ROI Statistics
Summarizes spectral values by ROI/class.
μ=(1/n)Σxi
s=√[(1/(n−1))Σ(xi−μ)²]
Best used when: checking class means, variability and outliers.
ROI Analysis
Extract and analyze raster values inside Regions of Interest (ROI). It supports descriptive statistics, class separability, spectral plots, n-D exploration, scatter plots, histograms, and correlation analysis.
Remote Sensing CoreEnd-User DocumentationTheory + Formula + Parameters1. What This Feature Does
Extract and analyze raster values inside Regions of Interest (ROI). It supports descriptive statistics, class separability, spectral plots, n-D exploration, scatter plots, histograms, and correlation analysis.
Cloud workflow: choose data → configure the scientific method → run → review the resulting layer/report. No programming is required.
2. Recommended Workflow
Choose ROI vector and raster.
↓
Choose the ROI class/value field and field type.
↓
Select bands and sampling limits.
↓
Choose an analysis method.
↓
Interpret statistics/plots/separability before classification or modelling.
3. Theory, Methods & Equations
ROI Separability — Jeffries–Matusita
Measures statistical separation between class distributions using Bhattacharyya distance.
B=⅛ΔμTΣ̄−1Δμ + ½ln(|Σ̄|/√(|Σ1||Σ2|))
JM=2(1−e−B)Common range: 0 to 2.
Best used when: selecting bands or checking whether classes can be separated.
ROI Separability — Transformed Divergence
Transforms a divergence measure into a bounded separability score that increases as class distributions separate.
TD = 2[1−exp(−D/8)]A common transformed-divergence form.
Best used when: comparing pairwise class separability.
n-D Visualizer
Views samples across multiple spectral dimensions to understand high-dimensional class geometry.
Best used when: class overlap cannot be understood from one or two bands.
Spectral Profile
Plots mean or sample response versus band/wavelength.
Best used when: comparing spectral signatures among ROI classes.
2D Scatter Plot
Plots one predictor against another, colored/grouped by ROI.
Best used when: visualizing pairwise class separation.
Histogram
Shows the distribution of values within selected ROIs/bands.
Best used when: checking skewness, overlap and outliers.
Correlation Analysis
Measures linear or rank association among bands and optionally a numeric ROI field.
r = Σ(x−x̄)(y−ȳ)/√[Σ(x−x̄)²Σ(y−ȳ)²]Pearson correlation.
ρ = corr(rank(x),rank(y))Spearman rank correlation.
Best used when: identifying redundant predictors or relationships with a numeric reference variable.
4. Input Data
| Input | Type | Requirement | Description |
|---|---|---|---|
| ROI Vector | vector | Required | Point/MultiPoint or Polygon/MultiPolygon ROI layer in the owner project vector layer database. |
| Raster Input | raster | Required | Raster source from which pixel values are extracted. |
5. Parameters Available in the Application
| Parameter | Default | Choices / Range | How to Use It |
|---|---|---|---|
| ROI Analysis Method | — | ROI Statistics, ROI Separability, n-D Visualizer, Spectral Profile, 2D Scatter Plot, Histogram, Correlation Analysis | Select the analysis first; method-specific parameters are shown afterward. |
| ROI Field | — | — | Vector attribute used as the ROI value. It may contain class labels (for example Forest, Water, 1, 2, 3) or continuous numeric values (for example biomass, DBH, temperature). |
| ROI Field Type | auto | Auto Detect, Categorical / Class, Numeric / Continuous | Use Categorical when numeric codes represent classes. Use Numeric when the ROI field is a continuous measurement. |
| Select Input Bands | — | — | 1-based raster band numbers. Default is All bands. Select specific bands only when needed. |
| Maximum Samples per ROI | 5000 | 1 to 100000 | Maximum extracted pixels for each polygon ROI. |
| Maximum Total Samples | 50000 | 1 to 500000 | Maximum extracted samples stored in the output table. |
| Include All Touched Pixels | Off | — | For polygon ROI extraction, include all pixels touched by the polygon. |
| Percentiles | [5, 25, 50, 75, 95] | — | Shown for: statistics |
| Separability Metric | both | Jeffries-Matusita + Transformed Divergence, Jeffries-Matusita, Transformed Divergence | Separability requires ROI Field Type = Categorical / Class. Shown for: separability |
| Covariance Regularization | 1e-06 | 0.0 to — | Shown for: separability |
| Maximum Visualization Samples | 5000 | 10 to 50000 | Shown for: nd_visualizer, scatter_plot |
| X Axis Band | 1 | 1 to — | Shown for: scatter_plot |
| Y Axis Band | 2 | 1 to — | Shown for: scatter_plot |
| Histogram Band | 1 | 1 to — | Shown for: histogram |
| Histogram Bins | 20 | 2 to 200 | Shown for: histogram |
| Correlation Method | pearson | Pearson, Spearman | Shown for: correlation |
| Include Numeric ROI Field in Correlation | On | — | When the ROI field is numeric, include it as an additional variable alongside raster bands in the correlation matrix. Shown for: correlation |
6. Output & Interpretation
The result should be interpreted according to the selected method and the physical meaning of the input data. Preserve source units, coordinate reference information, NoData meaning, acquisition date, and preprocessing level when comparing results.
Scientific interpretation: an algorithm can produce a numerically valid result even when the input data are unsuitable. Always check masks, units, sensor characteristics, spatial resolution and reference data.
7. Best Practices
- Use analysis-ready inputs and remove invalid/cloud/noise artifacts that are not part of the target phenomenon.
- Choose parameters from the physical scale of the data, not only from visual appearance.
- Keep categorical and continuous rasters conceptually separate when selecting interpolation/resampling methods.
- Compare the result with the source image and independent reference information.
- Document the settings used when results will be compared across dates, sensors or study areas.