IPB

Welcome Guest ( Log In | Register )

7 Pages V  < 1 2 3 4 > »   
Reply to this topicStart new topic
Reprocessing Historical Images - II, Restoring images from antiquated and/or poor quality sources
tedstryk
post Sep 2 2008, 09:51 AM
Post #16


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4248
Joined: 17-February 04
From: Powell, TN
Member No.: 33



Thanks. I eventually hope to do more of the images, both before and after closest approach.


--------------------
Go to the top of the page
 
+Quote Post
Stefan
post Sep 2 2008, 09:53 AM
Post #17


Junior Member
**

Group: Members
Posts: 45
Joined: 16-November 06
Member No.: 1364



QUOTE (tedstryk @ Sep 1 2008, 11:28 PM) *
If you notice, there aren't interpretable features on the left side except at the very top. There are tantalizing hints, but much less can be seen.

Back to the Proteus image... I did some "poor man's dark correction" by subtracting the average of images C1241506 and C1241758 from C1138920 and removed the reseaus. I got the following result:

Attached Image


At left the raw image, in the center the raw minus dark, at right a median filtered version of the center image. All images are stretched (the center and right images by the same amount). If you smooth the right image it would look very much like the official version from the Planetary Photojournal. Ted, I do not see any of the subtle features visible on the left side of your Proteus, and I am now convinced your method introduces artifacts (the "brush strokes") on single, underexposed images like this one. I still prefer the official version of Proteus, although on that one details on the left side are not real either (amplified noise).

One more thing. The images shows very clearly an illuminated feature on the right (dark) side of Proteus:

Attached Image


This must be part of the moon (i.e. the far rim of the big crater), but does not show up in any version I have seen so far! One must be careful when forcing "empty" space to be black.

Stefan.
Go to the top of the page
 
+Quote Post
tedstryk
post Sep 2 2008, 11:11 AM
Post #18


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4248
Joined: 17-February 04
From: Powell, TN
Member No.: 33



It is there, and I have brightened it in this version. It may have been lost in the 16-to-8 bit conversion. I have stretched it in this version. Stefan, you have every right to your opinion, but beyond some faint light and dark features, I am not sure what features are on the left part of my version of the image. All I see is a small, faint possible crater in the center and a long dark albedo feature stretching across the disk from top to bottom (has been interpreted as groves, but that is a guess). Given how well the features matched up in the common areas between the images, I decided not to suppress other little mottlings in the limb area any more than I would the overlapping areas. Stefan, what is frustrating is that you are not responding at all to my explanation of how I did it.

Attached Image


Here is a slightly earlier version, before I attempted to reconstruct the gamma curve. Stefan, the fact of the matter is that your averaged version kills the sharp, high contrast details. I have preserved them, while binning data to pull a bit more out of the images. You don't actually know that the how much of the left-hand detail is amplified noise and how much of it is detail right on the threshold of detection - that is why I chose not to wipe it out. It is also before reconstructed the bottom tip.

Attached Image


One more thing...Stefan, I have explained the effect you describe as brush stroking.


--------------------
Go to the top of the page
 
+Quote Post
djellison
post Sep 2 2008, 11:27 AM
Post #19


Administrator
****

Group: Chairman
Posts: 13852
Joined: 8-February 04
Member No.: 1



I would agree - Stefan, you're not reading what Ted has posted in response to your initial scepticism and doubts. To be honest, my patience would have evaporated long ago.
Go to the top of the page
 
+Quote Post
tedstryk
post Sep 2 2008, 12:02 PM
Post #20


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4248
Joined: 17-February 04
From: Powell, TN
Member No.: 33



This is another image that I find interesting. I have always wondered why some areas, even under low solar illumination, look very smooth on a fine scale, while some areas do not. I had always faulted the image data, but given the smoothness of portions of some of the small Saturnian moons, I have come to question that.

Attached Image


--------------------
Go to the top of the page
 
+Quote Post
Stefan
post Sep 2 2008, 02:28 PM
Post #21


Junior Member
**

Group: Members
Posts: 45
Joined: 16-November 06
Member No.: 1364



This discussion is obviously going in the wrong direction, so this will be my last contribution.

