Which MI metric are you using Mattes or MutualInformationImageToImageMetric ?<br><br>In the very last line of the metric you will notice that it is negated. ie the computed metric is H(x,y) - H(x) - H(y). Hence it is to be minimized (maximizing the true definition of Mutual information).<br>
<br>The other metric (MutualInformationImageToImageMetric) will yield positive values and needs to be maximized.<br><br>--<br>karthik<br><br><div class="gmail_quote">On Wed, Jan 14, 2009 at 5:08 PM, Dongfeng Han <span dir="ltr"><<a href="mailto:handongfeng@gmail.com">handongfeng@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">hello<br>
I have read the guide book on page 452<br>
<br>
I(a,b) =H(a)+h(B)-H(a,b)<br>
because H(a,b)<=H(a)+h(b), so I(a,b)should be >=0;<br>
so , we should maximize the I(a,b) for registration.<br>
But my output for registration two 3d volume is like below. (the<br>
second column is the MI value)<br>
At each iteraton is decrease and the value is non-positive.<br>
It conflict with my analysis.<br>
so anybody can give me some suggestion?<br>
thanks<br>
0 0 -0.271443 0<br>
0 0 -0.286633 0<br>
0 0 -0.286633 0.00620487<br>
1 0 -0.300877 0.00620487<br>
1 0 -0.300877 0.00304002<br>
2 0 -0.307195 0.00304002<br>
2 0 -0.307195 0.00132559<br>
3 0 -0.223647 0.00132559<br>
3 0 -0.300672 0.00132559<br>
3 0 -0.308016 0.00132559<br>
3 0 -0.308016 0.00116934<br>
4 0 -0.28022 0.00116934<br>
4 0 -0.307456 0.00116934<br>
4 0 -0.308148 0.00116934<br>
4 0 -0.308148 0.00112118<br>
5 0 -0.305431 0.00112118<br>
5 0 -0.308338 0.00112118<br>
5 0 -0.307334 0.00112118<br>
5 0 -0.308314 0.00112118<br>
5 0 -0.308314 0.00107307<br>
6 0 -0.276599 0.00107307<br>
6 0 -0.307406 0.00107307<br>
6 0 -0.308268 0.00107307<br>
6 0 -0.30831 0.00107307<br>
6 0 -0.308314 0.00107307<br>
6 0 -0.308314 0.00107307<br>
6 0 -0.308314 0.00107307<br>
_______________________________________________<br>
Insight-users mailing list<br>
<a href="mailto:Insight-users@itk.org">Insight-users@itk.org</a><br>
<a href="http://www.itk.org/mailman/listinfo/insight-users" target="_blank">http://www.itk.org/mailman/listinfo/insight-users</a><br>
</blockquote></div><br><br clear="all"><br>-- <br>Karthik Krishnan<br>R&D Engineer,<br>Kitware Inc.<br>Ph: 518 371 3971 x119<br>Fax: 518 371 3971<br>