排序方式： 共有8条查询结果，搜索用时 890 毫秒
1
1.
In this work we study the association between eruptive filaments/prominences and coronal mass ejections (CMEs) using machine learning-based algorithms that analyse the solar data available between January 1996 and December 2001. The support vector machine (SVM) learning algorithm is used for the purpose of knowledge extraction from the association results. The aim is to identify patterns of associations that can be represented using SVM learning rules for the subsequent use in near real-time and reliable CME prediction systems. Timing and location data in the US National Geophysical Data Center (NGDC) filament catalogue and the Solar and Heliospheric Observatory/Large Angle and Spectrometric Coronagraph (SOHO/LASCO) CME catalogue are processed to associate filaments with CMEs. In the previous studies, which classified CMEs into gradual and impulsive CMEs, the associations were refined based on the CME speed and acceleration. Then the associated pairs were refined manually to increase the accuracy of the training dataset. In the current study, a data-mining system is created to process and associate filament and CME data, which are arranged in numerical training vectors. Then the data are fed to SVMs to extract the embedded knowledge and provide the learning rules that can have the potential, in the future, to provide automated predictions of CMEs. The features representing the event time (average of the start and end times), duration, type, and extent of the filaments are extracted from all the associated and not-associated filaments and converted to a numerical format that is suitable for SVM use. Several validation and verification methods are used on the extracted dataset to determine if CMEs can be predicted solely and efficiently based on the associated filaments. More than 14?000 experiments are carried out to optimise the SVM and determine the input features that provide the best performance. 相似文献
2.
Machine-learning algorithms are applied to explore the relation between significant flares and their associated CMEs. The
NGDC flares catalogue and the SOHO/LASCO CME catalogue are processed to associate X and M-class flares with CMEs based on
timing information. Automated systems are created to process and associate years of flare and CME data, which are later arranged
in numerical-training vectors and fed to machine-learning algorithms to extract the embedded knowledge and provide learning
rules that can be used for the automated prediction of CMEs. Properties representing the intensity, flare duration, and duration
of decline and duration of growth are extracted from all the associated (A) and not-associated (NA) flares and converted to
a numerical format that is suitable for machine-learning use. The machine-learning algorithms Cascade Correlation Neural Networks
(CCNN) and Support Vector Machines (SVM) are used and compared in our work. The machine-learning systems predict, from the
input of a flare’s properties, if the flare is likely to initiate a CME. Intensive experiments using Jack-knife techniques
are carried out and the relationships between flare properties and CMEs are investigated using the results. The predictive
performance of SVM and CCNN is analysed and recommendations for enhancing the performance are provided. 相似文献
3.
Solar Feature Catalogues In Egso 总被引：1，自引：0，他引：1
V. V. Zharkova J. Aboudarham S. Zharkov S. S. Ipson A. K. Benkhalil N. Fuller 《Solar physics》2005,228(1-2):361-375
The Solar Feature Catalogues (SFCs) are created from digitized solar images using automated pattern recognition techniques
developed in the European Grid of Solar Observation (EGSO) project. The techniques were applied for detection of sunspots,
active regions and filaments in the automatically standardized full-disk solar images in Caii K1, Caii K3 and Hα taken at the Meudon Observatory and white-light images and magnetograms from SOHO/MDI. The results of automated
recognition are verified with the manual synoptic maps and available statistical data from other observatories that revealed
high detection accuracy. A structured database of the Solar Feature Catalogues is built on the MySQL server for every feature
from their recognized parameters and cross-referenced to the original observations. The SFCs are published on the Bradford
University web site http://www.cyber.brad.ac.uk/egso/SFC/ with the pre-designed web pages for a search by time, size and location. The SFCs with 9 year coverage (1996–2004) provide
any possible information that can be extracted from full disk digital solar images. Thus information can be used for deeper
investigation of the feature origin and association with other features for their automated classification and solar activity
forecast. 相似文献
4.
5.
