Color Image Super-Resolution Using Geometric Grouplets
Résumé
In this work, a technique for generating a super-resolution (SR) image from a single multi-valued low-resolution (LR) input image is proposed. This problem is approached from the perspective of image geometry-oriented interpolation. First, the geometry of the LR image is obtained by computing the grouplet transform. These grouplet bases are used to define a grouplet-based structure tensor to capture the geometry and directional features of the LR colour image. Then, the SR image is synthesised by an adaptive directional interpolation using the extracted geometric information to preserve the sharpness of edges and textures. This is accomplished by the minimisation of a functional, which is defined on the extracted geometric parameters of the LR image and oriented by the geometric flow defined by the grouplet transform. The proposed SR algorithm outperforms the state-of-the-art methods in terms of visual quality of the interpolated image.