Using a string to define Numpy array slice

Question

I have an image array that has an X times Y shape of 2048x2088. The x-axis has two 20 pixel regions, one at the start and one at the end, which are used to calibrate the main image area. To access these regions I can slice the array like so:

prescan_area = img[:, :20]
data_area = img[:, 20:2068]
overscan_area = img[:, 2068:]

My question is how to define these areas in a configuration file in order the generalise this slice for other cameras which may have different prescan and overscan areas and therefore require a different slice.

Ideally, something like the strings below would allow a simple representation in the camera specific configuration file, but I am not sure how to translate these strings into array slices.

prescan_area_def = "[:, :20]"
image_area_def = "[:, 20:2068]"
overscan_area_def = "[:, 2068:]"

Maybe there is something obvious that I am missing?

Thanks!

Answer 1

You can parse the string and use slice. The following generator expression within tuple will create the slice objects for you:

tuple(slice(*(int(i) if i else None for i in part.strip().split(':'))) for part in prescan_area_def.strip('[]').split(','))

Demo:

In [5]: import numpy as np

In [6]: 

In [6]: a = np.arange(20).reshape(4, 5)

In [7]: a
Out[7]: 
array([[ 0,  1,  2,  3,  4],
       [ 5,  6,  7,  8,  9],
       [10, 11, 12, 13, 14],
       [15, 16, 17, 18, 19]])

In [8]: 

In [8]: prescan_area_def = "[:, :3]"

In [9]: a[:, :3]
Out[9]: 
array([[ 0,  1,  2],
       [ 5,  6,  7],
       [10, 11, 12],
       [15, 16, 17]])

In [10]: indices = tuple(slice(*(int(i) if i else None for i in part.strip().split(':'))) for part in prescan_area_def.strip('[]').split(','))

In [11]: indices
Out[11]: (slice(None, None, None), slice(None, 3, None))

In [12]: a[indices]
Out[12]: 
array([[ 0,  1,  2],
       [ 5,  6,  7],
       [10, 11, 12],
       [15, 16, 17]])