{"id":{"repo_id":"njit","oai_identifier":"oai:digitalcommons.njit.edu:dissertations-1807"},"canonical_url":"https://search.dev.ndltd.org/etd/njit/oai:digitalcommons.njit.edu:dissertations-1807","repository":{"repo_id":"njit","name":"NJIT","base_url":"https://digitalcommons.njit.edu/do/oai/"},"display":{"title":"Automatic solar feature detection using image processing and pattern recognition techniques","abstract":"The objective of the research in this dissertation is to develop a software system to automatically detect and characterize solar flares, filaments and Corona Mass Ejections (CMEs), the core of so-called solar activity. These tools will assist us to predict space weather caused by violent solar activity. Image processing and pattern recognition techniques are applied to this system. For automatic flare detection, the advanced pattern recognition techniques such as Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), and Support Vector Machine (SVM) are used. By tracking the entire process of flares, the motion properties of two-ribbon flares are derived automatically. In the applications of the solar filament detection, the Stabilized Inverse Diffusion Equation (SIDE) is used to enhance and sharpen filaments; a new method for automatic threshold selection is proposed to extract filaments from background; an SVM classifier with nine input features is used to differentiate between sunspots and filaments. Once a filament is identified, morphological thinning, pruning, and adaptive edge linking methods are applied to determine filament properties. Furthermore, a filament matching method is proposed to detect filament disappearance. The automatic detection and characterization of flares and filaments have been successfully applied on H[alpha] full-disk images that are continuously obtained at Big Bear Solar Observatory (BBSO). For automatically detecting and classifying CMEs, the image enhancement, segmentation, and pattern recognition techniques are applied to Large Angle Spectrometric Coronagraph (LASCO) C2 and C3 images. The processed LASCO and BBSO images are saved to file archive, and the physical properties of detected solar features such as intensity and speed are recorded in our database. Researchers are able to access the solar feature database and analyze the solar data efficiently and effectively. The detection and characterization system greatly improves the ability to monitor the evolution of solar events and has potential to be used to predict the space weather.","abstract_html":"The objective of the research in this dissertation is to develop a software system to automatically detect and characterize solar flares, filaments and Corona Mass Ejections (CMEs), the core of so-called solar activity. These tools will assist us to predict space weather caused by violent solar activity. Image processing and pattern recognition techniques are applied to this system. For automatic flare detection, the advanced pattern recognition techniques such as Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), and Support Vector Machine (SVM) are used. By tracking the entire process of flares, the motion properties of two-ribbon flares are derived automatically. In the applications of the solar filament detection, the Stabilized Inverse Diffusion Equation (SIDE) is used to enhance and sharpen filaments; a new method for automatic threshold selection is proposed to extract filaments from background; an SVM classifier with nine input features is used to differentiate between sunspots and filaments. Once a filament is identified, morphological thinning, pruning, and adaptive edge linking methods are applied to determine filament properties. Furthermore, a filament matching method is proposed to detect filament disappearance. The automatic detection and characterization of flares and filaments have been successfully applied on H[alpha] full-disk images that are continuously obtained at Big Bear Solar Observatory (BBSO). For automatically detecting and classifying CMEs, the image enhancement, segmentation, and pattern recognition techniques are applied to Large Angle Spectrometric Coronagraph (LASCO) C2 and C3 images. The processed LASCO and BBSO images are saved to file archive, and the physical properties of detected solar features such as intensity and speed are recorded in our database. Researchers are able to access the solar feature database and analyze the solar data efficiently and effectively. The detection and characterization system greatly improves the ability to monitor the evolution of solar events and has potential to be used to predict the space weather.","abstract_has_math":false,"creators":["Qu, Ming"],"institution":null,"degree_name":"Doctor of Philosophy in Computing Sciences - (Ph.D.)","degree_level":null,"degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Frank Y. Shih","Haimin Wang","Alexandros V. Gerbessiotis"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2006,"date_issued":"2006-01-31T08:00:00Z","date_published":"2006-01-31T08:00:00Z","updated_at":"2026-07-24T03:23:04Z","subjects":["Computer science","Image processing","Pattern recognition","Solar feature detection","Computer Sciences"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.njit.edu/dissertations/752","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Frank Y. Shih","Haimin Wang","Alexandros V. Gerbessiotis"]},{"key":"dc:creator","label":"Author","values":["Qu, Ming"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:type","label":"Dc Type","values":["Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy in Computing Sciences - (Ph.D.)