Hi Luis,<br>
<br>
Thanks for the explanation, now i do understand and appreciate the
motive behind the metric, the way it is. It is only the name of the
metric "<span style="font-weight: bold;">Mean</span> Reciprocal Square
Metric" which caused the confusion. But in fact its behaving like a
(Sum of (Normalized (Squared Differences))) Metric ... <br>
<br>
may be a comment about this should be added in the documentation as well.<br>
<br>
with regards,<br>
Ashish.<br>
<br>
<br><br><div><span class="gmail_quote">On 7/8/05, <b class="gmail_sendername">Luis Ibanez</b> <<a href="mailto:luis.ibanez@kitware.com">luis.ibanez@kitware.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br><br>Hi Ashish,<br><br>No,<br>this Metrics *must not* be divided by the number of pixels counted.<br><br>In fact, the nice property of this metric is that it gets better<br>when you use many pixels.<br><br>Here is the rationale:
<br><br>When you use a metric such as mean squares, where the contributions<br>of individual pixels are "good" when they have numerical values,<br>you have the conflicting situation that you can get a good value
<br>because the pixels are very close in intensities, or because you<br>have just a few pixels. Therefore, in order to eliminate the second<br>case you are forced to divide by the number of pixels.<br><br>In the case of the reciprocal metric, the more pixels you count,
<br>the better the metric gets, and the closer their intensities are<br>between the fixed and the moving image, the better the metric gets.<br><br>Dividing by the number of pixels will destroy the most interesting<br>property of this metric.
<br><br><br>Another nice property of this metric is the fact that you know what<br>the optimal value would be. It is equal to the number of pixels<br>counted. E.g. this is obtained only when all the intensities of the<br>
fixed image pixels are equal to the intensities of the moving image<br>pixels.<br><br><br><br>Please let us know if you have any other questions,<br><br><br> Thanks<br><br><br><br> Luis<br><br><br>---------------------
<br>Ashish Poddar wrote:<br><br>> Hi,<br>><br>> in mean reciprocal square image to image metric, the value being<br>> returned is simply the sum of 1.0f / ( 1.0f + m_Lambda* ( diff * diff )<br>> ); for all the differences computed throughout the image. isnt it
<br>> supposed to be divided by the this->m_NumberOfPixelsCounted in the end<br>> before returning?<br>><br>> with regards,<br>> Ashish.<br>><br><br><br><br></blockquote></div><br><br><br>-- <br>Ashish Poddar
<br>Have an acceptable reason for accepting anything.<br>Y:ashish_poddar | <a href="mailto:MSN:ashish_poddar@yahoo.com">MSN:ashish_poddar@yahoo.com</a>