Abstract Details


Name: Dr. Sanmoy Bandyopadhyay
Affiliation: Aditya-L1 Support Cell, Aryabhatta Research Institute of Observational Sciences (ARIES), Nainital
Conference ID: TVS202510241
Title: A Survey on Fuzzy-Based Approaches in Solar Feature Analysis for Space Weather Forecasting
Authors and Co-Authors: Vaibhav Pant
Abstract Type: Contributory Presentation
Abstract: Being the closest star to Earth, the Sun significantly influences the surrounding space environment, with its activity causing notable variations in space weather. Space weather broadly refers to the physical characteristics and observable phenomena in natural space environments, which can adversely affect critical technological systems both on Earth and in orbit, including electricity grids, aviation tracking, GPS navigation, and satellite communication. Solar phenomena and features such as solar flares, coronal mass ejections (CMEs), sunspots, active regions (ARs), and coronal holes (CHs) are the primary drivers of these events. Among these, CHs, sunspots, and ARs are particularly important because of their strong association with high-speed solar wind streams, solar flare production, and CME initiation. Hence, detecting and analyzing these features is essential for reliable space weather forecasting. Accurate detection, however, is challenging due to the irregular structures in solar images and the blurred boundaries of regions of interest. Fuzzy logic provides a powerful framework to address these issues, as it can model imprecision and handle uncertainty using membership functions and linguistic variables. This review highlights several fuzzy-based methods designed for solar feature detection and prediction. Examples include triangular membership function-based fuzzy C-means for sunspot prediction, fuzzy energy-based dual contour models and fast fuzzy C-means for CHs detection, and the Spatial Possibilistic Clustering Algorithm (SPoCA) for identifying both ARs and CHs. These approaches enhance interpretability while maintaining competitive levels of accuracy. Nonetheless, no single method is universally effective for all image conditions, and challenges remain in terms of computational efficiency, limited labeled datasets, and the demand for real-time processing. Looking ahead, integrating fuzzy logic with hybrid intelligent systems—such as machine learning and physics-informed models—offers a promising pathway toward more robust solar feature detection and improved space weather forecasting.