cutting the full mosaic.
I saw this and always had a dude about it. You say EPSG:4326. I understand that is WGS84. Does Condor not work on UTM?

By the way, I have partial success to make the images using P. Strzelec method. You are part of this success, because I learned QGIS from your tutorial. It's not finish yet, but look good.

Anyway if you wish I can help you to fixed your spreadsheet problems

tberry
Posts: 91
Joined: Thu Aug 09, 2012 4:10 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by tberry » Wed May 03, 2017 11:57 pm

I don't know how follow from here. I'll do more test, but any help will be apreciated.
I calculated the corner points by hand and came out with the same answer as the spreadsheet. Before we can figure out if there is a calibration error, I want to make sure I can at least get the same corner points as you did with Lurik's spreadsheet.

Let us check a few things first. To make sure I understand:

1. The RED dots are the corners from my spreadsheet. The BLUE dots are the calibration points from Lurik's Excel spreadsheet, correct?

On my Google spreadsheet:
2. What number is in cells J16 and K16?
3. What number is in cells E24 and E25?

The reason I ask is that the width of the scenery is calculated mathematically and is based on values in those cells. Here's how it works:

-Each pixel of the SRTM data represents 90 meters. When you use Raster->Misc->Info on the SRTM data, it gives you the pixel size in decimal degrees and that number should be in cells E24 and E25 (it is the same for both width and height).
-Each terragen tile is 256 pixels of SRTM data. So, a terragen tile is 90meters X 256 pixels = 23,040 meters in width.
-Your scenery is 10 terragen tiles in width, so 23,040 X 10 = 230,400 meters in width. You may notice this value in the calibration points you posted.
-230,400 meters X the value in E24 (pixel size in decimal degrees) = width of the scenery across those 10 terragen tiles in decimal degrees.

I calculate the width of your scenery to be 2.133248 decimal degrees. The values you posted show it to be 2.72153945 degrees, which is slightly wider than I calculated.

I use that width to figure out the upper left, upper right, lower right, lower left corners of the scenery. If the pixel size in E24/E25 is changed, this would introduce an error in the size.

I still think there's probably an issue with shifting everything north, but first I wanted to figure out the correct corners.

luisbriones
Posts: 40
Joined: Fri Nov 14, 2014 4:56 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by luisbriones » Thu May 04, 2017 12:24 am

tberry wrote: 1. The RED dots are the corners from my spreadsheet. The BLUE dots are the calibration points from Lurik's Excel spreadsheet, correct?
No. The RED are from Lurik and BLUE are yours
tberry wrote: On my Google spreadsheet:
2. What number is in cells J16 and K16?
0.2133248 0.2133248
tberry wrote: 3. What number is in cells E24 and E25?
0.0008333
0.0008333

I send you screen captures from the spreadsheet and the calibration point generated by then.
S-1.jpg
S-2.jpg
S-3.jpg

