Earlier today I reported an issue in which SNAP coregistration of SAR images quit working when the initial offset information comes from georegistration. It was working a couple of months ago and now it does not. It occurred to me that there is another issue that I have not yet reported.
If you use SNAP to do coregistration of a stack of images, and there are more than 2 images (more than 1 slave), then the final interpolated registered slave image in the stack has no content — it is all zeros.
If you have one master and one slave, it appears to work OK.
I ran into a problem using SNAP for coregistration of PALSAR-2 images (Radar -> Coregistration). It worked fine a couple of months ago, and I’m wondering if something changed in a recent SNAP update.
In setting up the parameters for doing the coregistration, I tell it to use geolocation information rather than orbit information. However, when I click “Run,” it says that there is no geolocation information in the file, but the file is a geoTIFF file.
Again, this worked just a couple of months ago, so I’m not sure what has changed. In fact, I tried running the coregistration on the exact same images that I ran successfully a couple of months ago, and this time it failed, so it appears that something changed in SNAP.
Coregistration is based on a random initial distribution of GCPs which are then tested for correlation throughout all images. That means if it worked one time there is a chance that it fails another time by simple chance. You can increase the number of GCPs and run again. If this doesn’t work, you can also test DEM-based coregistration.
Did you just open the tif file or did you import the ALOS product via File > Import > SAR Sensors?
The latter is required for SNAP to read all metadata correctly.
It does NOT show the track. However, I would not expect the file to have track information since the files are GeoTIFF and not CEOS.
I have noticed that the dates are NOT correct.
I’m still a novice SNAP user. I’m open to any suggestions about loading metadata. I assumed that loading the files individually into the coregistration tool was sufficient, but apparently it’s not.
I downloaded the data from EastView Geospatial, a company in Minneapolis Minnesota, USA. They are a distributor of PALSAR-2 data through an agreement with JAXA. They do not do the processing themselves. They have it processed by whoever does the processing for JAXA.