| |
 |
|
|
Science Forum Index » Image Processing Forum » binary image rotation...
Page 1 of 1
|
| Author |
Message |
| Elmar Sack... |
 Posted: Sun Jun 29, 2008 11:02 am |
|
|
|
Guest
|
Hi all,
a known effect for BIR is that the general rotation formula produces holes
on closed blocks. I found different algorithms in wikis (e.g. the
shear-algorithm (Paeth)) and the paper "Hierarchical Block Matching Method
for Fast Rotation of Binary Images" (Chien,Baek) which led me to "Parallel
image transformation and its VLSI implementation” (Cheng,Tang,Suen). I
implemented the last two algorithms and they work as expected.
They all use a real to integer mapping using floor(x+0.5) (C syntax).
Is there a method not only rotating just the integer coordinate (x,y) but
also its associated cell built by floor((x + 0.5, y + 0.5))?
I investigated this "cell" rotation (what I call incidence rotation) and
developed a mapping which improves the results of the two papers above and
want to get a feeling wether I am on the right way. My results also shows
visually rather simple, how holes are created by rotation.
Thanks in advance!
Elmar |
|
|
| Back to top |
|
| |
|
Page 1 of 1
All times are GMT - 5 Hours
The time now is Thu Oct 16, 2008 3:51 pm
|
|