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer science","Image processing","Pattern recognition","Solar feature detection","Computer Sciences"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.njit.edu/dissertations/752"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The objective of the research in this dissertation is to develop a software system to automatically detect and characterize solar flares, filaments and Corona Mass Ejections (CMEs), the core of so-called solar activity. These tools will assist us to predict space weather caused by violent solar activity. Image processing and pattern recognition techniques are applied to this system. For automatic flare detection, the advanced pattern recognition techniques such as Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), and Support Vector Machine (SVM) are used. By tracking the entire process of flares, the motion properties of two-ribbon flares are derived automatically. In the applications of the solar filament detection, the Stabilized Inverse Diffusion Equation (SIDE) is used to enhance and sharpen filaments; a new method for automatic threshold selection is proposed to extract filaments from background; an SVM classifier with nine input features is used to differentiate between sunspots and filaments. Once a filament is identified, morphological thinning, pruning, and adaptive edge linking methods are applied to determine filament properties. Furthermore, a filament matching method is proposed to detect filament disappearance. The automatic detection and characterization of flares and filaments have been successfully applied on H[alpha] full-disk images that are continuously obtained at Big Bear Solar Observatory (BBSO). For automatically detecting and classifying CMEs, the image enhancement, segmentation, and pattern recognition techniques are applied to Large Angle Spectrometric Coronagraph (LASCO) C2 and C3 images. The processed LASCO and BBSO images are saved to file archive, and the physical properties of detected solar features such as intensity and speed are recorded in our database. Researchers are able to access the solar feature database and analyze the solar data efficiently and effectively. The detection and characterization system greatly improves the ability to monitor the evolution of solar events and has potential to be used to predict the space weather."]},{"key":"dc:title","label":"Title","values":["Automatic solar feature detection using image processing and pattern recognition techniques"]}]}],"canonical_facts":{"dc:contributor":["Frank Y. Shih","Haimin Wang","Alexandros V. Gerbessiotis"],"dc:creator":["Qu, Ming"],"dc:description.abstract":["The objective of the research in this dissertation is to develop a software system to automatically detect and characterize solar flares, filaments and Corona Mass Ejections (CMEs), the core of so-called solar activity. These tools will assist us to predict space weather caused by violent solar activity. Image processing and pattern recognition techniques are applied to this system. For automatic flare detection, the advanced pattern recognition techniques such as Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), and Support Vector Machine (SVM) are used. By tracking the entire process of flares, the motion properties of two-ribbon flares are derived automatically. In the applications of the solar filament detection, the Stabilized Inverse Diffusion Equation (SIDE) is used to enhance and sharpen filaments; a new method for automatic threshold selection is proposed to extract filaments from background; an SVM classifier with nine input features is used to differentiate between sunspots and filaments. Once a filament is identified, morphological thinning, pruning, and adaptive edge linking methods are applied to determine filament properties. Furthermore, a filament matching method is proposed to detect filament disappearance. The automatic detection and characterization of flares and filaments have been successfully applied on H[alpha] full-disk images that are continuously obtained at Big Bear Solar Observatory (BBSO). For automatically detecting and classifying CMEs, the image enhancement, segmentation, and pattern recognition techniques are applied to Large Angle Spectrometric Coronagraph (LASCO) C2 and C3 images. The processed LASCO and BBSO images are saved to file archive, and the physical properties of detected solar features such as intensity and speed are recorded in our database. Researchers are able to access the solar feature database and analyze the solar data efficiently and effectively. The detection and characterization system greatly improves the ability to monitor the evolution of solar events and has potential to be used to predict the space weather."],"dc:identifier":["https://digitalcommons.njit.edu/dissertations/752"],"dc:subject":["Computer science","Image processing","Pattern recognition","Solar feature detection","Computer Sciences"],"dc:title":["Automatic solar feature detection using image processing and pattern recognition techniques"],"dc:type":["Dissertation"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_name":["Doctor of Philosophy in Computing Sciences - (Ph.D.)"]},"updated_at":"2026-07-24T03:23:04Z"}