Zharkova V.V. Ipson S.S. Zharkov S.I. Benkhalil A. Aboudarham J. Bentley R.D. 《Solar physics》2003,214(1):89-105
Robust techniques are developed to put the H and Ca K line full-disk images taken at the Meudon Observatory into a standardised form of a `virtual solar image'. The techniques include limb fitting, removal of geometrical distortion, centre position and size standardisation and intensity normalisation. The limb fitting starts with an initial estimate of the solar centre using raw 12-bit image data and then applies a Canny edge-detection routine. Candidate edge points for the limb are selected using a histogram based method and the chosen points fitted to a quadratic function by minimising the algebraic distance using SVD. The five parameters of the ellipse fitting the limb are extracted from the quadratic function. These parameters are used to define an affine transformation that transforms the image shape into a circle. Transformed images are generated using the nearest neighbour, bilinear or bicubic interpolation. Intensity renormalisation is also required because of a limb darkening and other non-radial intensity variations. It is achieved by fitting a background function in polar coordinates to a set of sample points having the median intensities and by standardising the average brightness. Representative examples of intermediate and final processed results are presented in addition to the algorithms developed. The research was done for the European Grid of Solar Observations (EGSO) project. 相似文献
6.
We describe the automated extraction of active regions (ARs) or plages from the European Grid of Solar Observations (EGSO) Solar Feature Catalogue using a region-growing technique. In this work, Hα and Ca ii K3 solar images from the Meudon Observatory and EUV solar images from the SOHO/EIT instrument were used. For better detection accuracy, the statistical properties of each quarter of a full disk solar image are used to define local intensity thresholds for an initial segmentation that helps to define AR seeds. Median filtering and morphological operations are applied to the resulting binary image in order to remove noise and to merge broken regions. The centroids of each labelled region are used as seeds, from which a region-growing procedure starts. Statistics-based local thresholding is also applied to compute upper- and lower- threshold intensity values defining the spatial extents of the regions. The detection results obtained with the resulting automated thresholding and region-growing (ATRG) procedure are compared day-by-day with the synoptic maps manually generated by the Meudon Observatory and NOAA for 2 months in 2002 and more coarsely over a 5-year period. The moderate correlation found between our detection results and those produced manually on the other data sets reveals a need for a unified active region definition. As an application of the SFC for ARs we present the tracking of the active region AR NOAA 10484 during its appearance on the solar disk from 19–26 October 2003 and compare its intensity variations for Hα and Fe xii 195 Å wavelengths. 相似文献
7.
S. S. Ipson V. V. Zharkova S. Zharkov A. K. Benkhalil J. Aboudarham N. Fuller 《Solar physics》2005,228(1-2):399-421
We present an automated technique for comparison of magnetic field inversion-line maps from SOHO/MDI magnetograms with solar
filament data from the Solar Feature Catalogue created as part of the European Grid of Solar Observations project. The Euclidean
distance transform and connected component labelling are used to identify nearest inversion lines to filament skeletons. Several
filament inversion-line characteristics are defined and used to automate the decision whether a particular filament/inversion-line
pair is suitable for quantitative comparison of orientation and separation. The technique is tested on 551 filaments from
14 Hα images at various dates, and the distributions of angles and distances between filament skeletons and line-of-sight
(LOS) magnetic inversion lines are presented for six levels of magnetic field smoothing. The results showed the robustness
of the developed technique which can be applied for a statistical analysis of magnetic field in the vicinity of filaments.
The method accuracy is limited by the static filament detection which does not distinguish between filaments, fibrils, pre-condensations
and filament barbs and this may increase the asymmetries in magnetic distributions and broadening in angular distributions
that requires the incorporation of a feature tracking technique. 相似文献
8.
The first statistical results in sunspot distributions in 1996–2004 obtained from the Solar Feature Catalogues (SFC) are presented.
A novel robust technique is developed for automated identification of sunspots on SOHO/MDI white-light (WL) full-disk solar
images. The technique applies image standardization procedures for elimination of the limb darkening and non-circular image
shape, uses edge-detection methods to find the sunspot candidates and their edges and morphological operations to smooth the
features and fill in gaps. The detected sunspots are verified with the SOHO/MDI magnetograms by strong magnetic fields being
present in sunspots. A number of physical and geometrical parameters of the detected sunspot features are extracted and stored
in the relational SFC database including umbra/penumbra masks in the form of run-length data encoding of sunspot bounding
rectangles. The detection results are verified by comparison with the manual daily detection results in Meudon and Locarno
Observatories in 2002 and by correlation (about 96%) with the 4 year sunspot areas produced manually at NOAA. Using the SFC
data, sunspot area distributions are presented in different phases of the solar cycle and hemispheres which reveals a periodicity
of the north–south asymmetry with a period of about 7–8 years. The number of sunspots increases exponentially with the area
decrease with the index slightly increasing from −1.15 (1997) to −1.34 (2001). 相似文献
1