Compute Coherence (Image Analyst Tools)

Summary

Computes the similarity between the reference and secondary input complex radar data.

The output is a coherence raster with a value range of 0 to 1 in which 0 indicates no coherence and 1 indicates perfect coherence. A value of 0.3 or above is considered a good coherence value.

Usage

The inputs must be a pair of coregistered complex radar datasets.
The output coherence raster can be used for coherent change detection. The output coherence raster can be used in InSAR and DInSAR workflows to assess the reliability of the interferogram pixels.

For InSAR and DInSAR applications, the input radar data must have the same satellite geometry.
In the Geoprocessing pane, the Range Window Size and Azimuth Window Size parameters include information about the estimate size required to create an approximate square pixel for the Input Radar Data parameter value.

Baseline considerations for InSAR applications are shown in the following table:


Baseline type	Definition	Smaller baseline	Larger baseline
Perpendicular baseline	The component of the physical separation between the two satellite acquisition positions that is perpendicular to the line of sight (LOS).	Preserves phase coherence, which is ideal for surface deformation monitoring. Sensitivity to elevation differences is reduced, making it less ideal for DEM generation.	Increases sensitivity to topography, which is ideal for DEM generation.
Temporal baseline	The time separation between the two acquisitions.	Better coherence in dynamic areas and captures rapid changes such as landslides and earthquakes.	Suitable for detecting slow, cumulative deformation such as tectonic motion, although it carries a risk of decorrelation.

Perpendicular and temporal baselines

Label	Explanation	Data type
Input Reference Radar Data	The input reference complex radar data.	Raster Dataset; Raster Layer
Input Secondary Radar Data	The input secondary complex radar data.	Raster Dataset; Raster Layer
Output Radar Data	The output coherence radar data.	Raster Dataset
Polarization Bands (Optional)	The polarization bands that will be corrected. The first band is selected by default.	String
Range Window Size (Optional)	The range window size in pixels. The default value is 10.	Long
Azimuth Window Size (Optional)	The azimuth window size in pixels. The default value is the minimum number of pixels required to create an approximate square window. For example, if the Range Window Size parameter value is 10, the default will be 3.	Long

ComputeCoherence(in_reference_radar_data, in_secondary_radar_data, out_radar_data, {polarization_bands}, {range_window_size}, {azimuth_window_size})

Name	Explanation	Data type
in_reference_radar_data	The input reference complex radar data.	Raster Dataset; Raster Layer
in_secondary_radar_data	The input secondary complex radar data.	Raster Dataset; Raster Layer
out_radar_data	The output coherence radar data.	Raster Dataset
polarization_bands [polarization_bands,...] (Optional)	The polarization bands that will be corrected. The first band is selected by default.	String
range_window_size (Optional)	The range window size in pixels. The default value is 10.	Long
azimuth_window_size (Optional)	The azimuth window size in pixels. The default value is the minimum number of pixels required to create an approximate square window. For example, if the `range_window_size` parameter value is 10, the default will be 3.	Long

Code sample

ComputeCoherence example 1 (Python window)

This example computes the similarity between the reference and secondary SAR images using a range and azimuth window size of 10 pixels.

import arcpy
arcpy.env.workspace = r"C:\Data\SAR"

outRadar = arcpy.ia.ComputeCoherence("Reference_SAR.crf",
    "Secondary_SAR_Coreg.crf", "VV", 10, 10)
outRadar.save("Reference_Secondary_SAR_Coh.crf")

ComputeCoherence example 2 (stand-alone script)

This example computes the similarity between the reference and secondary SAR images using a range and azimuth window size of 10 pixels.

# Import system modules
import arcpy
from arcpy.ia import *

# Set local variables
in_reference_radar_data = r"C:\SAR\Reference_SAR.crf"
in_secondary_radar_data = r"C:\SAR\Secondary_SAR_Coreg.crf"
polarization_bands ="VV"
range_window_size = 10
azimuth_window_size=10

# Run
outRadar = arcpy.ia.ComputeCoherence(in_reference_radar_data,
    in_secondary_radar_data, polarization_bands, range_window_size,
    azimuth_window_size)
outRadar.save(r"C:\SAR\Reference_Secondary_SAR_Coh.crf")

Environments

Cell Alignment, Compression, Current Workspace, NoData, Parallel Processing Factor, Pyramid, Raster Statistics, Resampling Method, Scratch Workspace, Snap Raster, Tile Size

Licensing information