QUOTE (djellison @ Sep 2 2008, 01:27 PM) *
I would agree - Stefan, you're not reading what Ted has posted in response to your initial scepticism and doubts. To be honest, my patience would have evaporated long ago.

Doug, I do not understand why you interfere. In science, scepticism and doubt are accepted, and often lead to more understanding, sometimes on both sides. I was interested in discussing one aspect of Ted's work, and I believe I did this in a respectful manner.

QUOTE (tedstryk @ Sep 2 2008, 01:11 PM) *
It is there, and I have brightened it in this version.

You are right, I see it now. It is almost invisible on my TFT monitor, and perhaps even my eye rejected it automatically as a dust speck.

QUOTE (tedstryk @ Sep 2 2008, 01:11 PM) *
what is frustrating is that you are not responding at all to my explanation of how I did it.

That is the point: I do not know how you got from the raw image to the end product, and I do not get the story from your posts. Binning? Smoothing? How do you treat empty space? You say that my averaged version (I presume you refer to the median filtered one) kills contrast details. True, but it also kills some noise. Binning is a form of averaging, so you kill contrast details too. And you preserve some noise. I believe these are all worthy topics for discussion, and I regret frustration has crept in.

I was interested in discussing your approach. How can I agree that your Proteus image is an improvement over the original version if I do not understand what you did? These images are so important, they will be the only ones for a very long time.

Stefan.
Go to the top of the page
 
+Quote Post
t_oner
post Sep 2 2008, 02:48 PM
Post #22


Junior Member
**

Group: Members
Posts: 97
Joined: 26-September 05
Member No.: 508



I must agree with Stefan's skepticism. Ted's images have been bothering me for sometime. I do not want to offend him as I admire his work and enthusiasm, but Ted, as I have told you in our private conversation it is sometimes better to leave the noise removal to the human brain. Your method removes real information from well exposed images and adds non-existant information to noisy ones and it could be misleading for people outside this forum. Your images are already great without noise removal.
Go to the top of the page
 
+Quote Post
tedstryk
post Sep 2 2008, 05:11 PM
Post #23


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4248
Joined: 17-February 04
From: Powell, TN
Member No.: 33



Stefan, I discussed this earlier in the thread, but I will add that I convert the images to 16-bit before binning to reduce damage.

Tayfun, the Mariner 6 and 7 images are a very different breed in there are some areas of the images that are totally saturated in analog and are therefore only available in the undersampled digital format, which makes part of the images better than others. I think that is what you are getting at there, but if not, I would appreciate your showing me (I am not saying that to be sarcastic - I literally mean it, I respect your work and knowledge, and I would like your input). However, on some images, there is the fact dealing with desmeared/deconvoluted and underexposed images can lead to a sort of terracing effect is something I have worked to find ways to get rid of that did not wipe out the high contrast details as well. As a result, low contrast features are less distinct than high contrast features. I never thought of someone thinking this is real.


--------------------
Go to the top of the page
 
+Quote Post
t_oner
post Sep 2 2008, 06:10 PM
Post #24


Junior Member
**

Group: Members
Posts: 97
Joined: 26-September 05
Member No.: 508



Ted, let's take your Proteus image as an example, due to its low DN values and high noise ratio there is a mathematical limit to the information that this image can hold. When I look at the NASA version my brain realizes that this is a noisy picture and interprets it accordingly. When I look at your version which looks perfect and as if taken under ideal conditions, my brain interprets all the details as real and that is the problem. Not only the real craters, but also any chance alignment of light and dark pixels looks like real features.

Also some of the real information that our brains would have no difficulty seeing it from the noisy background is also lost in the softening process. Try removing noise from Cassini SAR images, you will not be able to see many of the fine details.

This is a little bit like restoring historical artefacts to the limit that will make them look newly made. You will both lose information and add false information at the same time. Maybe you can make the noise removal only at the last stage of your processing workflow and display two versions.
Go to the top of the page
 
+Quote Post
tedstryk
post Sep 2 2008, 06:21 PM
Post #25


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4248
Joined: 17-February 04
From: Powell, TN
Member No.: 33



Are you referring to the smallscale 'features" on the left side? Perhaps because I dealt with them all the way through, I never really thought of them as looking like real features. Perhaps this is why when I was debating this with Stefan, he kept mentioning the features on the lower left side of the image and my reply was always to the effect of "what features on the left side of the image?"


