Page 1 of 1

Suggestion for APOD

Posted: Fri Jul 06, 2007 3:22 pm
by Pegasus
I don't know if this will get back to the folks who decide what to choose for the APOD, but I'll post it here anyway. :D I would love to see an APOD that shows the difference between an untouched versus touched up pic of a galaxy or nebula, for example. So often we see that the pic is a "false color image." I understand that this is so that we can make out what we're looking at better. I'm not an astronomer (just a fan), so I'm left not knowing what a "real color" pic looks like and how different the false/enhanced color image is from the raw image that it starts out with. I'd love to see an APOD that talks about this and shows examples. Thanks to anyone who will kindly pass along this suggestion, or let me know where else I should send the suggestion.

Posted: Fri Jul 06, 2007 5:14 pm
by BMAONE23
Pegasus,
An interestng thought! They could have the images linked to a visible light image that appears when you place your curser over the false color image

remove extraneous objects from photos?

Posted: Fri Jul 06, 2007 6:04 pm
by billmorley
I realize that it may not be entirely "scientifically honest", but I believe that it would still be beneficial if APOD would "clean up" photos of distant deep sky objects by removing the foreground, and therefore irrelevant, stars from those photos.

In my opinion, the foreground stars in our galaxy that appear in the APOD photos of distant galaxies are subject to misinterpretation as relevant extraordinarily bright objects close to the subject galaxy.

If the subject of the photo is a lone galaxy, for example, I think we should see the photo of that galaxy in an apparently dark empty setting. Foreground stars just cause confusion.

Re: Suggestion for APOD

Posted: Fri Jul 06, 2007 6:05 pm
by jgabany
Hello Pegasus:

You have touched on an interesting aspect about astrophotography that I constantly wrestle with when producing a new astronomical image.

Interestingly, the hues shown in the image on today's APOD are not artificial or falsely colored. They have simply been exagerated through increased saturation.

Producing an astronomical image requires the photographer to increase contrast because the raw images are often very shadowy in appearance. By stretching the variances between dark and light areas, structures hidding in the shadows become more apparent. However, doing this causes the color to become very pale. The solution is to boost the color saturation.

The amount of saturation increase is subjective. I have found that folks tend to like more saturation than less because it makes it easier to recogize hues and that results in a more interesting picture.

For what it's worth, if you could climb inside a spaceship and fly close to this object, or most others for that matter, you would be very disappointed in what you saw outside your porthole window.

Most astronomical subjects (like galxies and nebulas) have a very low surface brightness therefore they appear very shadowy and exhibit little if any color other than green. This is because our eyes need a certain amout of light before their color receptors are stimulated. Since we have more green receptors than those for other colors, most astronomical objects, other than planets and stars, look green even through the largest telescopes on Earth. You would have a similar impression if you could physically travel much closer to them.

Therefore, most astrononical images of galaxies and nebulae are actually a graphic representation (versus a portrait) of the subject.

Re: remove extraneous objects from photos?

Posted: Fri Jul 06, 2007 6:16 pm
by jgabany
billmorley wrote:I realize that it may not be entirely "scientifically honest", but I believe that it would still be beneficial if APOD would "clean up" photos of distant deep sky objects by removing the foreground, and therefore irrelevant, stars from those photos.

In my opinion, the foreground stars in our galaxy that appear in the APOD photos of distant galaxies are subject to misinterpretation as relevant extraordinarily bright objects close to the subject galaxy.

If the subject of the photo is a lone galaxy, for example, I think we should see the photo of that galaxy in an apparently dark empty setting. Foreground stars just cause confusion.

Hi Bill:

This is an interesting idea. In fact, it has been tried and produced some unusal and unexpected results.

Last year, there was an APOD image of the Lagoon Nebula. Although it is situated inside of our galaxy I believe it is still conceptually relevant to your suggestion. Anyway, the photographer produced two images- one with the foreground stars and one without them. The colors used in the picture were artificially applied to highlight the different materials in this subject, by the way.

The result was kinda cool. Here's the link:

http://antwrp.gsfc.nasa.gov/apod/ap060825.html


Jay

Re: remove extraneous objects from photos?

Posted: Sat Jul 07, 2007 4:11 am
by billmorley
jgabany wrote:
billmorley wrote:I realize that it may not be entirely "scientifically honest", but I believe that it would still be beneficial if APOD would "clean up" photos of distant deep sky objects by removing the foreground, and therefore irrelevant, stars from those photos.

In my opinion, the foreground stars in our galaxy that appear in the APOD photos of distant galaxies are subject to misinterpretation as relevant extraordinarily bright objects close to the subject galaxy.

If the subject of the photo is a lone galaxy, for example, I think we should see the photo of that galaxy in an apparently dark empty setting. Foreground stars just cause confusion.

