Abstract
An image up-scaling algorithm that can also reduce salt and pepper noise is presented. The algorithm is based on a least mean square optimisation framework which trains on original images and the downsampled images corrupted by salt and pepper noise. Hybrid filters which are composed of structure-adaptive linear filters and rank-order filters are used. Experimental results show the effectiveness of the proposed algorithm.
| Original language | English |
|---|---|
| Pages (from-to) | 746 |
| Journal | Electronics Letters |
| Volume | 43 |
| Issue number | 14 |
| DOIs | |
| Publication status | Published - 5 Jul 2007 |