Abstract
The main causes and characteristics of stripe noises in OMIS hyperspectral data were analyzed. Based on the previous analysis, two improved destriping methods of OMIS images were developed. The result of comparing the tested images with data indicates that both the new methods could effectively remove stripe noises and preserve the textures and details of the OMIS images better than the existed methods.
| Original language | English |
|---|---|
| Pages (from-to) | 2148-2152 |
| Number of pages | 5 |
| Journal | Guangzi Xuebao/Acta Photonica Sinica |
| Volume | 36 |
| Issue number | 11 |
| State | Published - Nov 2007 |
Keywords
- Destriping
- Moment matching
- OMIS
- Wavelet transform