despeck.py usage
stoormgeo opened this issue · 4 comments
What is the best way to use despeck.py?
It gives an error then on startup. The same window sizes used.
despeck.py -i /data/merged/fringe/coreg_stack/slcs_base.vrt -o /data/merged/fringe/Despeck/despeck -w /data/merged/fringe/KS2/count -l 100 -r 100000 -x 25 -y 35 -b 1
Input Dataset: /data/insar/Proc/Specgeo/Marre/proc/merged/fringe/coreg_stack/slcs_base.vrt
Weights Dataset: /data/insar/Proc/Specgeo/Marre/proc/merged/fringe/KS2/count
Output Dataset: /data/insar/Proc/Specgeo/Marre/proc/merged/fringe/Despeck/despeck
Window size: 25 35
Memsize: 100000 Mb
Blocksize: 100 lines
Master band: 2
Number of uint32 bytes for mask: 114
Number of rows = 4840
Number of cols = 13909
Number of bands = 17
Number of bands mismatch for weights and window size
Exiting with error code ....(108)
The wts/nmap.hdr file should report what window sizes were used to generate it in the first place. You are providing it the count file, when it needs the wts/nmap file as input. This looks like a mismatch in x and y arguments. And you are requesting 100GB of RAM with the memory option.
The wts/nmap.hdr file should report what window sizes were used to generate it in the first place. You are providing it the count file, when it needs the wts/nmap file as input. This looks like a mismatch in x and y arguments. And you are requesting 100GB of RAM with the memory option.
Thanks Piyushrpt,
Recalculated with a smaller window size without any errors. The output is a file with the .hdr
How can I use further with evd.py/sequential.py?
Thx
Those already use the despeck operation internally for all computations. This is just a utility to despeck a SAR image / individual interferogram using the neighborhood map - for visualization or other applications that build on it.
Those already use the despeck operation internally for all computations. This is just a utility to despeck a SAR image / individual interferogram using the neighborhood map - for visualization or other applications that build on it.
Ok. Thanks for guidance!