--------------------
Go to the top of the page
 
+Quote Post
t_oner
post Sep 2 2008, 06:47 PM
Post #26


Junior Member
**

Group: Members
Posts: 97
Joined: 26-September 05
Member No.: 508



Yes I am, I have no problem with the features that you have identified.
Go to the top of the page
 
+Quote Post
tedstryk
post Sep 2 2008, 07:58 PM
Post #27


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4248
Joined: 17-February 04
From: Powell, TN
Member No.: 33



This is an alternate take in which I tried to wipe that out, but it never looked quite right.
Attached Image


Edit: Here is a slightly improved version.

Attached Image


--------------------
Go to the top of the page
 
+Quote Post
tedstryk
post Sep 2 2008, 09:53 PM
Post #28


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4248
Joined: 17-February 04
From: Powell, TN
Member No.: 33



Here is my Borrelly image at its original size. The other version was made to enlarge well for a size comparison Emily was doing. Both original frames for this image are smeared, which also creates some effects. This is the image at 133% of its original size. The left hand side is raw, the right hand side is my version before I worked to make it bigger for the comparison image. The right hand image has a partial overlay from the next closest image to reduce noise.

Attached Image


--------------------
Go to the top of the page
 
+Quote Post
4th rock from th...
post Sep 2 2008, 10:08 PM
Post #29


Member
***

Group: Members
Posts: 346
Joined: 21-April 05
From: Portugal
Member No.: 347



Hi,

Just my small contribution to this topic. Here's a quick processing of the Proteus image. I was completelly conservative on appliying gausian blur filters to reduce noise and levels adjustment. There's some cliping of the darkest parts of the image in an atempt to reduce uncertain details at high noise levels.

The general details visible on Ted's images are present, and a hint of the finer ones.

I think that the issue here is the balance between further processing and the introduction of processing artifacts. It's a complicated decision and is related to image scale and sampling.

Attached Image


--------------------
_______________________
www.astrosurf.com/nunes
Go to the top of the page
 
+Quote Post
elakdawalla
post Sep 2 2008, 10:29 PM
Post #30


Bloggette par Excellence
****

Group: Admin
Posts: 4503
Joined: 4-August 05
From: Pasadena, CA, USA, Earth
Member No.: 454



I think it's fascinating to see a comparison of the results of different people's efforts to process a single image, especially one of such marginal quality but such great significance as this one shot of Proteus (honestly, when does anybody think we'll get another chance to look at Proteus?) You each have a slightly different approach and a range of goals, as well as a range of willingnesses to accept the possible introduction of artifacts in order to generate a pleasing product. Yet all the approaches appear to be systematic (in that they should be reproducible -- no actual subjective painting, for instance). Anybody else want to try? smile.gif

--Emily


--------------------
My blog - @elakdawalla on Twitter - Please support unmannedspaceflight.com by donating here.
Go to the top of the page
 
+Quote Post

7 Pages V  < 1 2 3 4 > » 
Reply to this topicStart new topic

 



RSS Lo-Fi Version Time is now: 27th November 2014 - 02:51 AM
RULES AND GUIDELINES
Please read the Forum Rules and Guidelines before posting.

IMAGE COPYRIGHT
Images posted on UnmannedSpaceflight.com may be copyrighted. Do not reproduce without permission. Read here for further information on space images and copyright.

OPINIONS AND MODERATION
Opinions expressed on UnmannedSpaceflight.com are those of the individual posters and do not necessarily reflect the opinions of UnmannedSpaceflight.com or The Planetary Society. The all-volunteer UnmannedSpaceflight.com moderation team is wholly independent of The Planetary Society. The Planetary Society has no influence over decisions made by the UnmannedSpaceflight.com moderators.
SUPPORT THE FORUM
Unmannedspaceflight.com is a project of the Planetary Society and is funded by donations from visitors and members. Help keep this forum up and running by contributing here.