A Space & astronomy forum. SpaceBanter.com

Go Back   Home » SpaceBanter.com forum » Astronomy and Astrophysics » CCD Imaging
Site Map Home Authors List Search Today's Posts Mark Forums Read Web Partners

Some Questions about Hough Transforms...



 
 
Thread Tools Display Modes
  #1  
Old April 17th 08, 12:26 AM posted to comp.graphics.algorithms,sci.image.processing,sci.astro.ccd-imaging
SpreadTooThin
external usenet poster
 
Posts: 6
Default Some Questions about Hough Transforms...

How can I determine the length of the line that the hough transform
detects?

What goes on with aliased lines...
for example..

X
X
X
X
X
X
X


Is this three vertical lines or is it one line?

I seem to remember there being a modified hough transform that would
tell you were the line started and ended.. anyone else seen that
paper?
If so.. Got a link?

  #2  
Old April 17th 08, 02:02 AM posted to comp.graphics.algorithms,sci.image.processing,sci.astro.ccd-imaging
lomas
external usenet poster
 
Posts: 1
Default Some Questions about Hough Transforms...

On Apr 17, 7:26*am, SpreadTooThin wrote:
How can I determine the length of the line that the hough transform
detects?

What goes on with aliased lines...
for example..

X
X
* X
* X
* X
* * X
* * X

Is this three vertical lines or is it one line?

I seem to remember there being a modified hough transform that would
tell you were the line started and ended.. anyone else seen that
paper?
If so.. Got a link?


In my opinion,
whether they are one line or not is depend on the bin you use in hough
transform parameter space.
and after hough transfrom, the bin where peak lying holds the
length(in pixel) and line parameters (e.g. K and B in y = K*x+ B).
With these parameters, the ending of line is possible to be found by
some discussion.

Best Regards,
lomas
  #3  
Old April 18th 08, 12:12 PM posted to comp.graphics.algorithms,sci.image.processing,sci.astro.ccd-imaging
JS[_2_]
external usenet poster
 
Posts: 1
Default Some Questions about Hough Transforms...

Is this three vertical lines or is it one line?
That will depend on the resolution of your accumulator.

Determining the length of your line: if the input to the hough
transform is a binary image, then the length of any possible line
could be considered to be the value in the corresponding accumulator
bin. However, I would not recommend using this as the length. I
think that the HT is good for getting general positions of lines, but
if you want length/position of lines you need to then use other
methods as well.

  #4  
Old April 18th 08, 10:57 PM posted to comp.graphics.algorithms,sci.image.processing,sci.astro.ccd-imaging
SpreadTooThin
external usenet poster
 
Posts: 6
Default Some Questions about Hough Transforms...

On Apr 18, 5:12*am, JS wrote:
Is this three vertical lines or is it one line?


That will depend on the resolution of your accumulator.

And this is really the problem I'm facing.. How do I determine the
size of the accumulator?
I can find the maximum peak easily, but what about other peaks in the
image...
How can I go about finding those?

I'm wondering if I should find a peak and then use the neighbors as a
weighted sum to get
the actual R Theta... Make sence?



Determining the length of your line: if the input to the hough
transform is a binary image, then the length of any possible line
could be considered to be the value in the corresponding accumulator
bin. *However, I would not recommend using this as the length. *I
think that the HT is good for getting general positions of lines, but
if you want length/position of lines you need to then use other
methods as well.


  #5  
Old April 19th 08, 12:34 PM posted to comp.graphics.algorithms,sci.image.processing,sci.astro.ccd-imaging
Nicolas Bonneel
external usenet poster
 
Posts: 1
Default Some Questions about Hough Transforms...

SpreadTooThin a écrit :
How can I determine the length of the line that the hough transform
detects?

What goes on with aliased lines...
for example..

X
X
X
X
X
X
X


Is this three vertical lines or is it one line?

I seem to remember there being a modified hough transform that would
tell you were the line started and ended.. anyone else seen that
paper?
If so.. Got a link?


You can increase the dimension of your Hough transform to make it 4
dimensions instead of 2 (for example the 4 dimensions are the
x1,y1,x2,y2) and find your maximum in this 4 dimensional space.

You will directly get your x1,y1,x2,y2 and the corresponding length.

Otherwise, I don't really know how to get the length in the usual Hough
transform since it only encodes directions.


--
Nicolas Bonneel
http://www.bonneel.com
  #6  
Old May 6th 08, 02:18 PM posted to comp.graphics.algorithms,sci.image.processing,sci.astro.ccd-imaging
Norberto Goussies
external usenet poster
 
Posts: 1
Default Some Questions about Hough Transforms...

How can I determine the length of the line that the hough transform
detects?


MATLAB has the houghlines function to get the length. I have tried it
is very useful. You must provide two parameters : 'FillGap' and
'MinLength'
  #7  
Old May 7th 08, 04:13 AM posted to comp.graphics.algorithms,sci.image.processing,sci.astro.ccd-imaging
SpreadTooThin
external usenet poster
 
Posts: 6
Default Some Questions about Hough Transforms...

On May 6, 7:18*am, Norberto Goussies wrote:
How can I determine the length of the line that the hough transform
detects?


MATLAB has the houghlines function to get the length. I have tried it
is very useful. You must provide two parameters : 'FillGap' and
'MinLength'


Oddly enough thats exactly what I did in my application.
Cool.

 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
EINSTEIN 1905 DERIVATION OF LORENTZ TRANSFORMS Pentcho Valev Astronomy Misc 7 September 30th 07 05:48 PM
More Questions Macro Amateur Astronomy 3 September 13th 07 11:26 PM
Lorentz transforms physical incoherence [email protected] Astronomy Misc 223 June 24th 05 12:48 AM
ISS O2 questions? JJR2 Space Station 12 May 22nd 05 12:45 AM
A Few Questions: Bob Carlson Astronomy Misc 4 May 13th 04 04:37 PM


All times are GMT +1. The time now is 02:23 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 SpaceBanter.com.
The comments are property of their posters.