Date of Award
Master of Science
Department of Engineering Physics
Ariel O. Acebal, PhD.
An automatic sunspot detection and classification method is developed combining HMII and HMIM imagery procured from the Solar Dynamics Observatory. Iterative global thresholding methods are employed for detecting sunspots. Groups are selected based on heliographic distance between sunspots via area-based grouping lengths. Classifications are applied through logical operators adhering to the standard McIntosh classification system. Calculated sunspot parameters and classifications are validated in three way comparisons between code output, Holloman AFB and the Space Weather Prediction Center. Accuracy is achieved within the margin of difference between Holloman and SWPC reports for sunspot area, number of groups, number of spots, and McIntosh classification using data spanning 6 July 2012 to 29 June 2013: SWPC/Holloman (33.38%,57.48%,87.67%), SWPC/SDO (20.22%,51.25%,83.80%), and SDO/Holloman (24.54%,50.91%,80.65%). The automatic classification system is used to evaluate bias inherent in Holloman classification methods. Parameters are altered to reach optimal match percentages with Holloman, indicating differences between computed parameter values and hand-calculated counterparts. Group length cutoffs are shown to differ by 2.5°, eccentricity is quantified at 0.8, and penumbra length cutoffs are shown to exceed differences of 1.4° from McIntosh values.
DTIC Accession Number
Spahr, Gordon M., "Fully Automated Sunspot Detection and Classification Using SDO HMI Imagery in MATLAB" (2014). Theses and Dissertations. 662.