Image Fusion Based on Wavelet and Curvelet Transform

Image Fusion is a combination of two or more different images. It extracts the information from multiple source images. Wavelet based image fusion is suitable for representing the point singularities in one dimensions, but it fails to represent the edges across the curves in two dimensions. In this method, pixel level fusion based on curvelet transform detects the discontinuities across the curves. Curvelet improves the spatial characteristics and at the same time preserves the frequency characteristics. In curvelet transform operation, the image is first decomposed using the discrete wavelet transform to produce subbands at various levels. Then smooth partitioning is carried out by applying the grid of squares to the subband image, and a smooth window function is applied on that square. In the next step, each square is renormalized into unit square and then ridgelet transform is applied to get the fused image.