Conversion to multi-look intensity (.mli)
The ALOS data was supplied as amplitude and had to be first converted to multi-look intensity (.mli). This was performed using GAMMA software which can be is driven using the script mli_script given here , This asks for a "fileroot" meaning the convenient name that you want to give the resulting file rather than the original ALOS label. I used a convention of using a fileroot based on the path, frame and cycle number of the file e.g. P443_F7170_C20. The code given in mli_script reads the long ALOS filenames using a wildcard and renames the result as fileroot.mli and fileroot.hdr. Versions for single look complex (SLC) data are also available here.
Co-Registration
Stacks of up to 12 ScanSAR images were used in our work and before analysis these had to be co-registered using one image (e.g. the earliest) as a reference. This procedure was carried our semi-automatically using GAMMA scripts. Describes the co-registration of two images of the same scene taken at different times Unix/Linux scripts The same procedures were used to co-register earlier and later FBD images before applying techniques to identify change.
Multi-Channel filtering
The use of multiple images enables reduction of speckle using the Multi-Channel Filtering technique described in
Quegan, S., & Yu, J.J. (2001). Filtering of Multichannel SAR images. IEEE Transactions on Geoscience and Remote Sensing, 39, 2373-2379
This was implemented using IDL codes running on IDL version 6.4.1 decribed in the document MCF processing and given the extension .mcf.
ScanSAR-FBD combination
This involved using Envi to co-register files. Although not strictly pre-processing it is processing before combination and a non-matlab procedure, so wins a place here.
Reading envi files to MATLAB
This is covered under Matlab Processing but codes are also given here.