Hi Bill:

This is an interesting idea. In fact, it has been tried and produced some unusal and unexpected results.

Last year, there was an APOD image of the Lagoon Nebula. Although it is situated inside of our galaxy I believe it is still conceptually relevant to your suggestion. Anyway, the photographer produced two images- one with the foreground stars and one without them. The colors used in the picture were artificially applied to highlight the different materials in this subject, by the way.

The result was kinda cool. Here's the link:

http://antwrp.gsfc.nasa.gov/apod/ap060825.html


Jay
You are correct, sir. That with stars/without stars image is kinda (actually very) cool. Thanks for that link.

Here's a recent APOD that includes foreground stars and background galaxies. I think it would help us amateurs a little if the irrelevant foreground stuff was masked out of the image.
http://antwrp.gsfc.nasa.gov/apod/ap070615.html

In some photos, I think the foreground stars are like dirt specks on a window pane.

Thanks for considering my modest notion.

Re: remove extraneous objects from photos?

Posted: Sat Jul 07, 2007 1:30 pm
by BMAONE23
jgabany wrote:
billmorley wrote:I realize that it may not be entirely "scientifically honest", but I believe that it would still be beneficial if APOD would "clean up" photos of distant deep sky objects by removing the foreground, and therefore irrelevant, stars from those photos.

In my opinion, the foreground stars in our galaxy that appear in the APOD photos of distant galaxies are subject to misinterpretation as relevant extraordinarily bright objects close to the subject galaxy.

If the subject of the photo is a lone galaxy, for example, I think we should see the photo of that galaxy in an apparently dark empty setting. Foreground stars just cause confusion.

Hi Bill:

This is an interesting idea. In fact, it has been tried and produced some unusal and unexpected results.

Last year, there was an APOD image of the Lagoon Nebula. Although it is situated inside of our galaxy I believe it is still conceptually relevant to your suggestion. Anyway, the photographer produced two images- one with the foreground stars and one without them. The colors used in the picture were artificially applied to highlight the different materials in this subject, by the way.

The result was kinda cool. Here's the link:

http://antwrp.gsfc.nasa.gov/apod/ap060825.html


Jay
Jay,
How long does it take to "Scrub" an image like this? What about the Trifid APOD image from 07-07-07 Any Scrubbing time estimates?

Re: Suggestion for APOD

Posted: Sun Jul 08, 2007 2:24 am
by Pegasus
BMAONE23, you've got the idea. APOD did something similar not long ago (can't find it now) where there were 4 pics of the same thing, each showing a different colour and then the 3 colours put together to form a full colour picture. It was interesting but not quite what I was referring to.

jgabany wrote:Hello Pegasus:

For what it's worth, if you could climb inside a spaceship and fly close to this object, or most others for that matter, you would be very disappointed in what you saw outside your porthole window.

Most astronomical subjects (like galxies and nebulas) have a very low surface brightness therefore they appear very shadowy and exhibit little if any color other than green.
jgabany, thanks for the explanation, even if disappointing. :? I was afraid it would be something like that. When I see all these beautiful nebulae and galaxies, I can't help but imagine how spectacular they would be if I saw them in person. I've wondered if that would be true or if, as you say, I would find something far less exciting.

It's a bit of a strange feeling to see these pictures now and realise that no one will ever see them like that in real life. In a sense, they're not "real," at least in terms of subjective experience. It's too bad. :( Oh well...

Thanks for answering my question. I still think it would make an interesting APOD piece though. :wink:

Pegasus

Re: remove extraneous objects from photos?

Posted: Sun Jul 08, 2007 7:38 pm
by jgabany
BMAONE23 wrote:
Jay,
How long does it take to "Scrub" an image like this? What about the Trifid APOD image from 07-07-07 Any Scrubbing time estimates?
Hi BMAONE23:

Removing unwanted foreground stars and background galaxies from an astronomical photograph can be very challenging when they appear superimposed on the picture's principal subject. If the removal is not carefully handled, then information about the main subject (that you want to keep) can be accidentally erased or distorted in the process.

Removing undesirable objects does not just require that they are erased because simply removing them would leave the image filled with holes! Therefore, something has to be substituted in their place. This becomes a highly subjective process and reduces the value of the picture, for scientific purposes, especially. Generally, the hole created by the removal of the foreground star or background galaxy can be filled by averaging its immediate surroundings, by blurring or by copy and pasting a small sample from a nearby, similar looking, area. The result may make the image appear less confusing but it can also lead to a distortion of the truth.- that's why you do not see this being used in most, in fact hardly any, images.

