<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
  <channel>
    <link>http://www.mathworks.com/matlabcentral/newsreader/view_thread/235909</link>
    <title>MATLAB Central Newsreader - dependent threshol</title>
    <description>Feed for thread: dependent threshol</description>
    <language>en-us</language>
    <copyright>&amp;copy;1994-2012 by MathWorks, Inc.</copyright>
    <webmaster>webmaster@mathworks.com</webmaster>
    <generator>MATLAB Central Newsreader</generator>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <ttl>60</ttl>
    <image>
      <title>MathWorks</title>
      <url>http://www.mathworks.com/images/membrane_icon.gif</url>
    </image>
    <item>
      <pubDate>Fri, 12 Sep 2008 02:37:01 -0400</pubDate>
      <title>dependent threshol</title>
      <link>http://www.mathworks.com/matlabcentral/newsreader/view_thread/235909#600005</link>
      <author>ahmed</author>
      <description>hi all&lt;br&gt;
&lt;br&gt;
&lt;br&gt;
i have this question. i complete project in on line/off line signature verification &lt;br&gt;
but now my problem is how to choose a threshold for every person, can any one help me or give me an advise in this. my probelm is that forged and genuine signatures are very close in some times and too different in another time &lt;br&gt;
so this for me very difficult to classify them into two classes </description>
    </item>
    <item>
      <pubDate>Fri, 12 Sep 2008 03:17:22 -0400</pubDate>
      <title>Re: dependent threshol</title>
      <link>http://www.mathworks.com/matlabcentral/newsreader/view_thread/235909#600009</link>
      <author>ImageAnalyst</author>
      <description>On Sep 11, 10:37=A0pm, &quot;Ahmed &quot; &amp;lt;mogwari2...@yahoo.com&amp;gt; wrote:&lt;br&gt;
&amp;gt; hi all&lt;br&gt;
&amp;gt;&lt;br&gt;
&amp;gt; i have this question. i complete project in on line/off line signature ve=&lt;br&gt;
rification&lt;br&gt;
&amp;gt; but now my problem is how to choose a threshold for every person, can any=&lt;br&gt;
&amp;nbsp;one help me or give me an advise in this. my probelm is that forged and ge=&lt;br&gt;
nuine signatures are very close in some times and too different in another =&lt;br&gt;
time&lt;br&gt;
&amp;gt; so this for me very difficult to classify them into two classes&lt;br&gt;
&lt;br&gt;
---------------------------------------------------------------------------=&lt;br&gt;
----------------------&lt;br&gt;
Sorry, one Ph.D. is enough.  I'm not going to do yours for you.  This&lt;br&gt;
is a very challenging problem and not one that can be answered in such&lt;br&gt;
a forum.&lt;br&gt;
Good luck with it,&lt;br&gt;
ImageAnalyst</description>
    </item>
    <item>
      <pubDate>Fri, 12 Sep 2008 03:37:30 -0400</pubDate>
      <title>Re: dependent threshol</title>
      <link>http://www.mathworks.com/matlabcentral/newsreader/view_thread/235909#600012</link>
      <author>Walter Roberson</author>
      <description>Ahmed wrote:&lt;br&gt;
&lt;br&gt;
&amp;gt; i have this question. i complete project in on line/off line signature verification &lt;br&gt;
&amp;gt; but now my problem is how to choose a threshold for every person, can any one &lt;br&gt;
&amp;gt; help me or give me an advise in this. my probelm is that forged and genuine &lt;br&gt;
&amp;gt; signatures are very close in some times and too different in another time &lt;br&gt;
&amp;gt; so this for me very difficult to classify them into two classes &lt;br&gt;
&lt;br&gt;
I think I have read that modern signature verification depends not just&lt;br&gt;
on the final visual representation but also upon the measured pressure&lt;br&gt;
as the signature is written.&lt;br&gt;
&lt;br&gt;
-- &lt;br&gt;
Q = quotation(rand);&lt;br&gt;
if isempty(Q); error('Quotation server filesystem problems')&lt;br&gt;
else sprintf('%s',Q), end</description>
    </item>
  </channel>
</rss>

