Akademik Veri Yönetim Sistemi
32nd IEEE Signal Processing and Communications Applications Conference (SIU), Mersin, Türkiye, 15 - 18 Mayıs 2024, (Tam Metin Bildiri)
The widespread mucilage (sea-snot) formation in the Sea of Marmara in 2021 and 2022 has again shown the importance of environmental protection, monitoring of environmental disasters, and the use of remote sensing for this purpose. The processing of multispectral and hyperspectral data acquired by Earth observation satellites provides important alternatives to costly field and laboratory studies, for the detection and analysis of environmental disasters such as mucilage. In this work, mucilage is detected and analyzed from data acquired by GOKTURK-I, which is one of Turkiye's Earth observation satellites, using an unmixing approach that takes spectral variability into consideration. The proposed approach enables the detection and spectral and spatial analysis of mucilage with high accuracy, without requiring ground truth labels or precisely set thresholds.