QUOTE (KS15 @ Jun 25 2008, 08:43 AM)

No, the images on these pages are not valid science data. They're press release images. No serious scientist would dream of using these images for any real analysis because there's no record of how the data was processed to get to these pretty pictures. They're beautiful images to look at, but they're not meant for research, especially if you're trying to interpret tiny details at or below the resolution limits of the data. The real science data is carefully processed step by step, with complete records of the exact work that was done. There's a good overview of the image processing done both for the science data and for the press release images in this conference paper by the HRSC team:
PUBLIC OUTREACH AND ARCHIVING OF DATA FROM THE HIGH RESOLUTION STEREO CAMERA ONBOARD MARS EXPRESS: 2004 – THE FIRST YEARPay special attention to this paragraph:
QUOTE (LPI 2202)
Press Release Images.
HRSC scenes selected for publication by the PI and produced by the PI group in cooperation with DLR will contain several images: Highest resolution is provided with a black and white nadir image; the identical scene is also displayed in color and by an anaglyph image – the latter based on a digital elevation model calculated from the stereo channels; this applies also for one or more perspective views; all images make use of the high resolution of HRSC data, and therefore have considerable dimensions of several thousand pixels per line and column to allow large-format printing; the resolution, though, is slightly reduced for convenience, to allow better online handling. Interpretational text, context images and geographical location of the scene on a Mars globe are also provided. The release sets go online simultaneously at ESA and DLR.
The HRSC scientists state flat-out that the highest resolution is provided by the black and white nadir image. That's what I've been trying to tell you all along, and that's why I'm posting black and white images. They also state that they're reducing the resolution of the press release images for better online handling. Were you aware of this? The resolution figures you keep posting are for the full-scale original images, not the reduced-size press-release images. As I said, you can't use press-release images for serious science because you just don't know how they've been manipulated.
QUOTE (KS15 @ Jun 25 2008, 08:43 AM)

JPEG images are valid images. In fact, The Mars Express camera transmits images to Earth in JPEG form.
Yes, Mars Express typically uses JPEG-type compression to save bandwidth. It's a tradeoff between sending back just a few uncompressed images or many more compressed images with slightly lower quality. The science teams have carefully weighed the alternatives and have selected a compression level that gives the best balance between quality and quantity.
Once the images are back on Earth, though, no serious scientist would use press release images when the original science data is available. Heck, you haven't even been using the best available press release images! The press release JPEGs have been highly compressed to make the file sizes small for easier web downloading. They have much, much lower quality than the original images transimitted from Mars Express. On the sites you've been using, though, ESA has also provided a much better TIFF version of each of the press release images. These files are very large, but they generally have much better resolution and much lower compression than the JPEG versions.
You've been using this:
385-090408-5212-6-co1-01-NepenthesMensae_H1.jpg (cropped)When you just as easily could have been using this:
385-090408-5212-6-co1-01-NepenthesMensae_H2.tiff (cropped)Is there a reason you don't want to use the best available images?
QUOTE (KS15 @ Jun 25 2008, 08:43 AM)

The Mars Express images are COLOR images. There is absolutely no reason to convert these images to black and white (Grayscale).
There is no reason to compare B&W images to color….There is no comparison.
No, the HRSC camera does NOT record color images. If you would read your own
link, you'd know that the HRSC camera has nine image sensors, each of which is a line of pixels that record grayscale data. The entire camera works like a giant flatbed scanner or photocopy machine. As the spacecraft orbits over Mars, the camera stares downward and the sensor lines sweep over the surface like the scan head on a scanner. To get color information, three of the nine sensor lines are fitted with color filters (red, blue, and green). These sensors still only record grayscale information, but the three channels can be assembled into a color image back on Earth.
The nadir channel is the primary image channel, so its data is normally saved at the full 12.5 meters per pixel resolution. To save memory, though, the color channels are typically downsampled and saved at either 50 meters per pixel or 100 meters per pixel resolution.
Reference:
The high-resolution stereo camera (HRSC) experiment on Mars Express: Instrument aspects and experiment conduct from interplanetary cruise through the nominal mission
QUOTE
In order to decrease the data volume to accomplish spacecraft downlink requirements, the pixels of a CCD line can be summed up on-chip in an analog way in flight direction by increasing the exposure time, and in a crosstrack direction by multiple pixel binning 2, 4 or 8 times digitally. A typical mode for high-resolution imaging with emphasis on surface mapping is operating the nadir sensor without pixel summation, the 4 multi-spectral sensors with 4x4 or 8x8 pixel summation, the outer stereo sensors with 2x2 pixel summation and the inner stereo sensors (the photometry sensors) with 4x4 pixel summation.
The color images of Cydonia you've been posting over and over are actually created from four individual black and white images taken on two different orbits. You can see this by looking at the FU Berlin site for the Cydonia press-release images here: (
link). Look at the image download table near the bottom of the page and you will see:
#co1 rgb color scene, large, (data: HRSC orbit 3253 (nadir), HRSC orbit 1216 (rgb))
The Cydonia color images are made up from nadir channel data from orbit 3253 (at 12.5 meter per pixel resolution) and from the three channel red-green-blue data from orbit 1216 (at 50 meters per pixel resolution). I've used the original Level 3 science data files for each of these four images to generate my own version of the color Cydonia image. Here's how it all goes together:

The three images at the top of the picture are the red, green, and blue data from orbit 1216. Since this data was saved at only 50 meters per pixel, it doesn't show much detail at all. I've used Photoshop to combine these three grayscale images into a single color image, shown in the second row. You can see that the colors are good, but that there is very, very little image detail visible. That's where the final step comes in. The nadir image from orbit 3253 is a very good, 12.5 meter per pixel resolution grayscale image of the "face." By using the low-res colors from the orbit 1216 data to tint the orbit 3253 nadir image, I end up with a final color image of the "face." This is essentially the same way that ESA made their color press-release image. Frankly, I'm not that impressed with their results. I think their final images have far too much contrast and far too much sharpening.
QUOTE (KS15 @ Jun 25 2008, 08:43 AM)

Some examples of an attempt to obscure images….
Good image:

By the way, This is not a crater.
Poor image:

This image is an attempt to remove color and detail.
Nope. It's my attempt to show you the image with the best detail. From the
DLR Nepenthes Mensae web page (near the bottom):
The colour scenes have been derived from the three HRSC-colour channels and the nadir channel. The perspective views have been calculated from the digital terrain model derived from the stereo channels. The anaglyph (3D) image was calculated from the nadir and one stereo channel. The black and white high-resolution images were derived from the nadir channel which provides the highest detail of all channels.
As I've described above, the color images are typically created by tinting the black and white nadir image. All of the image detail comes from the nadir image. Still, if you want to see color, try this:
385-090408-5212-6-co1-01-NepenthesMensae_H2.tiff (cropped, 1.5x enlargement)This IS a crater. The feature's shadows and the 3D anaglyph image show that it's a shallow depression in the surface. It looks just like several other craters in the area. What evidence do you have that it's NOT a crater?
QUOTE (KS15 @ Jun 25 2008, 08:43 AM)

It seems certain few individual(s) are determined to make this topic more complicated than it really is.
I will keep it simple.
But this stuff ISN'T simple! It really
is rocket science, and there are a lot of very, very smart people working very hard to get these amazing images back to Earth.