Digital Image Restoration
| Hi-index | 0.00 |
Signal reconstruction is one of the knowledge extration problems. Especially in this problem, knowledge means underlying signal or image, which is extracted from the downsampled one. In this paper, we propose a novel reconstruction filter which perfectly reconstructs predetermined preferential components, and makes a reconstructed sigal/image agree with the oblique projection of an original one. It enables us to get rid of artifacts which arise in reconstructed signals by the conventional partial projection filter when the number of samples is small compared with the dimension of the approximation subspace. By simulations, we show that the proposed filter performs better than the conventional methods.