CODEWITHSUNDEEP
pythonnumpy
shadow dk
shadow dk

Reputation: 503

How to separate image data into x, y coordinate? (python)

I have a image data, How can I separate it into x, y and the value?

image data

[[38  0  0 ...  0  0  0]
 [46  3  0 ...  0  0  0]
 [46  3  0 ...  0  0  0]
 ...
 [74  0  0 ...  0  0  0]
 [74  0  0 ...  0  0  0]
 [74  0  0 ...  0  0  0]]

and their should it be?

x = 0, y = 0, value = 38
x = 0, y = 1, value = 46
...

How can I separate it in to :

x = []
y = []
value = []

only for loop method is work?

Thanks for any help

Upvotes: 2

Views: 2384

Answers (2)

sacuL
sacuL

Reputation: 51395

IIUC, I think you can use np.indices. Take your example:

>>> img
array([[38,  0,  0,  0,  0,  0],
       [46,  3,  0,  0,  0,  0],
       [46,  3,  0,  0,  0,  0],
       [74,  0,  0,  0,  0,  0],
       [74,  0,  0,  0,  0,  0],
       [74,  0,  0,  0,  0,  0]])

value = img.flatten()
y,x = np.indices(img.shape).reshape(-1,len(value))

>>> x
array([0, 1, 2, 3, 4, 5, 0, 1, 2, 3, 4, 5, 0, 1, 2, 3, 4, 5, 0, 1, 2, 3,
       4, 5, 0, 1, 2, 3, 4, 5, 0, 1, 2, 3, 4, 5])
>>> y
array([0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3,
       3, 3, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5])
>>> value
array([38,  0,  0,  0,  0,  0, 46,  3,  0,  0,  0,  0, 46,  3,  0,  0,  0,
        0, 74,  0,  0,  0,  0,  0, 74,  0,  0,  0,  0,  0, 74,  0,  0,  0,
        0,  0])

So where x is 0 and y is 0, you get value 38, where x is 0 and y is 1, you get value 46, and so on.

Edit: In your comment, you said you want to filter out the zeros. You can do this with np.where and np.nonzero:

y,x = np.where(img)

value = img[np.nonzero(img)]

>>> y
array([0, 1, 1, 2, 2, 3, 4, 5])
>>> x
array([0, 0, 1, 0, 1, 0, 0, 0])
>>> value
array([38, 46,  3, 46,  3, 74, 74, 74])

Upvotes: 4

Thanh Nguyen
Thanh Nguyen

Reputation: 912

if your image is colored, it's often in RGB format(3 channels), if it's grayscale, it will have 1 channel. So the shape of the array will be (img_height, img_width, number_of_channels)

By understanding the shape, you can properly use the imread from PIL or imread from matplotlib to load the image, and then turn them to array using myarray = numpy.array(your_loaded_img). As it's a numpy array, you can call the cell values by myarray[x,y]

Upvotes: 0

Related Questions