Having said that, scrubbing an image, like the Trifid APOD from 07-07-07, would probably take a few hours, a day at most, if you wanted to do it really well and inflict the least amount of damage. :)

Jay

Posted: Sun Jul 08, 2007 7:46 pm
by BMAONE23
Jay,
Thanks for the reply I thought it would be time consuming. I tried a very simple diffraction spike removal on a Pleiades image http://antwrp.gsfc.nasa.gov/apod/ap060109.html
and it took about an hour so I can appreciate the time needed.
Here is the updatd image
http://s189.photobucket.com/albums/z159 ... pdated.jpg

Nice images though

Re: removing extraneous info

Posted: Mon Jul 09, 2007 2:53 pm
by billmorley
BMAONE23 wrote:Jay,
Thanks for the reply I thought it would be time consuming. I tried a very simple diffraction spike removal on a Pleiades image http://antwrp.gsfc.nasa.gov/apod/ap060109.html
and it took about an hour so I can appreciate the time needed.

Nice images though
Silly me - I hadn't thought about how much time/expense might be involved.

If the benefit is minimal, scientifically speaking, then never mind all that. But if there is a way to clean up the easy stuff (the big stars that aren't superimposed over the object image), then maybe there's a cost:benefit ratio that'd make that more modest effort worthwhile.

I appreciate your efforts. And I've enjoyed our little dialog. Thanks.

Scrubbing stars

Posted: Wed Jul 11, 2007 12:44 pm
by kjardine
It is not at all out of line to suggest scrubbing stars from certain astronomical images.

In fact, scientists do it themselves if they want to examine the fine detail of a nebula for example.

(This is not to suggest that the good astronomers who run the APOD would have the time to do it themselves.)

This hydrogen-alpha image of the entire sky has had the stars removed to better emphasize the gas emission:

http://www.astro.princeton.edu/~dfink/halpha/

I've started a nebula by nebula commentary of this image here:

http://galaxymap.org/drupal/node/35

on my http://galaxymap.org website.

Re: Suggestion for APOD

Posted: Wed Jul 11, 2007 1:01 pm
by bystander
Pegasus wrote:... I would love to see an APOD that shows the difference between an untouched versus touched up pic of a galaxy or nebula, for example. So often we see that the pic is a "false color image." I understand that this is so that we can make out what we're looking at better. I'm not an astronomer (just a fan), so I'm left not knowing what a "real color" pic looks like and how different the false/enhanced color image is from the raw image that it starts out with. ...
Most "false color" images are just black and white images taken through filters to select certain wavelengths of radiation (not necessarily visible light). The seperate pictures are then "colored" with colors representing the different wavelengths and recombined.

Even most "true color" images are treated the same way. Black and white images taken through red, green and blue filters, colored red, green and blue, and recombined.

Posted: Wed Jul 11, 2007 3:35 pm
by Pegasus
Thanks for replying, Bystander. It sounds like what I was hoping APOD would do would still have to be approximated manually (if that's a good word for it). From what you're saying, it seems there are very few actual raw color images. Would that not have changed with digital photography, though? I can see the old film pics being that, but did I read on APOD recently that digital photography has allowed us to detect better color? Or is that still starting with b&w taken through a color filter?

See how complex this can get? I think it would make a great APOD subject! :D

P

Posted: Wed Jul 11, 2007 4:17 pm
by bystander
It is my understanding that due to the long exposure times required, b&w imaging provides the best contrast and relative exposure values. See an explanation of the process:

http://www.noao.edu/outreach/aop/glossary/lrgb.html

Posted: Wed Jul 11, 2007 4:49 pm
by Pegasus
Hmm... Not exactly your days of Polaroid snapshots, is it? :wink: Although... in terms of what's involved in putting the image together, it's probably not that far off!

Fascinating stuff. I guess what I'm left still puzzled about is what that nebula would look like if I took the local space bus and went to view it in person, or what would it look like through a visual telescope. Would it have discernible color, and if so, how saturated or bright would those colors be? Or would it look more like the b&w images?

From what BMAONE23 was saying, it sounded like it would look more like the b&w. The colorized pics look so beautiful, I've always hoped they would look more like that in real life too. :?

P

Starship vision

Posted: Wed Jul 11, 2007 5:32 pm
by kjardine
I've always imagined that starships would carry along their own Hubble-like photon collectors and instantly process the images. So the beautiful images would appear on the starship bridge - and include infrared, ultraviolet, radio, gamma, x-rays - the whole works. The Captain would say things like "zoom in on that star and emphasize the X-ray spectrum".

If you put on a space suit and floated outside, most of the time you would just see faint black and white. Our eyes are fine for Planet Earth but quite inadequate for space!

