Akademik Veri Yönetim Sistemi
TRAITEMENT DU SIGNAL, cilt.42, sa.6, ss.3033-3044, 2025 (SCI-Expanded)
This study explores the effective use of multispectral data from Landsat 8 and 9 satellites by applying various band combinations and Intensity-Hue-Saturation (IHS) Afusion techniques for the analysis of land surface features. The research highlights the advantages of false-color composites, particularly in identifying vegetation health, soil structure, and water bodies. Band combinations such as B5-B4-B3, B6-B5-B4, and B7-B5-B3 proved successful in distinguishing thematic differences, while the IHS transformation enhanced spatial resolution. Quantitative metrics including Entropy, Structural Similarity Index (SSIM), Root Mean Square Error (RMSE), Universal Image Quality Index (UIQI) and correlation coefficients were used to evaluate the comparative performance of different bands. Findings indicate that Band 6 (SWIR 1) Aprovided the highest structural similarity and image quality, whereas Band 11 (Thermal Infrared (TIR)), despite its rich information content, showed lower structural coherence. In conclusion, the integration of Landsat imagery with appropriate band selection and fusion methods offers a robust and reliable decision-support tool for environmental monitoring, agricultural analysis, and natural resource management.