Comments for Steve on Image Processing

Comment on Escher, images, and chess by ismail

ismail — Sat, 06 Sep 2008 01:59:12 +0000

i love chess keep posting :)
can we make a web cam identify a chess set ? so we have a roboarm plays for real with us ? it will look fun .. it’s like the computer having grew arm and playing with you ..

Comment on Escher, images, and chess by Navan

Navan — Fri, 05 Sep 2008 17:54:39 +0000

While black is going to win with a smothered mate, it is hard to see what moves would have led to this position. For black, the queen must have started from one of the squares in the diagonal c5-f2. White must have made a blunder to allow this to happen. If it is an artificially set up position why is there a rook at a3. Probably just to connect with the castle on the left.

Comment on Escher, images, and chess by Doug

Doug — Fri, 05 Sep 2008 17:29:47 +0000

Forced ’smothermate’ is about to happen, with the added insult of threatening the queen on the final move too.

Comment on Escher, images, and chess by Viton

Viton — Fri, 05 Sep 2008 16:50:48 +0000

Let’s give it a try:
- RxQ : White Rook takes black Queen (White King was checked, can’t take Queen, due to Bishop)
- Nf2 checkmate : Black Knight at f2

Comment on Spatial transformations: maketform, tformfwd, and tforminv by Omar

Omar — Fri, 05 Sep 2008 15:42:23 +0000

Hi Steve,
when using tformfwd to find corresponding points in the new space, the resulting co-ordinates from tformfwd do not match the transformed image display (using for instance datacursor to select a couple of points). As an example, I’m working on the following piece of code:

tTransf = eye(3);
NormMat = [0.6851 -0.3149; -0.3149 0.6851];
tTransf(1:2,1:2)= NormMat;
tform=maketform(’affine’,tTransf);
new_im = imtransform(im,tform);

The problems occur when I call:
new_pt = tformfwd(tform, pt);

The co-ordinates given by new_pt do not match the diplayed point!!!I should add that in my work, I have not control over NormMat, it is computed iteratively. But I have noted that tformfwd work fine when the values in NormMat are all positive.
Any suggestion please ?

Comment on Dilation algorithms—Introduction by Steve

Steve — Thu, 04 Sep 2008 17:40:32 +0000

Cris—You’re right, I should have caught the plot scaling issue. I wasn’t actually trying hide the point. I updated the post to make it more clear. Thanks. Also, thanks for the TrIP pointer; I’ll take a look.

Comment on Dilation algorithms—Introduction by Cris Luengo

Cris Luengo — Thu, 04 Sep 2008 15:04:15 +0000

Not to spoil your upcoming bog entry too much, but if you scale the first graph (times vs Q) by setting the lower ylim to 0, you’ll see it is pretty close to a horizontal line. There’s only a 10% difference from low to high.

However, isn’t this supposed to be an O(1) algorithm? Where’s the increase coming from? Actually, only yesterday I saw some paper with a timing plot, where operations with a linear SE actually had lower times for longer SE. That one seriously surprises to me! (Urbach and Wilkinson, IEEE TIP 17(1):1-8, fig 9)

Comment on Dilation algorithms—Introduction by Steve

Steve — Thu, 04 Sep 2008 01:58:37 +0000

Jim—Thanks for adding your comment showing how the syntax works for matrices.

Comment on Dilation algorithms—Introduction by Jim

Jim — Wed, 03 Sep 2008 21:33:30 +0000

for i = A
…statements…
end;

Description:
The …statements… are executed (as MATLAB commands) for i equal sequentially to each column in A. If A is 3D or greater, the …statements… are executed for i equal sequentially to the vector of first index values for each combination of the other indices of A.

Example:
A = [1 2 3; 4 5 6];
for i = A
this_is_i = i
end;

Result:
this_is_i =

1.00
4.00

this_is_i =

2.00
5.00

this_is_i =

3.00
6.00

Comment on Dilation algorithms—Introduction by Steve

Steve — Wed, 03 Sep 2008 20:51:14 +0000

Omar—Nice work.