[NatureNS] Desaturated 180 Degree Double Rainbow Photo

Date: Tue, 14 Jun 2011 15:34:12 -0300
From: "Stephen R. Shaw" <srshaw@Dal.Ca>
To: naturens@chebucto.ns.ca
References: <31916D37B7504C1DB130342973076FD7@hans57ff9baf63>
User-Agent: Internet Messaging Program (IMP) H3 (4.3.4)
Precedence: bulk
Return-Path: <naturens-mml-owner@chebucto.ns.ca>
Original-Recipient: rfc822;"| (cd /csuite/info/Environment/FNSN/MList; /csuite/lib/arch2html)"

next message in archive
next message in thread
previous message in archive
previous message in thread
Index of Subjects

Index of Subjects
Hans: Agreed -- I actually looked at your colour picture last night  
and thought the same thing (great picture as always by the way, as  
others have said).

Another potential 'enemy', though, is non-linear processing.  I don't  
know if consumer digital cameras use a non-linear algorithm to  
compress a scene's extended brightness scale from the CCD chip into  
the 256 steps usually available in the conversion from light energy to  
pixel memory record.  I became concerned about this when taking pics  
of fluorescence down a microscope, and had to go to some length to  
check that the fancy camera/software there was giving a linear output  
up to CCD saturation, so we could make proper relative brightness  
estimates from the images.  Don't know the answer for consumer  
cameras.  In addition, some of Photoshop's filters must act  
non-linearly to distort the initially recorded image, for instance the  
wonderful Unsharp Mask: wondered if you may have used it on the colour  
picture of the rainbow (the telephone cables have unexpectedly bright  
white adjacent dots).

Another perhaps curious feature is that the few clouds seem quite  
bright when present just below and extending into your primary  
rainbow, but wash out where they look to extend beyond into  
Alexander's (putatively dark) band.  But that could just be in the  
cloud distribution itself, not in the processing.

It would be interesting to know if you or anyone on the list knows  
whether consumer cameras (or some of them) operate linearly, that is  
(roughly), give out twice the number (0-256) to the memory pixel  
record where the scene is twice as bright objectively (not visually).   
Or do some/all use intensity compression to render scenes more  
acceptable or manageable?
Steve
  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Quoting Hans Toom <htoom@hfx.eastlink.ca>:
> When analyzing photographs colour can sometimes be our enemy so I've  
> desaturated the first rainbow photo to black and white.  The visible  
> brightness of the image is clearly brightest below or within the  
> primary rainbow and darkest between rainbows and the sky above the  
> outer rainbow.  The visible brightness in the space between rainbows  
> and the background sky above the rainbows appears to my naked eye as  
> identical.  No dark band is visible to my eyes anywhere.
>
> The overall effect of the desaturated image is much like a giant  
> eyeball with the space from the inner rainbow skyward becoming the  
> imaginary eyelid.  The visible brightness below the lower rainbow is  
> stunning and much brighter than the background sky as seen above the  
> outer rainbow.
>
> http://www.hanstoom.com/StockPhotos/Visions3/0179.html
>
> Hans Toom
> Portuguese Cove, Nova Scotia, Canada
> http://www.hanstoom.com/
>

next message in archive
next message in thread
previous message in archive
previous message in thread
Index of Subjects