Preprocessing of Sentinel-1 SAR data via Snappy Python module

Preprocessing of Sentinel-1 SAR data via Snappy Python module

m

August 1, 2016

This chapter demonstrates the Snappy Python module for the automatization of the ESA SNAP tool.

Code examples will be shown for an automated processing chain for the preprocessing of Sentinel-1 SAR data including Calibration, Subsetting and Terrain Correction of GRD (Ground Range Detected data).

A detailed installation tutorial for snappy can be found here: https://senbox.atlassian.net/wiki/display/SNAP/How+to+use+the+SNAP+API+from+Python

First, import the needed Python modules:

 

import snappy

from snappy import ProductIO
from snappy import HashMap

import os, gc   
from snappy import GPF

GPF.getDefaultInstance().getOperatorSpiRegistry().loadOperatorSpis()
HashMap = snappy.jpy.get_type('java.util.HashMap')

Now loop through all Sentinel-1 data sub folders that are located within a super folder (of course, make sure, that the data is already unzipped):

path = "D:\\SENTINEL\\"
 for folder in os.listdir(path):

   gc.enable()
   
   output = path + folder + "\\"  
   timestamp = folder.split("_")[4] 
   date = timestamp[:8]

Then, read in the Sentinel-1 data product:

   sentinel_1 = ProductIO.readProduct(output + "\\manifest.safe")    
   print sentinel_1

If polarization bands are available, spolit up your code to process VH and VV intensity data separately. The first step is the calibration procedure by transforming the DN values to Sigma Naught respectively. You can specify the parameters to output the Image in Decibels as well.

   pols = ['VH','VV'] 
   for p in pols:  
      polarization = p    
    
      ### CALIBRATION
  
      parameters = HashMap() 
      parameters.put('outputSigmaBand', True) 
      parameters.put('sourceBands', 'Intensity_' + polarization) 
      parameters.put('selectedPolarisations', polarization) 
      parameters.put('outputImageScaleInDb', False)  

      calib = output + date + "_calibrate_" + polarization 
      target_0 = GPF.createProduct("Calibration", parameters, sentinel_1) 
      ProductIO.writeProduct(target_0, calib, 'BEAM-DIMAP')

Next, specify a subset AOI to reduce the data amount and processing time. The AOI specified by its outer polygon corners and is formatted through a Well Known Text (WKT).

      ### SUBSET

      calibration = ProductIO.readProduct(calib + ".dim")    
      WKTReader = snappy.jpy.get_type('com.vividsolutions.jts.io.WKTReader')

      wkt = "POLYGON((12.76221 53.70951, 12.72085 54.07433, 13.58674 54.07981, 
                      13.59605 53.70875, 12.76221 53.70951))"

      geom = WKTReader().read(wkt)

      parameters = HashMap()
      parameters.put('geoRegion', geom)
      parameters.put('outputImageScaleInDb', False)

      subset = output + date + "_subset_" + polarization
      target_1 = GPF.createProduct("Subset", parameters, calibration)
      ProductIO.writeProduct(target_1, subset, 'BEAM-DIMAP')

Apply a Range Doppler Terrain Correction to correct for layover and foreshortening effects, by using the SRTM 3 arcsecond product (90m) that is downloaded automatically. You could also specify an own DEM product with a higher spatial resolution from a local path:

      ### TERRAIN CORRECTION
 
      parameters = HashMap()     
      parameters.put('demResamplingMethod', 'NEAREST_NEIGHBOUR') 
      parameters.put('imgResamplingMethod', 'NEAREST_NEIGHBOUR') 
      parameters.put('demName', 'SRTM 3Sec') 
      parameters.put('pixelSpacingInMeter', 10.0) 
      parameters.put('sourceBands', 'Sigma0_' + polarization)
 
      terrain = output + date + "_corrected_" + polarization 
      target_2 = GPF.createProduct("Terrain-Correction", parameters, subset) 
      ProductIO.writeProduct(target_2, terrain, 'GeoTIFF')

Fergana_Sentinel

you may also like:

Department of Remote Sensing awarded for insect-friendly cooperation

Department of Remote Sensing awarded for insect-friendly cooperation

For German version see below Together with other collaborative partners, the Department of Remote Sensing at the University of Würzburg was awarded for "Insect-friendly Cooperation" by the Federal Ministry of Food and Agriculture (BMEL) in the federal competition for...

UAV Lidar data Acquisition in the Alps

UAV Lidar data Acquisition in the Alps

  Lidar data on board one of our UAVs was acquired in the last field campaign early July for mapping forest structure in an alpine valley. The data delivered very good and interesting vertical information about the vegetation structure and the terrain of this...

most recent news:

Department of Remote Sensing awarded for insect-friendly cooperation

Department of Remote Sensing awarded for insect-friendly cooperation

For German version see below Together with other collaborative partners, the Department of Remote Sensing at the University of Würzburg was awarded for "Insect-friendly Cooperation" by the Federal Ministry of Food and Agriculture (BMEL) in the federal competition for...

UAV Lidar data Acquisition in the Alps

UAV Lidar data Acquisition in the Alps

  Lidar data on board one of our UAVs was acquired in the last field campaign early July for mapping forest structure in an alpine valley. The data delivered very good and interesting vertical information about the vegetation structure and the terrain of this...