Code: Select all

92160	92160	-39.63997269	-67.40667725
92160	69120	-39.85329819	-67.40667725
92160	46080	-40.06662369	-67.40667725
92160	322560	-37.50672531	-67.40667725
92160	299520	-37.72005081	-67.40667725
92160	276480	-37.93337631	-67.40667725
92160	253440	-38.14670181	-67.40667725
92160	230400	-38.36002731	-67.40667725
92160	23040	-40.27994919	-67.40667725
92160	207360	-38.573349	-67.40667725
92160	184320	-38.7866745	-67.40667725
92160	161280	-39	-67.40667725
92160	138240	-39.2133255	-67.40667725
92160	115200	-39.426651	-67.40667725
92160	0	-40.49327469	-67.40667725
69120	92160	-39.63997269	-67.19335175
69120	69120	-39.85329819	-67.19335175
69120	46080	-40.06662369	-67.19335175
69120	322560	-37.50672531	-67.19335175
69120	299520	-37.72005081	-67.19335175
69120	276480	-37.93337631	-67.19335175
69120	253440	-38.14670181	-67.19335175
69120	230400	-38.36002731	-67.19335175
69120	23040	-40.27994919	-67.19335175
69120	207360	-38.573349	-67.19335175
69120	184320	-38.7866745	-67.19335175
69120	161280	-39	-67.19335175
69120	138240	-39.2133255	-67.19335175
69120	115200	-39.426651	-67.19335175
69120	0	-40.49327469	-67.19335175
46080	92160	-39.63997269	-66.98002625
46080	69120	-39.85329819	-66.98002625
46080	46080	-40.06662369	-66.98002625
46080	322560	-37.50672531	-66.98002625
46080	299520	-37.72005081	-66.98002625
46080	276480	-37.93337631	-66.98002625
46080	253440	-38.14670181	-66.98002625
46080	230400	-38.36002731	-66.98002625
46080	23040	-40.27994919	-66.98002625
46080	207360	-38.573349	-66.98002625
46080	184320	-38.7866745	-66.98002625
46080	161280	-39	-66.98002625
46080	138240	-39.2133255	-66.98002625
46080	115200	-39.426651	-66.98002625
46080	0	-40.49327469	-66.98002625
230400	92160	-39.63997269	-68.68662262
230400	69120	-39.85329819	-68.68662262
230400	46080	-40.06662369	-68.68662262
230400	322560	-37.50672531	-68.68662262
230400	299520	-37.72005081	-68.68662262
230400	276480	-37.93337631	-68.68662262
230400	253440	-38.14670181	-68.68662262
230400	230400	-38.36002731	-68.68662262
230400	23040	-40.27994919	-68.68662262
230400	207360	-38.573349	-68.68662262
230400	184320	-38.7866745	-68.68662262
230400	161280	-39	-68.68662262
230400	138240	-39.2133255	-68.68662262
230400	115200	-39.426651	-68.68662262
230400	0	-40.49327469	-68.68662262
23040	92160	-39.63997269	-66.76670074
23040	69120	-39.85329819	-66.76670074
23040	46080	-40.06662369	-66.76670074
23040	322560	-37.50672531	-66.76670074
23040	299520	-37.72005081	-66.76670074
23040	276480	-37.93337631	-66.76670074
23040	253440	-38.14670181	-66.76670074
23040	230400	-38.36002731	-66.76670074
23040	23040	-40.27994919	-66.76670074
23040	207360	-38.573349	-66.76670074
23040	184320	-38.7866745	-66.76670074
23040	161280	-39	-66.76670074
23040	138240	-39.2133255	-66.76670074
23040	115200	-39.426651	-66.76670074
23040	0	-40.49327469	-66.76670074
207360	92160	-39.63997269	-68.47329712
207360	69120	-39.85329819	-68.47329712
207360	46080	-40.06662369	-68.47329712
207360	322560	-37.50672531	-68.47329712
207360	299520	-37.72005081	-68.47329712
207360	276480	-37.93337631	-68.47329712
207360	253440	-38.14670181	-68.47329712
207360	230400	-38.36002731	-68.47329712
207360	23040	-40.27994919	-68.47329712
207360	207360	-38.573349	-68.47329712
207360	184320	-38.7866745	-68.47329712
207360	161280	-39	-68.47329712
207360	138240	-39.2133255	-68.47329712
207360	115200	-39.426651	-68.47329712
207360	0	-40.49327469	-68.47329712
184320	92160	-39.63997269	-68.25997162
184320	69120	-39.85329819	-68.25997162
184320	46080	-40.06662369	-68.25997162
184320	322560	-37.50672531	-68.25997162
184320	299520	-37.72005081	-68.25997162
184320	276480	-37.93337631	-68.25997162
184320	253440	-38.14670181	-68.25997162
184320	230400	-38.36002731	-68.25997162
