Change values inside array based on logic - Python

Question

I have an array (called 'img') that I want to modify.

img
array([[[244, 244, 244],
    [248, 248, 248],
    [249, 249, 249],

I want to change the values in the array to 0 if they are below 200 and convert to 255 if they are above or equal to 200:

for value in img:
    if value < 200:
         value = 0
    else:
         value = 255

However, I am getting this error:

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

How can I get this code to work?

You are currently iterating through the img variable, which is a two dimensional list. So value is a list as well. — Pieter van den Ham
– Pieter van den Ham, Commented Apr 29, 2016 at 13:29

user2285236 · Accepted Answer · 2016-04-29 13:35:07Z

1

np.where(img<200, 0, 255)

For the example you provided all values are above 200 so it will return 255 all the time but for 245:

np.where(img<245, 0, 255)
Out[4]: 
array([[[  0,   0,   0],
        [255, 255, 255],
        [255, 255, 255]]])

answered Apr 29, 2016 at 13:35

user2285236

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

Comments

user3062459 · Accepted Answer · 2016-04-29 13:36:19Z

0

Simple solution using Numpy where method. Pass Logic as first argument, if True (second argument), if False (third argument).

img = np.where(img<200.0, 0.0, 255.0)

answered Apr 29, 2016 at 13:36

user3062459

1,6318 gold badges29 silver badges41 bronze badges