Up: Solar image segmentation by
Sunspots and sunspot groups are the most conspicuous features of solar
activity. They appear as dark areas in the photosphere of the sun,
with sizes and numbers that vary in the 11-year cycle and they are
caused by a lower surface temperature.
The area covered by the sunspot is one of the most relevant parameters used to investigate
the solar activity since the sunspot number is only an empirical index.
In many occasions a phase delay between the value of the sunspot number and the projected
areas of sunspots has been observed (e.g. Pap 1985).
There is some recent evidence pointing to a relation between magnetic field strength,
continuum intensity and sunspot size (Collados et al. 1994).
The ratio of the sunspot and facular areas is linked with the age of
the active regions (young or old sunspot groups) (Pap 1985).
The energy balance of individual active regions seems to depend mainly
on the area ratio of the plage and the corresponding sunspot group
(Steinegger et al. 1996). During the solar minimum, when
only a few active regions exist on the Sun, the main contribution to
the irradiance variations arises from the active network, which is
formed by the breakup and dispersion
of the active regions (Fröhlich
& Pap 1989).
The determination of sunspot area becomes
more complicated when one attempts to separate the individual areas
of umbra and penumbra.
Steinegger et al. (1997) compare
two methods of measuring sunspot areas from
photospheric observations; one based on histogram modifications and a
new method termed inflexion point method (IPM).
Therefore, it is of great importance to isolate the
basic characteristics of the sunspot or sunspot group, in order to compare
them with active regions and to study the results of solar activity.
In , the chromosphere is particularly bright
and the matter is apparently dominated by the chromospheric magnetic field lines.
The purpose of segmentation is to divide the image into meaningful
regions that correspond to structural units in the scene or distinguish
objects of interest.
In this case it could be useful if one wants to
isolate automatically the basic characteristics of the sunspot or
sunspot group and study their relation with solar activity using
observations.
The segmentation is separated here in two stages, the
classification in q labels (Sect. 2) and the
relaxation labeling process (Sect. 3).
Up: Solar image segmentation by
Copyright The European Southern Observatory (ESO)