Mining Coronal Loops in Solar Images from the SOHO collection

Abstract

We present our preliminary findings as part of a coronal condensations and bright spots, are common new data mining application aiming at the automatic detection around the time of solar maximum (the sun goes through of images with coronal loops from one of NASA's solar image a cycle of activity every 11 years), while larger faint databases, known as EIT. Coronal loops are immense arches of hot gas on the surface of the Sun, thought to be jets of hot onest last d solr weeks a ty pical o e plasma flowing along in the alleys between the strong coronal quiet corona, when solar activity is low. The two ends magnetic fields. We use various data mining techniques including of a loop, known as footprints, lie in regions of the combining crisp and fuzzy classifiers for automated detection of photosphere of opposite magnetic polarity to each other. blocks extracted from EIT solar images. Our data mining and Coronal loops have been linked to complex phenomena retrieval system helps provide relevant data to astrophysicists who need such data to study the solar corona, and whose work onates sn, such asvolent corona mas rejtions, solar is traditionally hindered by the need to manually sift through flares, and solar storms, that affect the rest of the solar thousands of images in order to locate the very few that are useful system, by shaking the Earth's magnetic field and power for further analysis. Our data-driven approach is distinct from grids, and possibly harming satellites and astronomers related image processing based approaches that cannot scale to in space. Measurements of the temperature distribution large image databases because they rely mostly on semi-automated along the loop length can be used to support or eliminate detection and on heavy and computationally intensive local shape analysis.'234 various classes of coronal temperature models [7]. In order to make progress, scientific analysis requires data observed by instruments such as EIT, TRACE, and I. INTRODUCTION SXT [7]. The combination of EIT, TRACE, and SXT The Coronal Heating Problem [8,9] is one of the information provides a powerful data set that will yield longest standing unsolved mysteries in astrophysics, and unprecedented detail on the plasma parameters of a is essentially concerned with understanding and model-variety of coronal loop structures. The biggest obstacle ing the exact properties of temperature distribution along to completing studies of the solar loop properties has coronal loops. The corona is the uppermost level of the been putting the data set together, so that the subseSun's (or another star's) atmosphere, lying immediately quent scientific analysis is performed. The search for above its visible surface. Coronal loops (see Figure 1) interesting images (the ones with coronal loops) is by far are immense arches of hot gas on the surface of the the most time consuming aspect of this scientific task, Sun, thought to be jets of hot plasma flowing along in amounting to a "search for a needle in a haystack". In the alleys between the st...

Coronal loop detection from solar images and extraction of salient contour groups from cluttered images.

“…Skewness, Jl3 given in Eq. (2)(3)(4)(5)(6)(7)(8), measures the symmetry of distribution. Skewness is 0 for symmetric histograms, positive for right skewed histograms, and negative for left skewed histograms.…”

“…An ellipse can be described by its center coordinates (p, q), semi major axis length (a), semi minor axis length (b), and orientation (8). Finding the ellipse that passes through the given points requires calculating these five parameters (McLaughlin, 1998;Duda, et aI., 1972;Tsuji, et aI., 1978).…”

“…(3)(4)(5)(6)(7)(8), where n is the number of segments in segment list, S. If there is a single segment in S, then 8 is assigned the value 0°. To reduce the influence of shorter segments, we also take into account the length of the segments as shown in Eq.…”

“…To reduce the influence of shorter segments, we also take into account the length of the segments as shown in Eq. (3)(4)(5)(6)(7)(8). Shorter segments will have less effect on the average curvature value.…”

“…[4][5][6][7] Another observation about the residual space is that even if two contours do not lie on the same ellipse, the comparison might still not reflect this fact. For example, in [4][5][6][7][8] If there are N contours in the contour group, we derive the Tjump statistics between consecutive contours and take the average of all of them as given in Eq. 4-9.…”