Abstract
Algorithms of super-resolution interpolation for high resolution images are studied. Some common interpolation algorithms and super-resolution interpolation algorithms are analyzed. Following the study of MPMAP super-resolution algorithm a kind of interpolation combined bi-cubic interpolation with MPMAP algorithm was proposed. The bi-cubic interpolation is applied to replace the reverse imaging procedure during the MPMAP iterative calculation and preliminary treatment of initial high resolution image. The pulse signal to noise ratio (PSNR) of restored image is improved more than 2 dB compared with bilinear interpolation algorithm. It is further proved that it is effective in improving the PSNR of infrared thermal images and achieve good effects in restoration.
| Original language | English |
|---|---|
| Pages (from-to) | 161-165 |
| Number of pages | 5 |
| Journal | Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology |
| Volume | 27 |
| Issue number | 2 |
| Publication status | Published - Feb 2007 |
Keywords
- Bi-cubic interpolation
- Infrared image
- Markov random field
- Maximum posterior probability
- Super-resolution interpolation