MATLAB/Python array difference?

Question

I got the following:

redIdx: a 2x1 matrix with values (289485, 289486).
image: 366x791x3 uint8 matrix (an image).
image2: zeros matrix with the same sape as image.

In MATLAB, if I do image2(redIdx) it returns a 2x1 matrix with values (0,0) and if I do image(redIdx) it returns a 2x1 matrix with values (94, 83).

But in Python, if I do image2[redIdx] or image[redIdx], it returns the next error: index 2879485 is out of bounds for axis 0 with size 366.

How can I get the same result as MATLAB?

It seems like you want linear indexing in Python. See if this helps — Luis Mendo
– Luis Mendo, Commented Sep 7, 2021 at 10:46

Cris Luengo · Accepted Answer · 2021-09-07 18:31:20Z

1

MATLAB, when indexing an array with a single index (as opposed to multiple ones) uses linear indexing. Python, in the same situation, uses the index to index into the first dimension, returning a slice. The fact that redIdx contains multiple values is irrelevant, it's a 1D indexing operation.

To replicate linear indexing in Python, you can flatten the array, then index:

image.flatten('K')[redIdx]

This Q&A shows how to compute indices from the single linear index, which would be a more complex alternative to the above.

answered Sep 7, 2021 at 18:31

Cris Luengo

61.4k10 gold badges75 silver badges135 bronze badges

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

Collectives™ on Stack Overflow

MATLAB/Python array difference?

1 Answer 1

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Linked

Related