Introduction of a Symmetric Tight Wavelet Frame to Image Fusion Methods Based on Substitutive Wavelet Decomposition

A useful technique in various applications of remote sensing involves the fusion of panchromatic and multispectral images. Wavelet-based approaches to image fusion generally produce high-quality spectral content in fused images. However, the spatial resolution obtained by most wavelet-based methods is less than that obtained by the intensity-hue-saturation (IHS) method. Recent studies show that if an undecimated discrete wavelet transform (DWT) is used for image fusion, the spatial resolution of the fused images can be as good as that of images obtained by the IHS method. This effect occurs because an undecimated DWT is exactly shift-invariant. In this paper, the author introduces a symmetric tight wavelet frame to image fusion methods that are based on substitutive wavelet decomposition. The introduced tight wavelet frame transform is nearly shift-invariant with desired properties such as wavelet smoothness, short support, and symmetry. The experimental results show the possibility as an alternative DWT approach for image fusion. In addition, the author proposes a fast algorithm for an improved IHS method introduced by González-Audı́cana et al. The proposed approach enables a fast, easy, and extendable implementation. Hence, for the fusion of IKONOS panchromatic and multispectral images, the near-infrared band of IKONOS may be included in the definition of the intensity component. This approach produces satisfactory results, both visually and quantitatively.