select.html

this is code for finding and tracking feature points

<title> </title>

<a href = "../index.html">
<IMG SRC="../home_motif.gif" ALIGN=bottom>
</a>
<a href = "index.html">
<IMG SRC="../toc_motif.gif" ALIGN=bottom>
</a>
<a href = "index.html">
<IMG SRC="../previous_motif.gif" ALIGN=bottom>
</a>
<a href = "track.html">
<IMG SRC="../next_motif.gif" ALIGN=bottom>
</a>

<hr>

<h2> KLTSelectGoodFeatures() </h2> 

<h2> 
<code>
void KLTSelectGoodFeatures(<br>
     KLT_TrackingContext tc,<br>
     KLT_PixelType *img,<br>
     int ncols,<br>
     int nrows,<br>
     KLT_FeatureList fl);<br>
</code>
</h2>

KLTSelectGoodFeatures() takes an image pointed to by 
<code>img</code>. (NOTE: <code>KLT_PixelType</code>
is, unless modified by the user, an <code>unsigned char</code>.)  If
<code>tc->smoothBeforeSelecting</code> is set to <code>TRUE</code>, then the 
image is smoothed by
convolving with a Gaussian of sigma = <code>tc->smooth_sigma_fact *
max(tc->window_width, tc->window_height)</code>;  otherwise, the image
is not smoothed.  In either case,
gradients are computed from the resulting image by convolving with the
derivative of a Gaussian of sigma = <code>tc->grad_sigma</code>.  These 
gradients (one in
the <i>x</i> direction and the other in the <i>y</i> direction) are used 
to select the features.<p>

Pixels throughout the image are then measured as to their "goodness", which
is a measure of their trackability.
Generally each pixel within the image's interior is considered, where the
interior is defined by <code>tc->borderx</code> and 
<code>tc->bordery</code> (i.e., setting these
parameters to zero causes the interior to equal the whole image).
The parameter <code>tc->nSkippedPixels</code> can be used to speed up the process in
the following way:  its default value is zero, in which case every pixel within
the interior is considered; if it is set to one,
then every other pixel within the interior is considered; setting
it to two causes every third pixel to be considered; and similarly for
higher values.  Since neighboring 
pixels generally have similar goodness values, then skipping every
other one will probably not noticeably decrease performance.<p>

The goodness of each pixel is measured as the minimum eigenvalue of
the 2 by 2 gradient matrix computed from the <code>tc->window_width</code> by
<code>tc->window_height</code> window around the pixel.  After all the pixels have been
considered, they are sorted in descending order according to goodness.
Then, one by one the top <code>fl->nFeatures</code> 
features (or pixels) whose minimum eigenvalue is at least 
<code>tc->min_eigenvalue</code> are selected, ensuring that each new feature is at 
least <code>tc->mindist</code> pixels away from all the other features.<p>

If <code>tc->writeInternalImages</code> is <code>TRUE</code>, then 
the smoothed image and the image derivatives are written to 
<code>"kltimg_sgfrlf.pgm"</code>,
<code>"kltimg_sgfrlf_gx.pgm"</code>, and
<code>"kltimg_sgfrlf_gy.pgm"</code>, respectively.  
This allows the user to more intelligently select the parameters
for smoothing and differentiating.

<p>
<hr>

<a href = "../index.html">
<IMG SRC="../home_motif.gif" ALIGN=bottom>
</a>
<a href = "index.html">
<IMG SRC="../toc_motif.gif" ALIGN=bottom>
</a>
<a href = "index.html">
<IMG SRC="../previous_motif.gif" ALIGN=bottom>
</a>
<a href = "track.html">
<IMG SRC="../next_motif.gif" ALIGN=bottom>
</a>