الفهرس 
Introduction
BoostingOnly 14 pages are availabe for public view
 |  | from | 81 |  |  |
| | | from | 81 | | |
Abstract
Sunspots that appear as dark spots on the Sun’s surface have been the subject of interest to astronomers to understand their effect on the Earth’s weather. Sunspot recognition and classification are very difficult process because of the wide margin in the interpretation of classification rules, but the process could be automated if successfully learned by a machine. This report describes automatic sunspot recognition and classification using a combination of image processing and machine learning to extract individual sunspots with their attributes and classify them. The data was obtained from NASA SOHO/MDI satellite (http://www.solar.ifa.hawaii.edu /MWLT/archive.html) and the classification scheme attempted was the seven-class Modified Zurich scheme. In the training dataset sunspots were manually classified by comparing extracted sunspots with corresponding active region maps (ARMaps) from the Mees Observatory at the Institute for Astronomy, University of Hawaii (http://www.solar.ifa.hawaii.edu/ ARMaps/Search /ARMaps_200103.html) . A series of experiments were performed on the testing dataset and results showed that K-nearest neighbor approach gives 52% while the better success rate 63.7% is obtained by decision tree , and if boosting is added, the performance will be reach to (100%). This report consists of 5 chapters. Chapter 1 presents general introduction about sunspots properties, sunspots life-cycle , classification schema , effects of sunspots on the earth and Related work. Chapter 2 presents the formulation of the problem. Chapter 3 presents image processing techniques that used in both sunspot recognition and classification in details. Section 3.1 presents image enhancement technique which is dual tree complex wavelet that used as the first step in the hybrid system after the data are collected. Section 3.2 presents sunspot recognition techniques. Section 3.3 shows binary dilation and erosion to make a combination of them in the sunspots grouping step. Finally, section 3.4 presents machine learning techniques that used in sunspot classification. Chapter 4 presents the results. Conclusion and future work are summarized in chapter 5.