[GRASS-SVN] r65479 - grass-addons/grass7/raster/r.seg
svn_grass at osgeo.org
svn_grass at osgeo.org
Tue Jun 16 02:41:27 PDT 2015
Author: avitti
Date: 2015-06-16 02:41:27 -0700 (Tue, 16 Jun 2015)
New Revision: 65479
Modified:
grass-addons/grass7/raster/r.seg/r.seg.html
Log:
Service commit before module rename [from r.seg to r.smooth.seg]
Modified: grass-addons/grass7/raster/r.seg/r.seg.html
===================================================================
--- grass-addons/grass7/raster/r.seg/r.seg.html 2015-06-16 09:31:31 UTC (rev 65478)
+++ grass-addons/grass7/raster/r.seg/r.seg.html 2015-06-16 09:41:27 UTC (rev 65479)
@@ -1,6 +1,6 @@
<h2>DESCRIPTION</h2>
-<em><b>r.seg</b></em> generates a piece-wise smooth approximation of the
+<em><b>r.smooth.seg</b></em> generates a piece-wise smooth approximation of the
input raster map and a raster map of the discontinuities of the output
approximation. <br>
@@ -16,18 +16,20 @@
shortened too much when the parameter alpha is set to very high values,
(this happens very rarely). <br>
-Some examples of use of the module can be found <a
-href="http://www.ing.unitn.it/~vittia/sw/sw_index.html">here</a> and in <a
-href="http://download.OSGeo.org/OSGeo/foss4g/2009/SPREP/2Thu/Parkside%20GO4/1500/Thu
-G04 1545 Zatelli.pdf">this presentation [FOSS4G 2009 - pdf]</a>. <br>
-For details on the numerical implementation see [3].
+An overview of the underlying theory with some applications cab be found in
+<a href="http://dx.doi.org/10.1016/j.isprsjprs.2012.02.005">[3]</a>.<br>
+Other examples of use of the module can be found
+<a href="http://www.ing.unitn.it/~vittia/sw/sw_index.html">here</a>
+and in
+<a href="http://download.osgeo.org/osgeo/foss4g/2009/SPREP/2Thu/Parkside%20GO4/1500/Thu%20G04%201545%20Zatelli.pdf"
+>this presentation [FOSS4G 2009 - pdf]</a>. <br>
+For details on the numerical implementation see [4].
<h2>NOTES</h2>
-Remove any MASK before the execution of the module. If
-a MASK is present, the module stops after just one iteration.
+Remove any MASK before the execution of the module.
<br><br>
-Replace (<em>r.null</em>) any null data with the map average value (get with <em>r.univar</em>).
+Replace any NULL data (using <em>r.null</em>) with the map average value (get it with <em>r.univar</em>).
<br><br>
The segmentation depends on the parameters alpha and lambda:
@@ -35,33 +37,31 @@
<li> alpha controls how many discontinuities are allowed to exist.
<li> lambda controls the smoothness of the solution.
<li> It is not possible to select the values of the parameters in an
- automatic way. Test some different values to understand their
- influence on the results. Try the following procedure:
+ automatic way. Test some different values to understand their
+ influence on the results.<br>
+ Try the following procedure:
<ul>
<li> run the module with both alpha and lambda set to 1.0
- <li> run the module with alpha set to 1.0 and different values for lambda
- <br> e.g., 0.01, 0.1, 1, 10, 100
- <li> run the module with lambda set to 1.0 and different values for alpha
- <br> e.g., let's say 0.01, 0.1, 1, 10, 100
+ <li> run the module with alpha set to 1.0 and different values for lambda<br>
+ e.g., 0.01, 0.1, 1, 10, 100
+ <li> run the module with lambda set to 1.0 and different values for alpha<br>
+ e.g., 0.01, 0.1, 1, 10, 100
<li> see how the segmentations change and select the values that
- produce the result that best fits your requirements.
+ produce the result that best fits your requirements.
</ul>
</ul>
The module computes the segmentation by means of an iterative
procedure.<br>
-The module stops either when the number of iterations
-reaches the maximum number of iterations [mxi] or when the maximum
-difference between the solutions of two successive iterations is less than
-the convergence tolerance [tol].<br>
-To stop the iteration procedure,
-it is easier to act on the maximum number of iterations parameter [mxi]
-than on the convergence tolerance parameter [tol].<br>
-The number of
-iterations needed to reach the convergence tolerance increases for high
-values of the parameter lambda. The larger the total number of pixels
-of the input raster map the larger the number of iterations will be.
-<br><br>
+The module stops either when the number of iterations reaches the maximum
+number of iterations [mxi] or when the maximum difference between the solutions
+of two successive iterations is less than the convergence tolerance [tol].<br>
+To stop the iteration procedure, it is easier to act on the maximum number of
+iterations parameter [mxi] than on the convergence tolerance parameter [tol].<br>
+The number of iterations needed to reach the convergence tolerance increases
+for high values of the parameter lambda. The larger the total number of pixels
+of the input raster map the larger the number of iterations will be.<br>
+<br>
The data type of the output raster maps is DOUBLE PRECISION. <br><br>
@@ -71,16 +71,15 @@
To avoid to inappropriately re-sampled the input raster map, the settings
for the current region should be set so that:
<ul>
-<li> the resolution of the region matches the resolution of the
- input raster map;
-<li>the boundaries of the region are lined up along the edges of the nearest
-cells in the input raster map.
