PCA Based Generalized Interpolation for Image Super-Resolution.
C. V. Jiji, Subhasis Chaudhuri · 2004
In this paper we propose an eigenimage based super-resolution reconstruction technique. Eigenimages of a database of several similar low resolution images are ob-tained and the given low resolution image is projected on to the eigenimages to compute the eigenimage coefcients. The eigenimages are then interpolated using any conven-tional interpolation method and approximated to the near-est orthonormal bases. The high resolution image is recon-structed using these bases and the same coefcients deter-mined as above. This method is applicable to images of a particular class and results are demonstrated for both face and ngerprint images. The algorithm offers signicant ad-vantage when the input image is blurred and noisy. 1.