standard deviation stretch crash

[magdaw]

Hello jlaura,

I have been "partially" succesful with running a standard deviation stretch on my 8bit 1 band tif imagery. But pystretch crashes on some of them with the following (memory issue) message:
I use 32 bit version of Python, GDAL, numpy, scipy and pystretch.

"Processing on 8 cores.
Read band 1 of 1
Image segmented. Processing segment 1 of 1
Traceback (most recent call last):
File "c:\Python266\Scripts\pystretcher.py", line 185, in
main(options, args)
File "c:\Python266\Scripts\pystretcher.py", line 97, in main
array = numpy.ma.masked_values(array, options['ndv'], copy=False)
File "C:\Python266\lib\site-packages\numpy\ma\core.py", line 2206, in masked_values
condition = umath.less_equal(mabs(xnew - value), atol + rtol * mabs(value))
MemoryError"

All the input 8bit 1 band TIFs have been created in the same software.
However the one that fails, due to memory issue, is a larger matrix (6687,8190) comparing to one that completes OK (3230,8978), although the physical file size is actually smaller (~1.5MB) from the one that completes OK (~3MB).

Do I need to explicitly add the segmenting of an image to the following command?:

c:Python266\python c:\Python266\Scripts\pystretcher.py --std --standarddeviations 2.5 input_image.tif --output output_image.tif.

I tried adding --horizontal and --verical to the above command for splitting the memory (?):

c:Python266\python c:\Python266\Scripts\pystretcher.py --std --standarddeviations 2.5 --horizontal 4 --vertical 4 input_image.tif --output output_image.tif.

but the process fails with an error message:

Traceback (most recent call last):
File "c:\Python266\Scripts\pystretcher.py", line 185, in
main(options, args)
File "c:\Python266\Scripts\pystretcher.py", line 157, in main
del stats,array, jobs, shared_arr.data, p
UnboundLocalError: local variable 'p' referenced before assignment

I would like to avoid having to physically split the raster into smaller rasters if possible.

Your help much appreciated!
Many thanks
Magda