How do I index a numpy array of zeroes with a boolean datatype to True?

Question

So I'm recreating a Matlab project they made last year, part of which involves creating mask that pull out the RGB bands. They did this by an array of logical zeroes.

GMask_Whole = false(ROWS,COLS);

which I reconstructed as a numpy array.

self.green_mask_whole=np.zeros((self.rows, self.columns), dtype=bool)

The next part I can't for the life of me figure out how to do with numpy:

GMask_Whole(1:2:end,2:2:end) = true;

I've yet to find a numpy equivalent action. any Ideas?

btw, if your curious about what this is doing: https://en.wikipedia.org/wiki/Bayer_filter

edit: things I've tried:

wut(1:3:end, 1:2:end) = true
wut([1:3:end], [1:2:end]) = true
wut([1:3], [1:2]) = true
wut([1:3], [1:2]) = True
wut(slice(1:3), slice(1:2)) = True

editing my original post.

Spider Jerusalem
– Spider Jerusalem

2017-06-19 18:48:50 +00:00
Commented Jun 19, 2017 at 18:48 — Spider Jerusalem
– Spider Jerusalem, Commented Jun 19, 2017 at 18:48

JARS · Accepted Answer · 2017-06-20 18:46:09Z

1

You can translate Matlab's

GMask_Whole(1:2:end,2:2:end) = true;

to python by

green_mask_whole[::2,1::2] = True

(assuming green_mask_whole is a numpy array)

edited Jun 20, 2017 at 18:46

answered Jun 20, 2017 at 12:46

JARS

1,1297 silver badges10 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

JohanL · Accepted Answer · 2017-06-21 21:28:37Z

1

numpy can do slicing more or less as in Matlab, but the synax is a little bit different. In numpy, the order is [begin:end:step] and it is possible to leave both begin, end and step empty, which will give them their default values first element, last element and step size 1 respectively.

Further, `numpy´ has a nice system of 'broad casting' which allows a single value (or row/column) be repeated to make a new array of the same size as another. This makes it possible to assign a single value to a whole array.

Thus, in the current case, it is possible to do

self.green_mask_whole=np.zeros((self.rows, self.columns), dtype=bool)
self.green_mask_whole[::2,1::2] = True

answered Jun 21, 2017 at 21:28

JohanL

6,9211 gold badge16 silver badges32 bronze badges

Collectives™ on Stack Overflow

How do I index a numpy array of zeroes with a boolean datatype to True?

2 Answers 2

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related