PostGIS
Mobile
QGIS
MapBender
GeoServer
GeoNode
GeoNetwork
Novella
Solutions
64 lines
1.4 KiB
Markdown
64 lines
1.4 KiB
Markdown
# Kernel Density Estimation (KDE)
|
|
|
|
Generate density surfaces from point data using kernel density estimation.
|
|
|
|
## Overview
|
|
|
|
KDE creates a continuous density surface from point data, showing where points are concentrated. Higher values indicate greater point density.
|
|
|
|
## Inputs
|
|
|
|
- **Dataset**: Point dataset
|
|
- **Bandwidth**: Smoothing parameter (default: auto-calculated)
|
|
- **Cell Size**: Output raster cell size (default: auto-calculated)
|
|
- **Weight Field** (optional): Field to weight points
|
|
|
|
## Outputs
|
|
|
|
Raster dataset containing:
|
|
|
|
- Density values for each cell
|
|
- Higher values indicate greater point density
|
|
- Proper spatial reference
|
|
|
|
## Algorithm
|
|
|
|
1. Calculate optimal bandwidth (if not specified)
|
|
2. Create output raster grid
|
|
3. For each cell, calculate kernel-weighted sum of nearby points
|
|
4. Store density values in raster
|
|
|
|
## Example
|
|
|
|
```json
|
|
{
|
|
"dataset_id": 123,
|
|
"bandwidth": 1000,
|
|
"cell_size": 100
|
|
}
|
|
```
|
|
|
|
## Background Jobs
|
|
|
|
This analysis runs as a background job.
|
|
|
|
## Use Cases
|
|
|
|
- Population density mapping
|
|
- Crime hotspot visualization
|
|
- Species distribution modeling
|
|
- Event density analysis
|
|
|
|
## Notes
|
|
|
|
- Bandwidth controls smoothing (larger = smoother)
|
|
- Cell size controls output resolution
|
|
- Weight field allows importance weighting
|
|
- Results are sensitive to bandwidth selection
|
|
|
|
## Related Documentation
|
|
|
|
- [Analysis API](../api/analysis.md)
|
|
- [Raster Tools](raster.md)
|
|
|
