Interesting...

[m_k]

I've found it somewhere in the net...
Just look at the pic:

http://img24.echo.cx/img24/1104/ksc5xx.th.jpg (http://img24.echo.cx/my.php?image=ksc5xx.jpg)

:huh: What do you think? Do you know what is it? A new Landsat imagery?! :P

I know...
I'll leave this question unanswered :)




(Nobody knows? OK I'll post the answer tomorrow :)

[cobito]

I don´t know what is it... maybe a new Landsat imagery, but it is too good for be true. I think the resolution is better than USGS Urban Area (I can see water wave very well).

[Beansprout]

Originally posted by cobito@Apr 11 2005, 08:19 PM
I don´t know what is it... maybe a new Landsat imagery, but it is too good for be true. I think the resolution is better than USGS Urban Area (I can see water wave very well).
Quoted post

He's talking about what the pic shows :P

It's one of the launch pads, no idea what's being launched though.


Skynet!

[m_k]

Originally posted by TheBeansprout@Apr 11 2005, 10:22 PM
He's talking about what the pic shows :P
It's one of the launch pads, no idea what's being launched though.
Quoted post

No, no...
I mean the way the picture was made/obtained
(Pic shows Kennedy Space Center, Florida
Hint: see it in World Wind, change the image layer... ;)
Still nothing?)

[xipc]

Originally posted by m_k@Apr 12 2005, 02:29 AM
No, no...
I mean the way the picture was made/obtained
(Pic shows Kennedy Space Center, Florida
Hint: see it in World Wind, change the image layer... ;)
Still nothing?)
Quoted post


It's a color photograph of the USGS 1m Orthoset. Of course, in Worldwind it hows only in black-and-white....

This can be clearly established by the various reflections in the see waves, which are caued by parts of teh image taken at different times. They are identical in your color pic and the USGS 1m data...

Actually, if you look closely, you see that the Worldwind-data does not show the redish color variation in the vegatation seen in the the top-left corner of your color pic.

So I assume that the black-and-white worldwind data is either the blue channel or another channel taken by the scanner.

This is interesting... :) - it means that the USGS 1m data is actually somewhere present as a multi-channel image...

Btw: it can not be a merge between LANDSAT color data & USGS 1m data - the LANDSAT data does also not show the color variation noted above....

Where did you found the image? :) :)

[cobito]

I only can see it with USGS Digital ORTHO. The other layers doesn´t works in this place.
You can see an image here (http://www.galeon.com/cobitoo/capecanaveral.jpg).

[m_k]

Originally posted by xipc@Apr 12 2005, 01:33 PM
It's a color photograph of the USGS 1m Orthoset. Of course, in Worldwind it hows only in black-and-white....

This can be clearly established by the various reflections in the see waves, which are caued by parts of teh image taken at different times. They are identical in your color pic and the USGS 1m data...


Btw: it can not be a merge between LANDSAT color data & USGS 1m data - the LANDSAT data does also not show the color variation noted above....

Quoted post

Yesss... :)
Indeed, it's the USGS 1m combined with Landsat (pseudo color it seems) in a graphic program. Don't know exactly but I think it was combined from HSL channels - Luminance taken from USGS, Hue and Saturation from Landsat.
(I found it on a Polish astro forum - I will try to make similar images) :)

[xipc]

Originally posted by m_k@Apr 12 2005, 04:28 AM
Yesss... :)
Indeed, it's the USGS 1m combined with Landsat (pseudo color it seems) in a graphic program. Don't know exactly but I think it was combined from HSL channels - Luminance taken from USGS, Hue and Saturation from Landsat.
(I found it on a Polish astro forum - I will try to make similar images) :)
Quoted post


well, are you sure that it is a such a combination? The Landsat-data I know does not show the color variation (Hue + Saturation) of the vegation which is visible in the top-left corner of your image....

Also, some of the color information visible in your pic are too small-scaled to be recorded by LANDSAT, for example near the launch pads. So I still assume that the color data comes from other satellite imagery....

:)

[TomServo]

Nowak has been talking off an on about panshaping LandSat with 1-meter magery a few times... he hasn't done anything yet about that.. but it can be done.

[m_k]

Originally posted by xipc@Apr 12 2005, 06:03 PM
well, are you sure that it is a such a combination?
Quoted post

OK Here is what i've done:
* take a screenshot of Landsat pseudo color (level 4?)[attachmentid=547]
* then screenshot of USGS 1m (level 6)[attachmentid=548]
* split the first image to HSL channels in Paint shop pro
* convert USGS screenshot to greyscale ;)
*combine Hue and Saturation with USGS as a luminance
* voila! B) [attachmentid=549]

[bull]

Very cool, now do all the other tiles ;) nice work, maybe someone will take up the challenge and process all the data now they can see it is possible.

[TomServo]

Well.. I couldn't work out EXACTLY how to do that in Photoshop 7...

But here is what I managed to do:

[attachmentid=552]

Used layer blending in my example. I'll play around and improve it some later.