184320	23040	-40.27994919	-68.25997162
184320	207360	-38.573349	-68.25997162
184320	184320	-38.7866745	-68.25997162
184320	161280	-39	-68.25997162
184320	138240	-39.2133255	-68.25997162
184320	115200	-39.426651	-68.25997162
184320	0	-40.49327469	-68.25997162
161280	92160	-39.63997269	-68.04664612
161280	69120	-39.85329819	-68.04664612
161280	46080	-40.06662369	-68.04664612
161280	322560	-37.50672531	-68.04664612
161280	299520	-37.72005081	-68.04664612
161280	276480	-37.93337631	-68.04664612
161280	253440	-38.14670181	-68.04664612
161280	230400	-38.36002731	-68.04664612
161280	23040	-40.27994919	-68.04664612
161280	207360	-38.573349	-68.04664612
161280	184320	-38.7866745	-68.04664612
161280	161280	-39	-68.04664612
161280	138240	-39.2133255	-68.04664612
161280	115200	-39.426651	-68.04664612
161280	0	-40.49327469	-68.04664612
138240	92160	-39.63997269	-67.83332825
138240	69120	-39.85329819	-67.83332825
138240	46080	-40.06662369	-67.83332825
138240	322560	-37.50672531	-67.83332825
138240	299520	-37.72005081	-67.83332825
138240	276480	-37.93337631	-67.83332825
138240	253440	-38.14670181	-67.83332825
138240	230400	-38.36002731	-67.83332825
138240	23040	-40.27994919	-67.83332825
138240	207360	-38.573349	-67.83332825
138240	184320	-38.7866745	-67.83332825
138240	161280	-39	-67.83332825
138240	138240	-39.2133255	-67.83332825
138240	115200	-39.426651	-67.83332825
138240	0	-40.49327469	-67.83332825
115200	92160	-39.63997269	-67.62000275
115200	69120	-39.85329819	-67.62000275
115200	46080	-40.06662369	-67.62000275
115200	322560	-37.50672531	-67.62000275
115200	299520	-37.72005081	-67.62000275
115200	276480	-37.93337631	-67.62000275
115200	253440	-38.14670181	-67.62000275
115200	230400	-38.36002731	-67.62000275
115200	23040	-40.27994919	-67.62000275
115200	207360	-38.573349	-67.62000275
115200	184320	-38.7866745	-67.62000275
115200	161280	-39	-67.62000275
115200	138240	-39.2133255	-67.62000275
115200	115200	-39.426651	-67.62000275
115200	0	-40.49327469	-67.62000275
0	92160	-39.63997269	-66.55337524
0	69120	-39.85329819	-66.55337524
0	46080	-40.06662369	-66.55337524
0	322560	-37.50672531	-66.55337524
0	299520	-37.72005081	-66.55337524
0	276480	-37.93337631	-66.55337524
0	230400	-38.36002731	-66.55337524
0	23040	-40.27994919	-66.55337524
0	207360	-38.573349	-66.55337524
0	184320	-38.7866745	-66.55337524
0	161280	-39	-66.55337524
0	138240	-39.2133255	-66.55337524
0	115200	-39.426651	-66.55337524
0	0	-40.49327469	-66.55337524
You do not have the required permissions to view the files attached to this post.

tberry
Posts: 91
Joined: Thu Aug 09, 2012 4:10 pm

Re: GUI based Photorealistic Condor Scenery Tutorial - draft

Post by tberry » Thu May 04, 2017 12:54 am

The obvious difference is that Lurik's calculations use a larger pixel size.

If you subtract his upper left corner from the upper right corner, you get: 68.903259310-066.181719860 = 2.72153945 degrees width for 10 terragen tiles.
That means each terragen tile in the 10 row width is 2.72153945 / 10 = .272153945 degrees wide.

If you divide that width by 256 (the number of pixels per terragen tile):.272153945 / 256 = .0010631013476563 <--value for cells E24/E25

My value is smaller (0.0008333) and comes from the SRTM data. Thus, the difference between the two spreadsheets.

What I'm having difficulty figuring out is why my calibration points, which are CLOSER together would make distances on your flight planner to be FARTHER apart than Lurik's calculation. I would expect the opposite to be true. Are you SURE that you didn't accidentally use Lurik's calibration points in the flight planner example you posted?

Out of curiousity, what technique are you using to cut the imagery tiles and where are you getting the numbers to cut them from?

Post Reply