Hubble macht gar keine bilder!
Published by Chris Doering,
Die Bilder werden vor der populärwissenschaftlichen Veröffentlichung coloriert, bilden also nicht die Realität ab sondern eine Vorstellung
Farbschema:
red for far infrared
green for middle infrared
blue for near infrared
Are the images from Hubble retouched to be shown to the public?
Follow
6 answers
Report Abuse
Answers
Relevance
Best Answer: They have to.
For one thing, all CCD cameras can only take B & W shots. In order to build a color photo, we must take images through filters.
The simple way is to use the primary colors:
Take one image through a red filter; the camera simply registers the level of light that makes it through the red filter. As far as the camera itself is concerned, it is still taking a black & white picture.
Take another image of the same object, this time with a green filter.
Take a third image with a blue filter.
In many cases, we also take an image without any filters (called "luminosity" shot) in order to make the final image better.
Then, in a computer, we mix the images: the computer puts the appropriate quantity of red color for all pixels where the "red" picture shows white.
Green color where the "green" image shows white and blue for blue. It uses the "luminosity" shot to balance the mix of colors. The final addition of the three (or four) images gives a color photo.
---
In addition:
CCD cameras are very useful for scientific work because they are consistent in the way they record light. If star A is twice as bright as star B, the camera will record exactly twice as much light from A as it does from B.
However, as a "camera", it is not perfect.
1. When you turn the camera on before each image, the pixel values are supposed to reset to zero. They don't. Each pixel begins with a different value. We take an "image" (with zero second exposure) of these residual values and subtract them from the astronomical image (to correct). This image is called the "bias".
2. Some pixels may be affected by processes inside the camera. Most of the time, heat from close-by wires will cause an excess of electrons to be released in some part of the camera. So we take a "dark" image -- same exposure as the astro-image, but with the shutter closed. Thus, the pixels will be subjected to the heat from inside the camera, without receiving any light from the outside. This "image" must also be subtracted from the astro-images.
3. Some parts of the camera could receive more light (or less light) than other parts, due to the design and construction of lenses and mirrors. There is also a problem with dust on the lenses and mirrors. Even though they are greatly out of focus, they may affect the quantity of light that is blocked before reaching the camera. For a earth telescope, we take a "flat" field by taking a picture of the sky while it is still lit by twilight. For Hubble, flats were done in a lab before launch. It is possible to do the occasional flat by taking very long exposures while "scanning" the sky -- allowing the image to roam over the camera, so as to be completely blurred out -- but that costs a lot of time.
4. Putting a telescope in space is costly. Once there, though, it is possible to take pictures in wavelengths that our eyes cannot see (for example ultra-violet and infrared). In such cases, of course, the final image must use "false colors". If we were to print the final image in infrared, for example, it would be useless (we can't see infrared!). So, false colors are used to represent the invisible colors.
One common arrangement is to use:
red for far infrared
green for middle infrared
blue for near infrared
5. Editing. When an image is published (on the Web, in print or whatever), the process itself will change the image (for example, the settings of your home printer might be slightly different than the settings on mine).
To make the image pleasant to most applications, we do play with the saturation, hue, contrast, etc.
Scientists who need the images for research do have access to the original (uncorrected) images and to the corresponding correction images (bias, dark, flat).
Here is a "raw" (uncorrected) Hubble image of the Ring Nebula, using a specific filter
http://archive.stsci.edu/cgi-bin/mastpre...
Compare with a processed Hubble image:
http://www.spacetelescope.org/images/htm...
(The are not oriented the same way)
The scientist will use raw images. The popular magazines prefer processed images
For one thing, all CCD cameras can only take B & W shots. In order to build a color photo, we must take images through filters.
The simple way is to use the primary colors:
Take one image through a red filter; the camera simply registers the level of light that makes it through the red filter. As far as the camera itself is concerned, it is still taking a black & white picture.
Take another image of the same object, this time with a green filter.
Take a third image with a blue filter.
In many cases, we also take an image without any filters (called "luminosity" shot) in order to make the final image better.
Then, in a computer, we mix the images: the computer puts the appropriate quantity of red color for all pixels where the "red" picture shows white.
Green color where the "green" image shows white and blue for blue. It uses the "luminosity" shot to balance the mix of colors. The final addition of the three (or four) images gives a color photo.
---
In addition:
CCD cameras are very useful for scientific work because they are consistent in the way they record light. If star A is twice as bright as star B, the camera will record exactly twice as much light from A as it does from B.
However, as a "camera", it is not perfect.
1. When you turn the camera on before each image, the pixel values are supposed to reset to zero. They don't. Each pixel begins with a different value. We take an "image" (with zero second exposure) of these residual values and subtract them from the astronomical image (to correct). This image is called the "bias".
2. Some pixels may be affected by processes inside the camera. Most of the time, heat from close-by wires will cause an excess of electrons to be released in some part of the camera. So we take a "dark" image -- same exposure as the astro-image, but with the shutter closed. Thus, the pixels will be subjected to the heat from inside the camera, without receiving any light from the outside. This "image" must also be subtracted from the astro-images.
3. Some parts of the camera could receive more light (or less light) than other parts, due to the design and construction of lenses and mirrors. There is also a problem with dust on the lenses and mirrors. Even though they are greatly out of focus, they may affect the quantity of light that is blocked before reaching the camera. For a earth telescope, we take a "flat" field by taking a picture of the sky while it is still lit by twilight. For Hubble, flats were done in a lab before launch. It is possible to do the occasional flat by taking very long exposures while "scanning" the sky -- allowing the image to roam over the camera, so as to be completely blurred out -- but that costs a lot of time.
4. Putting a telescope in space is costly. Once there, though, it is possible to take pictures in wavelengths that our eyes cannot see (for example ultra-violet and infrared). In such cases, of course, the final image must use "false colors". If we were to print the final image in infrared, for example, it would be useless (we can't see infrared!). So, false colors are used to represent the invisible colors.
One common arrangement is to use:
red for far infrared
green for middle infrared
blue for near infrared
5. Editing. When an image is published (on the Web, in print or whatever), the process itself will change the image (for example, the settings of your home printer might be slightly different than the settings on mine).
To make the image pleasant to most applications, we do play with the saturation, hue, contrast, etc.
Scientists who need the images for research do have access to the original (uncorrected) images and to the corresponding correction images (bias, dark, flat).
Here is a "raw" (uncorrected) Hubble image of the Ring Nebula, using a specific filter
http://archive.stsci.edu/cgi-bin/mastpre...
Compare with a processed Hubble image:
http://www.spacetelescope.org/images/htm...
(The are not oriented the same way)
The scientist will use raw images. The popular magazines prefer processed images
2
0
Comment
Asker's rating 
Report Abuse