http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176
MATLAB Central Newsreader  can someone help me with this
Feed for thread: can someone help me with this
enus
©19942015 by MathWorks, Inc.
webmaster@mathworks.com
MATLAB Central Newsreader
http://blogs.law.harvard.edu/tech/rss
60
MathWorks
http://www.mathworks.com/images/membrane_icon.gif

Sun, 08 Mar 2009 23:42:02 +0000
can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#633375
Regina
hi!<br>
<br>
i really have a difficulty in analyzing the star skeleton of a given image.<br>
i need to get the angle and i don't seem to get how to do it.<br>
i hope someone will help me with this.<br>
<br>
<a href="http://www.vision.cs.chubu.ac.jp/04/pdf/VSAM08.pdf">http://www.vision.cs.chubu.ac.jp/04/pdf/VSAM08.pdf</a><br>
<br>
the pdf above explains how to get the angle and i dont get how.<br>
i have already the extremal points of the skeleton and my problem is to get the angle.<br>
if someone is willing to help me with this. <br>
i will post my codes here as well as my sample images<br>
<br>
thanks

Mon, 09 Mar 2009 00:56:57 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#633383
ImageAnalyst
Regina<br>
Did you read down as far as equation (7)?<br>
It says the angle is the arctangent of (x  xcentroid)/(y  ycentroid)<br>
What about that don't you understand?

Mon, 09 Mar 2009 01:18:02 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#633388
Roger Stafford
"Regina " <velasquezregina@rocketmail.com> wrote in message <gp1l4a$2ud$1@fred.mathworks.com>...<br>
> hi!<br>
> <br>
> i really have a difficulty in analyzing the star skeleton of a given image.<br>
> i need to get the angle and i don't seem to get how to do it.<br>
> i hope someone will help me with this.<br>
> <br>
> <a href="http://www.vision.cs.chubu.ac.jp/04/pdf/VSAM08.pdf">http://www.vision.cs.chubu.ac.jp/04/pdf/VSAM08.pdf</a><br>
> <br>
> the pdf above explains how to get the angle and i dont get how.<br>
> i have already the extremal points of the skeleton and my problem is to get the angle.<br>
> if someone is willing to help me with this. <br>
> i will post my codes here as well as my sample images<br>
> <br>
> thanks<br>
<br>
In the paper you mentioned they give equation (7):<br>
<br>
theta = arctan((lxxc)/(lyyc))<br>
<br>
where fig. 5a shows theta being measured counterclockwise from the downwardpointing positive yaxis toward the rightpointing positive xaxis. You could do this computation in matlab with:<br>
<br>
theta = atan((lxxc)/(lyyc));<br>
<br>
However I would recommend using<br>
<br>
theta = atan2(lxxc),lyyc);<br>
<br>
instead since its results remain valid even for "legs" that have been raised more than pi/2 (90 degrees) above horizontal. Also 'atan2' is more accurate for angles that are near pi/2 where 'atan' would have accuracy difficulties.<br>
<br>
Note that both functions give angles in radians, not degrees. To get degrees with 'atan2', multiply by 180/pi.<br>
<br>
The angle phi as shown in fig. 5b seems to be defined differently. If the x and y axes are defined the same, phi would be measured clockwise from the vertical negative yaxis toward the rightpointing positive xaxis, so you could find it with<br>
<br>
phi = atan2(lxxc,ycly);<br>
<br>
It seems a shame to introduce confusion by these two different definitions of angles. Perhaps that is the reason you had trouble with this, Regina.<br>
<br>
Roger Stafford

Tue, 10 Mar 2009 00:06:04 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#633629
Regina
hi mr stafford<br>
<br>
thanks for your help.<br>
i'll try our suggestion and i hope it works!<br>
<br>
thanks again<br>
<br>
regina

Wed, 11 Mar 2009 06:32:04 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#633970
Regina
hi mr stafford<br>
<br>
can you explain to me how to code equation 6 found in the pdf?<br>
i cant seem to understand it<br>
<br>
thanks

Wed, 11 Mar 2009 08:19:01 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#633982
Roger Stafford
"Regina " <velasquezregina@rocketmail.com> wrote in message <gp7lt4$34l$1@fred.mathworks.com>...<br>
> hi mr stafford<br>
> <br>
> can you explain to me how to code equation 6 found in the pdf?<br>
> i cant seem to understand it<br>
> <br>
> thanks<br>
<br>
The way I interpret equation (6), you are to select as pair (lx,ly), from among the set of those (xs,ys) for which ys < yc, that one for which xs is minimum (that is, farthest to the left.) If I have that right, you can do it this way. Suppose xs and ys are both vectors of the extremal point coordinates.<br>
<br>
p = find(ys<yc); % Limit the choice to ys's which are below yc<br>
[lx,q] = min(xs(p); % From among these choose the smallest xs<br>
ly = ys(p(q)); % Then choose its matching ys<br>
<br>
This presupposes that a ys element exists which is less than the threshold value yc. Otherwise p is empty and no valid answer is produced.<br>
<br>
This is simple to compute but in my opinion it is a rather crude criterion to use. If yc is too high, a hand may be chosen, and if it is too low, no answer is forthcoming. Ideally the angle should also play some role in the selection of (lx,ly).<br>
<br>
Roger Stafford

Wed, 11 Mar 2009 09:55:03 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#633996
Regina
hi mr stafford<br>
<br>
i have already implemented the code that you have given and im really thankful for your help.<br>
however, i have a difficulty in determining the value for phi because it just gives the inverse plot of theta.<br>
how would i implement equation 6 of the pdf for the torso angle or phi?<br>
thanks!

Thu, 12 Mar 2009 00:30:05 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#634203
Regina
hi mr stafford<br>
<br>
i have already implemented the code that you have given and im really thankful for your help.<br>
however, i have a difficulty in determining the value for phi because it just gives the inverse plot of theta.<br>
how would i implement equation 6 of the pdf for the torso angle or phi? and also how is the preemphasis filter done and how to use autocorrelation in the signal?<br>
<br>
thanks for all the help!

Wed, 16 Jun 2010 17:17:05 +0000
Re: can someone help me with this
http://www.mathworks.com/matlabcentral/newsreader/view_thread/246176#754974
Manal
"Regina " <velasquezregina@rocketmail.com> wrote in message <gp9l2c$jl9$1@fred.mathworks.com>...<br>
> hi mr stafford<br>
> <br>
> i have already implemented the code that you have given and im really thankful for your help.<br>
> however, i have a difficulty in determining the value for phi because it just gives the inverse plot of theta.<br>
> how would i implement equation 6 of the pdf for the torso angle or phi? and also how is the preemphasis filter done and how to use autocorrelation in the signal?<br>
> <br>
> thanks for all the help! <br>
<br>
Hi regina<br>
could you help me in this algorithm<br>
I am don't know how to extract the extreme points by using the zerocrossings <br>
<br>
waiting to hair from you<br>
thanks