An adaptive PCA-based approach to pan-sharpening
- V. Shah, N. Younan, R. King
- Proc. of SPIE Image and Signal Processing for…
1. INTRODUCTION Pansharpening has been an active research topic in the last few decades and numerous methods have been developed. These methods are generally categorized as arithmetic combination based (AMC) and component substitution (COS) techniques. The AMC methods involve direct arithmetic operation such as multiplication, addition division, weighted adding, etc. on the low resolution multispectral (MS) images to obtain high resolution images. The commonly known methods are Brovey method, Syntheic Variable Ratio (SVR) method, and high pass filtering . The COS-based substitution methods are performed after taking spectral or spatial transformation of the low resolution MS image. The popular COS approaches are the intensity-hue-saturation (IHS), the principal component analysis (PCA), and Multiresolution Analysis (MRA) based pansharpening. The PCA approach has been very commonly used for spectral transformation due to its ability to optimally compress the high dimension data . For this approach, the first principal component (PC) is substituted with the high resolution histogram-matched Pan image. However, the PCA approach is data dependent. For images with mostly vegetation/agricultural contents, this method yields very poor results with high spectral distortion . To alleviate this problem, González-Audícana et. al. proposed a PCA-wavelet merger pan-sharpening method that took the advantage of the component substitution and the currently popular multiresolution approach . Recently our work showed that for the PCA based methods, the standard approach of selecting the first PC is not always a suitable choice, and presented an adaptively method for selecting the PC required to be replaced or injected with high spatial details. The spectral distortion in pan-sharpened images obtained by adaptive PCA (A-PCA) approach was much less than the standard PCA-based approach . There are different approaches to perform spectral transformation – PCA, IHS, Discrete Cosine Transform (DCT), Fast Fourier transform (FFT), wavelets, etc. The transformation obtained by these methods is very data dependent. Previous work indicated that substitution of the component with high variance always yielded good results. Similarly, a single type of transformation does not always yield an optimal component required for substitution or transformation. In order to alleviate this problem, this paper proposes the method to adaptively select the component required for the substitution or injection of the high frequency information.