Speaker
Description
First Name: Sanmoy
Last Name: Bandyopadhyay
Affiliation: Sister Nivedita University, Kolkata
All Authors: Sanmoy Bandyopadhyay, Vaibhav Pant
Abstract: This work presents a method for detecting chromospheric plages in 393.416 nm images captured by the Precision Solar Photometric Telescope (PSPT) at the Mauna Loa Solar Observatory (MLSO), Hawaii. Plages are bright chromospheric regions associated with sunspots and active areas, and their detection plays a central role in understanding solar magnetic evolution and long-term solar cycle variability. Like most solar imaging data, these observations exhibit intensity inhomogeneity, noise, limb darkening effects, and diffuse or irregular boundaries, all of which make accurate plage identification challenging and often limit the reliability of classical methods. To address these issues, a novel adaptive Fuzzy C-Means (FCM) clustering approach is introduced, beginning with an intensity-based pre-masking step that highlights the most prominent bright regions as the initial foreground. The remaining mixed-intensity pixels are then segmented using FCM to achieve finer separation between foreground and background. The combined output is further refined using connected component analysis to remove spurious detections, merge fragmented structures, and isolate the true plage regions. Visual comparisons with traditional techniques show that this method achieves improved accuracy and consistency in detecting plage structures in Ca II K images, and it can be extended to NB03 SUIT/Aditya-L1 data.