Posted: Wed Jul 11, 2007 6:48 pm
by Pegasus
Kjardine, I've often wondered why the folks behind various scifi series never thought of that. I think you're probably right about the special viewing ability on a starship bridge. We'll probably have individual ocular instruments (like Geordi on ST-TNG) that will do the same thing, too.

I see what you mean about our eyes not being suited to pick up more detail and light in space.

P

Posted: Wed Jul 11, 2007 7:26 pm
by BMAONE23
I could see a heads up display that images from a camera mounted on top of your space helmet and projects/imposes the image over the inside of the face glass of the helmet so you could see deeper as an optional view. That could likely be done today.

Starship visuals

Posted: Wed Jul 11, 2007 8:58 pm
by kjardine
One modification on what I said before.

A starship would need to travel at speeds at a reasonable fraction of the speed of light in order to get very far.

So material in front would get blue shifted and material behind would get red shifted.

A starship imaging system would need to correct for that.

Posted: Thu Jul 12, 2007 6:42 am
by jgabany
bystander wrote:It is my understanding that due to the long exposure times required, b&w imaging provides the best contrast and relative exposure values. See an explanation of the process:

http://www.noao.edu/outreach/aop/glossary/lrgb.html
Hello bystander:

When you gaze towards a clear moonless night sky, the stars appear as points of light - most are colorless. There are a few exceptions, however: Mars, Aldebaran and the star at the heart of the constellation Scorpius, Antares, can be seen to have a very slight reddish hue. Through a small telescope, star and planetary colors become more apparent but galaxies and nebulas remain un-pigmented and monochromatic. These objects begin to take on a greenish ting when viewed through very large telescopes but rarely show the rainbow of hues seen in many deep space pictures.

This begs the question that is often asked of astrophotographers: are those the real colors or did you make them up?

The human eye's retina contains two types of photoreceptors called rods and cones. There are about 120 million rods compared to approximately 7 million cones. Rods are more sensitive to light but only cones detect color. This is why we can make out objects that surround us, in dimly lit situations, but we cannot discern their hue. Light is comprised of three primary colors, red, blue and green. Of these, the cones in our eyes are most sensitive to the later, which makes some evolutionary sense if your ancestor's survival was dependant upon discerning plants (plus, our Sun's light peaks in green wavelengths).

Astronomical telescopes are essentially used for two purposes: 1) to help separate distant but closely spaced objects and 2) to collect a lot of light. The amount of light collected by even the world's largest telescopes is still insufficient for the cones in our eyes to detect color in faint nebula and galaxies other than green. Therefore, the full color of distant astronomical places, other than stars and planets, is something that still eludes direct observation. It should be noted, however, that there have been some rare claims of seeing other colors by a few observers who may simply have eyes with more color sensitivity.

But film and digital cameras do not have this type of color bias. Film emulsion contains crystals that are sensitive to each of the three primary colors of light. Color digital cameras place microscopic red, green or blue filters on top of their grayscale sensing pixels then colorize them inside the camera. Manufacturers use various schemes to place these filters, it should be noted, but here's the point: only a portion of the pixels (in any color digital camera) are dedicated to one color. Regardless, this enables cameras to detect color much more efficiently than human eyes.

Digital astronomical cameras go one-step further- they use every pixel for each color.

Cameras specifically designed for taking deep space images are unsurpassed for detecting very faint light but they only produce results in black and white. To create a full color picture, astronomers, both professional and amateur, place a red, green or blue filter in front of the camera so that every pixel is limited to detecting one specific color reflecting or shining from the astro-subject. This, by the way, is a very time consuming process. To create a full color picture, the astronomer digitally combines separate red, green and blues images using commercially available software like Photoshop. Thus, the colors seen in deep space objects taken through a camera are very real and, unless mis-handled during processing, they are also accurate.

Interestingly, false color images also are created by placing a filter in front of the astronomical camera. However, these filters are not broadly tuned to a specific color- they only pass the glow emitted by specific ionized atoms. These are usually the wavelengths associated with a particular element such as hydrogen, sulfur or oxygen- these are just examples, there are many others. Astronomers will assign a color to each of the filtered images produced using this technique so that color can be used as a method of visulizing not only how the astronomical subject physically appears but what it is chemically made of. For example, hydrogen atoms are often tinted green, sulfur is colored red and the hue for oxygen is blue. Of course, there is nothing to prevent different color assignments.

I apologize for the length of this post but it seems that there is a lot of curiosity about how astro images are produced and I thought this might help...


Jay

Posted: Thu Jul 12, 2007 12:29 pm
by jimmysnyder
jgabany wrote:This begs the question that is often asked of astrophotographers: are those the real colors or did you make them up?
No, it raises the question. "Begs the question" means something else.
http://en.wikipedia.org/wiki/Begging_the_question