Abstract
A new method is proposed for the 3D (three-dimensional) image restoration of the wide-field microscope based on the MPMAP algorithm (Poisson-MAP Super-resolution image restoration algorithm with Markov constraint) according to the 3D features of the microscopic image. The neighborhood of the Markov random field in MPMAP algorithm is extended to 3D, and the regularization parameter α of the bound term in MPMAP is simplified. As a result, the restoration of 3D image of wide-field microscope is achieved, and the more perfect effect of the image restoration is got. When images within noise are restored by different value α, different attainable resolution and signal-to-noise ratio (SNR) in the restored image. Experiment results show that it is necessary to select appropriately value of α, and take tradeoff between resolution and SNR in the restored image so that the more perfect effect of the image restoration is got.
| Original language | English |
|---|---|
| Article number | 60260L |
| Journal | Proceedings of SPIE - The International Society for Optical Engineering |
| Volume | 6026 |
| DOIs | |
| Publication status | Published - 2006 |
| Event | ICO20: Biomedical Optics - Changchun, China Duration: 21 Aug 2005 → 26 Aug 2005 |
Keywords
- 3D continuation of the neighborhood
- 3D wide-field microscopic image
- Image restoration
- MPMAP algorithm
- Markov random field
- Regularization parameter α