PROProduction

Coregister moving SAR imagery to a reference grid (translation/affine/local-offset-grid modes).

workflow pro

Workflow Narrative

SAR Coregistration

Problem It Solves

Can we put a moving SAR scene on the reference scene grid with enough geometric confidence for pairwise analytics?

Who It Is For

SAR analysts, infrastructure monitoring teams, and registration-first EO pipelines.

Primary User

InSAR/infrastructure monitoring teams, geospatial intelligence groups, and SAR QA operations.

What It Does

Aligns a moving SAR raster or resolved SAR-bundle measurement onto a reference SAR grid.
Supports `translation`, `affine`, and `local_offset_grid` modes (`affine` and `local_offset_grid` remain experimental).
Emits aligned output plus transform/summary diagnostics, including compatibility and acceptance gates for machine-checkable downstream use.

How It Works

Harmonizes the moving SAR raster to the reference CRS/grid when needed and evaluates strict scene compatibility where metadata allows.
Applies amplitude-domain matching with bounded search and subpixel refinement, then executes the selected transform model.
Emits per-run QA including burst diagnostics (Sentinel-1), continuity checks, and Phase-A acceptance gates.
Indicative formula: $\rho(\Delta x, \Delta y) = \mathrm{corr}(\log(1 + I_{ref}), \log(1 + I_{mov}(x-\Delta x, y-\Delta y)))$.

Why It Wins

Accepts both direct rasters and supported SAR bundles while turning pair alignment into a reproducible audited workflow artifact instead of an opaque manual preprocessing step.

Typical Buying Trigger

A team needs a defensible alignment stage before coherence, pair differencing, or cross-sensor fusion.

Typical Presets

translation: default global shift path.
affine: global affine refinement (experimental).
local_offset_grid: local residual warp model (experimental).

Inputs

Parameter	Optional	Description
reference_sar or reference_sar_bundle	no	Reference SAR source provided either as a raster path or a supported SAR bundle.
moving_sar or moving_sar_bundle	no	Moving SAR source provided either as a raster path or a supported SAR bundle.
reference_measurement_key, moving_measurement_key	yes	Optional bundle measurement selectors when either SAR bundle contains multiple measurement assets.
coreg_mode	yes	Coregistration mode: `translation`, `affine`, or `local_offset_grid` (experimental for affine/local).
max_offset_px, decimation, min_overlap_fraction	yes	Search-radius and sampled-overlap controls for correlation-based shift estimation.
input_dem, dem_z_factor	yes	Optional DEM-assisted initialization controls used for geometry-informed matching support.
phase_a_* thresholds	yes	Optional acceptance/continuity threshold overrides for deterministic quality gating.
resample_method, output_prefix	yes	Output resampling mode and artifact prefix.

Outputs

Parameter	Type	Description
moving_aligned	GeoTIFF	Moving SAR raster resampled onto the reference SAR grid.
offset_x	GeoTIFF	Constant x-offset surface in map units for the estimated translation.
offset_y	GeoTIFF	Constant y-offset surface in map units for the estimated translation.
transform	JSON	Machine-readable transform and QA summary for the estimated alignment.
summary	JSON	Machine-readable workflow report containing parameters, QA diagnostics, and artifact paths.
html_report	HTML	Human-readable report generated from the summary contract for review and traceability.

Python Example

import whitebox_workflows as wbw

wbe = wbw.WbEnvironment(include_pro=True, tier="pro")

result = wbe.sar_coregistration(
    reference_sar_bundle="data/S1_reference.SAFE",
    reference_measurement_key="vv",
    moving_sar_bundle="data/S1_pair.SAFE",
    moving_measurement_key="vv",
    max_offset_px=24,
    decimation=4,
    resample_method="bilinear",
    output_prefix="output/sar_coreg",
)

print(result)

License Notice

Use of this function requires a license for Whitebox Workflows Professional (WbW-Pro). Please visit www.whiteboxgeo.com to purchase a license.

Project Links

WbW Homepage User Manual Learn More