+<li> the resolution of the region matches the resolution of the input raster map;
+<li> the boundaries of the region are lined up along the edges of the nearest
+ cells in the input raster map.
</ul>
-The discontinuity thickness should be changed for test purposes only.
-<br><br>
+The discontinuity thickness should be changed for test purposes only.<br>
+<br>
-The actual need to use the MSK model should be very rare, see [3].
+The actual need to use the MSK model should be very rare, see [4].
Due to a different implementation of the MSK model with respect to MS
one, the values of the parameters lambda and alpha in MSK have to be
set independently from the values used in MS.
@@ -99,8 +98,8 @@
# select a smaller region:
g.region n=221725 s=220225 w=638350 e=639550
-# run r.seg:
-r.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u_OF out_z=z_OF lambda=10 alpha=200 mxi=250
+# run r.smooth.seg:
+r.smooth.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u_OF out_z=z_OF lambda=10 alpha=200 mxi=250
# for a better visualization of the output raster map <em>u_OF</em>, set its color table to:
r.colors u_OF rast=ortho_2001_t792_1m
@@ -117,26 +116,34 @@
# for a better visualization of the output raster map <em>u_OF</em>, set its color table to:
r.colors z_OF color=bgyr
-# run r.seg with different parameter values:
-r.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u1_OF out_z=z1_OF lambda=10 alpha=65 mxi=250
-r.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u2_OF out_z=z2_OF lambda=10 alpha=600 mxi=250
-r.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u3_OF out_z=z3_OF lambda=0.1 alpha=200 mxi=250
-r.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u4_OF out_z=z4_OF lambda=1 alpha=200 mxi=250
+# run r.smooth.seg with different parameter values:
+r.smooth.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u1_OF out_z=z1_OF lambda=10 alpha=65 mxi=250
+r.smooth.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u2_OF out_z=z2_OF lambda=10 alpha=600 mxi=250
+r.smooth.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u3_OF out_z=z3_OF lambda=0.1 alpha=200 mxi=250
+r.smooth.seg in_g=ortho_2001_t792_1m at PERMANENT out_u=u4_OF out_z=z4_OF lambda=1 alpha=200 mxi=250
# visualize and compare the different results
</pre></div>
-<h2>REFERENCE</h2>
+<h2>REFERENCES</h2>
-<ul> <li> <b>[1]</b> D. Mumford and J. Shah. <em>Optimal Approximation by
+<ul>
+<li> <b>[1]</b> D. Mumford and J. Shah. <em>Optimal Approximation by
Piecewise Smooth Functions and Associated Variational Problems</em>. <br>
-Communications on Pure Applied Mathematics, 42:577-685, 1989.
+Communications on Pure Applied Mathematics, 42(5):577-685, 1989.<br>
+DOI: 10.1002/cpa.3160420503
<li> <b>[2]</b> R. March and M. Dozio. <em>A variational method for the
recovery of smooth boundaries</em>. <br> Image and Vision Computing,
-15:705-712, 1997.
+15(9):705-712, 1997.<br>
+DOI: 10.1016/S0262-8856(97)00002-4
-<li> <b>[3]</b> A. Vitti. <em>Free discontinuity
+<li> <b>[3]</b> A. Vitti. <em>The Mumford-Shah variational model
+for image segmentation: An overview of the theory, implementation and use</em>. <br>
+ISPRS Journal of Photogrammetry and Remote Sensing, 69:50-64, 2012.<br>
+DOI: 10.1016/j.isprsjprs.2012.02.005
+
+<li> <b>[4]</b> A. Vitti. <em>Free discontinuity
problems in image and signal segmentatiion</em>. <br>
Ph.D. Thesis - University of Trento (Italy), 2008. <br> <a
href="http://www.ing.unitn.it/~vittia/misc/vitti_phd.pdf">http://www.ing.unitn.it/~vittia/misc/vitti_phd.pdf</a>
@@ -145,6 +152,7 @@
<h2>SEE ALSO</h2>
+<em><a href="r.clump.html">r.clump</a></em>,
<em><a href="i.smap.html">i.smap</a></em>,
<em><a href="i.zc.html">i.zc</a></em>,
<em><a href="r.mfilter.html">r.mfilter</a></em>
@@ -153,9 +161,8 @@
<h2>AUTHOR</h2>
Alfonso Vitti <br>
- Dept. Civil and
-Environmental Engineering <br>
+ Dept. Civil, Environmental and Mechanical Engineering <br>
University of Trento - Italy<br>
- alfonso.vitti [at] ing.unitn.it
+ alfonso.vitti [at] unitn.it
-<p><i>Last changed: $Date: 2010-08-10 12:00:00 +0200 (Tue, 10 Aug 2010)$</i>
+<p><i>Last changed: $Date: 2015-06-16 12:00:00 +0200 (Tue, 16 Jun 2015)$</i>
More information about the grass-commit
mailing list