Tarbiat Modares University The Modares Journal of Electrical Engineering 2228-527X 11 3 2011 10 01 Multispectral and Panchromatic Image Fusion by Combining Spectral PCA and Spatial PCA Methods ترکیب روش‌های تحلیل مولفه اصلی مکانی و تحلیل مولفه اصلی طیفی به منظور ادغام تصاویر چند طیفی و تکرنگ 19 27 12895 EN Hamid Reza Shahdoosti Ph.D. Student, Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran, hamidreza. Hesan Ghassemian Professor, Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran, Journal Article An ideal fusion method preserves the spectral information in fused image without spatial distortion. The PCA is believed to be a well-known pan-sharpening approach and being widely used for its efficiency and high spatial resolution. However, it can distort the spectral characteristics of multispectral images. The current paper tries to present a new fusion method based on the same concept. In the conventional standard PCA method, PCA transform is applied to spectral bands of multispectral images, but we applied the PCA transform to pixel blocks instead. Since PCA coefficients are extracted from statistical properties of the image, it is more consistent with type and texture of remotely sensed image compared to other kernels such as wavelets. After that, a new hybrid algorithm is proposed which uses both the spatial PCA and the spectral PCA method to improve the quality of the merged images. Visual and statistical analyses show that the proposed algorithm clearly improves the merging quality in terms of RASE, ERGAS, SAM, correlation coefficient and UIQI; compared to fusion methods such as IHS, Brovey, PCA, HPF, and HPM. https://mjee.modares.ac.ir/article_12895_e85f439bc16dc2341eb5957a1e9c2f5f.pdf