Quote:
Originally Posted by Bad. Guardian
What I meant by "normal 3D shading cues" was this: large-scale variations in brightness, which allow a normal human visual system to perform a kind of "shape from shading" process, and thus perceive the three-dimensional shape of the landform.
What I meant by "aggressively suppressed" is that these large-scale variations in brightness, which serve a very important purpose in helping a normal human viewer to perceive the 3D shape of the landform, were _selectively_ eliminated from the image. Smaller-scale variations, which help the viewer to perceive the shape of small lumps and boulders, but which do not help the viewer to perceive the overall 3D shape of the landform, were _not_ eliminated from the image.
|
The primary image is low in contrast, but this does not mean to say that they selectively removed. Given the hazy atmosphere and the high sun angle the image is naturally and inevitably lacking in contrast with small shadows. The effect of the processing has been the opposite to what you claim. A low contrast image has had that increased.
Note that using the expressions "aggresively suppressed" and selectively eliminated" implies deliberate distortion by the people do did the processing. This is a very serious accusation of intellectual dishonesty and you musst substantiate it. So faar you have not done this, only expressed your opinion.
Quote:
Originally Posted by Bad. Guardian
To help you understand this, I'm attaching a new animation to this message. This animation was prepared in the same way as "before-after.gif" above, except that instead of including the entire image reduced in size, this time I've extracted only a small square, and highlighted an important 3D feature.
This animation shows a small lump or boulder (informally known as "the teardrop"), and the approximately level surface (or "platform") on which it lies, and the slope below the platform, and the approximately level plain below the slope. The slope is highlighted.
If you look at the animation, you'll notice that in the original raw image the platform is brighter than the level plain, and much brighter than the slope. This phenomenon is of course caused by the position of the sun in the sky, and by the shape of the landform. This large-scale variation in brightness helps the normal human viewer to perceive the 3D shape of the platform, and the slope, and the plain.
|
When you look at the animation, you'll also notice that in the image that was used in the press release, the difference in brightness between the platform and slope and plain has been very aggressively suppressed. There's almost no difference. It's much more difficult for a normal human viewer to perceive the platform, and the slope, and the level plain.
Does this suppression result from a simple reduction in contrast?
No!
Look at the "teardrop." Like other small lumps and boulders in the image, the teardrop's 3D shading cues are actually _more_ prominently visible in the press release image than they were in the original raw image that was returned by the spacecraft.
In summary: the overall shape of the landform was suppressed, while the shape of individual lumps and boulders was enhanced.[/QUOTE]
Sorry, I don't find the animation useful. It's of too small an area, the extact context of each sub part is not clear, and it is not obvious which processed frames have beeen used.
What would be better a side by side comparison of the different images with labels highlighting the differences between them.
Quote:
Originally Posted by Bad. Guardian
I had an interesting result in forensic image analysis, and wanted to share it. Contemplation of the implications of that result led to an interesting object lesson in the Ethics of Bad Astronomy, so I decided to share it here.
|
Unfortunately your "forensic image analysis" is simply wrong. Your "reconstructed" a processing methology has no relation to what was actually used and your claim that contrast was suppressed is contrary to the simple fact that contrast enhanced. In faact, almost nothing can be seen in the raw image.
The cropped raw MOC image can be seen at
http://www.msss.com/mars_images/moc/...elease/raw.gif . Vastly more detail is visible in the image as processed by MIPL
http://mpfwww.jpl.nasa.gov/mgs/targe...ydonia1c-s.gif , as well as those by Parker and MSSS.
Quote:
Originally Posted by Bad. Guardian
Most of my followups in this thread have just been clarifications for people who are confused about the basic facts, e.g. which image was returned by the spacecraft, and which image was used in the "catbox" press release.
|
Not confused at all as to which basic image was being referred to, MOC2-41.
Quote:
Originally Posted by Bad. Guardian
Trying to make something out of nothing? Pshaw, I'm only doing what this site was designed for. We find clueless misrepresentations of real astronomical data, and we criticise them mercilessly. Right?
|
Let's see, so far you have been show links with the following facts:
Observing conditions were well below par.
The actual processing of the image used by MIPL, MSSS and Tim Parker
The original, image, washed out because of the poor observing conditions, was processed to enhance contrast enhanced, not supress it.
Since you don't appear to have taken any of these facts on board, I advise you to be careful who you call clueless, especially when passing judgement on people and groups of the calibre of the MIPL, Tim Parker, and MSSS.
Note that the different processed versions of the same image were all available on the same day. That's honest, showing different ways of processing the same data and letting people shose which one they prefer.
Jon