[tig] RE: This is very grey to me...

Slanzi Christian cslanzi
Fri Apr 23 14:20:36 BST 2004


> So, my question is: why?  Why do both boxes appear
> different when they are in fact the same?  What is making
> my brain perceive the boxes as different and better yet:
> how do I get my brain to stop doing that!?!??  
> It is not the prescence of the letters and if I crop
> each box, and paste both into a new image, they now look
> the same, thus proving that the letters are not causing the
> boxes to appear different in the original image. Nor is it the
> prescence of the cylinder removing it, or altering it's color
> does not change anything.  Therefore, I am figuring it is not
> the boxes themselves, but the SURROUNDING boxes but
> what gives?  Why would the shade of any of the surrounding
> boxes affect how others are perceived?

That illusion is due to our human visual system, that it is not only a
simple physical meter system but it is a perceptive system.

The checker shadow illusion is a mix of several behaviours of human
vision. 

First at all, the local contrast. A check surrounded by darker checks
will appear lighter than what really it is. And the darker checks are
surrounded by lighter checks, so they will enhance their darkness and so
lighting more  the central check (B).

The cast shadow is a trick, because human vision system tends to
compensate and ignore them to determine the real color of a surface. The
shadow is not a real shadow btw you see the object that seems to cast
that shadow and the shadow has soft edges, so you put that check in that
context and so your human vision system remove that shadow letting you
see a lighter colour.

But also because you are looking a checker, usually formed by
alternatively light grey and dark grey checks. That tends you to
recognize the edges of checks.        

> And from that, comes two questions:
> 
> 1.  If the board were oriented differently, would this affect how
> the brain perceives the boxes?
> 

No, it doesn't affect your perception.


> 2.  How about looking at the same box in real life as opposed
> to on a computer screen would that make any difference?
 
Oh... this is cool :-) You will have a more marked illusion... having a
natural illuminant with the reflection of the light (that is the thing
that bring the information to your eyes), while the light of the monitor
is emitted.  

> Anyone who can properly explain this will be owed a night of
> heavy drinking.

That sounds good... Would you drop in Milan? ;-)

Slanzi Christian
GIC - Graphics, Imaging and Colour research group
Wiener - computer graphics lab
Polo Didattico e di Ricerca di Crema
University of Milan 
 

> 
> Original Message: 
> Date: Thu, 22 Apr 2004 12:58:06 -0400
> From: "jeffh" <jeffh at nvbb.net>
> Reply-To: <jeffh at nvbb.net>
> To: "tig"  <tig at tig.colorist.org>
> Subject: [tig] This is very grey to me...
> 
> 
> http://www.ebaumsworld.com/checkershadow.html
> 
> Some of you may have seen this before, or other things
> like it; I just found this yesterday.  Look at box A, and
> then box B.  They look like they are different shades=85.
> but in fact they are not - they are identical.  If you believe
> me, proceed to paragraph 2.  If you do not believe me,
> save the pic, and open it up in a paint program.  Select
> and move a part of A into B, or vice-versa.  See?
> Alternately, hold your color dropper tool over box A and
> see what color is displayed (Jasc=92s Paint Sho Pro has
> this; I am sure other programs do as well, tho I am not
> sure if the =93dropper=94 would be called something else).
> See what RGB value is displayed; I get 107 107 107.
> Then hold the tool over box B (try to keep away from the
> edges) =96 the same 107 107 107, right?
> 
> So, my question is: why?  Why do both boxes appear
> different when they are in fact the same?  What is making
> my brain perceive the boxes as different =96 and better yet:
> how do I get my brain to stop doing that!?!??  <cough>
> It is not the prescence of the letters =93A=94 and =93B=94 =96 if I
crop
> each box, and paste both into a new image, they now look
> the same, thus proving that the letters are not causing the
> boxes to appear different in the original image. Nor is it the
> prescence of the cylinder =96 removing it, or altering it=92s color
> does not change anything.  Therefore, I am figuring it is not
> the boxes themselves, but the SURROUNDING boxes =96 but
> what gives?  Why would the shade of any of the surrounding
> boxes affect how others are perceived?
> 
> And from that, comes two questions:
> 
> 1.  If the board were oriented differently, would this affect how
> the brain perceives the boxes?
> 
> 2.  How about looking at the same box in real life as opposed
> to on a computer screen =96 would that make any difference?
> 
> Anyone who can properly explain this will be owed a night of
> heavy drinking=85.
> 
> 







More information about the Tig mailing list