Asymmetric Least Squares
Applies asymmetric least-squares baseline smoothing to spectra.
Overview
Applies asymmetric least-squares baseline smoothing to spectra.
The method is intended to estimate and suppress broad baseline variation while preserving narrower spectral structure.
Smoothness controls the baseline penalty. Larger values favor a smoother baseline with less local variation.
Asymmetry controls the unequal weighting of points above and below the estimated baseline during iterative fitting.
Iterations controls how many times the baseline weights are updated.
More iterations can improve convergence but increase processing time.
Light: lower smoothness and fewer iterations.
Balanced: general-purpose default.
Strong: stronger baseline smoothing.
- Broad spectral baseline variation is present.
- Background curvature interferes with narrower spectral features.
- Baseline correction is needed before additional spectral analysis.
Excessive smoothing can alter scientifically meaningful broad spectral structure or distort weak features close to the baseline.
Workflow
1. Load a hyperspectral dataset.
2. Open Correct > Filtering & Enhancement.
3. Select Asymmetric Least Squares.
4. Choose a strength preset or open Advanced Settings.
5. Adjust Smoothness, Asymmetry, and Iterations if needed.
6. Inspect the corrected spectra.
7. Use Reset Filters if the result is not desired.