Hyperspectral Image Denoising Using a Spatial–Spectral Monte Carlo Sampling Approach
Bibliographic record
Abstract
Hyperspectral image (HSI) denoising is essential for enhancing HSI quality and facilitating HSI processing tasks. However, the reduction of noise in HSI is a difficult work, primarily due to the fact that HSI consists much more spectral bands than other remote sensing images. Therefore, comparing with other image denoising jobs that rely primarily on spatial information, efficient HSI denoising requires the utilization of both spatial and spectral information. In this paper, we design an unsupervised spatial-spectral HSI denoising approach based on Monte Carlo sampling (MCS) technique. This approach allows the incorporation of both spatial and spectral information for HSI denoising. Moreover, it addresses the noise variance heterogeneity effect among different HSI bands. In the proposed HSI denoising scheme, MCS is used to estimate the posterior distribution, in order to solve a Bayesian least squares optimization problem. Based on the proposed scheme, we iterate all pixels in HIS and denoise them sequentially. A referenced pixel in hyperspectral image is denoised as follows. First, some samples are randomly drawn from image space close to the referenced pixel. Second, based on a spatial-spectral similarity likelihood, relevant samples are accepted into a sample set. Third, all samples in the accepted set will be used for calculating the estimation of posterior distribution. Finally, based on the posterior, the noise-free pixel value is estimated as the discrete conditional mean. The proposed method is tested on both simulated and real hyperspectral images, in comparison with several other popular methods. The results demonstrate that the proposed method is capable of removing the noise largely, while also preserving image details very well.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".