Structure Tensor Based Image Interpolation Method

Feature-preserving image interpolation is an active area in image processing field. In this paper we propose a new edge directed image super-resolution algorithm based on structure tensors. Using an isotropic Gaussian filter, the structure tensor at each pixel of an image is computed. Based on the tangent eigenvector of the structure tensor, the edge direction is determined and used for interpolation. In comparison to some previous edge directed image interpolation methods, the proposed method achieves higher quality in both subjective and objective aspects, and requires less computational time. Also the proposed method outperforms previous methods in case of noisy images.