[Mongol]

How about this picture....
http://earthobservatory.nasa.gov/Newsroom/NewImages/Images/ISS010E23035_lrg.jpg
on Rollout of Shuttle Discovery, Kennedy Space Center (http://earthobservatory.nasa.gov/Newsroom/NewImages/images.php3?img_id=16874)


International Space Station Astronaut Leroy Chiao, like the rest of NASA, tracks key milestones for the Space Shuttle Return-to-Flight operations. A lucky overpass of the Space Station over Florida on April 6, 2005, allowed Leroy and his crew mate Salizhan Sharipov a unique view of the rollout of the Space Shuttle Discovery. At the time of his observations, Discovery was approximately midway between the Vehicle Assembly Building (VAB) and launch pad 39-B at Kennedy Space Center in Florida.

The Shuttle, mated with its solid-fuel booster rockets and orange external fuel tank, is visible on the wide Crawler-Transporter track (marked with white box and shown in an inset image). Moving at a maximum speed of 1 mile per hour (1.6 km/hr), the massive Crawler-Transporter, laden with a launch-configured Shuttle and Mobile Launch Platform, takes approximately 13 hours to move between the VAB and pad 39-B.

Launch of the Discovery will mark NASA’s “Return to Flight� following the loss of the Shuttle Columbia with its crew on February 1, 2003. The primary goal for the upcoming Discovery mission will be to test several vehicle modifications and damage-assessment data sources, tools, and procedures. These modifications include a modified design for the external fuel tank, more cameras focused on the launch and ascent of the Shuttle, and detailed imaging of the Shuttle’s heat-resistant tiles and reinforced carbon-carbon surfaces on the wings and nosecap.
http://earthobservatory.nasa.gov/Newsroom/NewImages/Images/ISS010E23035.jpg
Astronaut photograph ISS010-E-23035 was acquired April 6, 2005, with a Kodak 760C digital camera with an 800 mm lens, and is provided by the ISS Crew Earth Observations experiment and the Image Science & Analysis Group, Johnson Space Center. The International Space Station Program supports the laboratory to help astronauts take pictures of Earth that will be of the greatest value to scientists and the public, and to make those images freely available on the Internet. Additional images taken by astronauts and cosmonauts can be viewed at the NASA/JS

[5of0]

Concerning NASA's "Return to Flight" program, for those who are interested, there is an excellent (and rather stinging) article on it in this month's Popular Science, online at their website (http://www.popsci.com/popsci/aviation/article/0,20967,1048015,00.html).

[Guest]

Originally posted by m_k@Apr 12 2005, 08:41 AM
OK Here is what i've done:
* take a screenshot of Landsat pseudo color (level 4?)[attachmentid=547]
* then screenshot of USGS 1m (level 6)[attachmentid=548]
* split the first image to HSL channels in Paint shop pro
* convert USGS screenshot to greyscale ;)
*combine Hue and Saturation with USGS as a luminance
* voila! B) [attachmentid=549]
Quoted post


Yes - very nice work!

By the way: this is exactly the technique which is used to increase the resolution of the color LANDSAT imagery, namely by adding luminance information from a higher resolution channel.

But: if you look closely at your original post, you see a difference in color resolution between the image you created and the one you originally posted! If you do not believe, try to convert the RGB-images into HSI-images and have a look at the H- or S-channel only. You should see difference in the high frequency content of the images....

(Also notice the color bleeding visible in your and TomServo's example around roads and other small objects)

The higher resolution of your original post means that it was not created by using LANDSAT color - some higher resolution color data must have beeen used!

:) - xipc

[m_k]

Originally posted by Guest@Apr 13 2005, 11:02 AM
But: if you look closely at your original post, you see a difference in color resolution between the image you created and the one you originally posted! If you do not believe, try to convert the RGB-images intoÂÂ* HSI-images and have a look at the H- or S-channel only. You should see difference in the high frequency content of the images....
Quoted post
I think i get it... :)
TomServo was right - it's layer blending.
We don't get psychedelic colors and after splitting to HSL we have same effect with channels (saturation looks more detailed)...

Left: combined from HSL
Right: layers blending
[attachmentid=562]
(Colors from NLT Landsat)

[5of0]

I managed to do this in GIMP (www.gimp.org):
1. Open the two files (USGS and LandSat)
2. Go to the USGS. Click "Image->Mode->Decompose". Select "HSV"
3. This will create a new window with an image in it. Close the original window.
4. Repeat steps 2 & 3 on LandSat.
5. Switch to USGS. Click Edit->Copy
6. Back to LandSat. Edit->Paste
7. In the Layer Manager, right-click on the pasted layer and click "New Layer."
8. Select the "Value" layer and click the trashcan on the lower-right to delete it.
9. Right-click on the new, untitled layer and click "Edit Attributes". Rename it "Value"
10. Click "Image->Compose"
11. Select the "HSV" radio button and choose the correct layer for each (Hue for Hue, Saturation for Saturation, Value for Value.)

If everything went well, you should have a new, colored and detailed picture.
Go GIMP. :)

[Guest]

Originally posted by m_k@Apr 13 2005, 03:20 AM
I think i get it... :)
TomServo was right - it's layer blending.


so, we won't never have the "Layer Blending Plugin" ?
:(

just to stop all the crap about GE having higher res imagery. and for the beauty of it of course. :D

[turboii]

here is a raw screen shot from the Google Earth program (just like world wind).