Numpy arrays as image, image channels

Question

I have a 3D Image represented in numpy array format. Shape is (60, 60, 15) , 60 width 60 height 15 frames deep.

Now I want to process this image further with another library. But this library requires an argument "channels". https://keras.io/layers/convolutional/#conv3d

What I don't get is how to find out how many channels my image has. Images are 3D MRI images: 1 slice:

PatriceG · Accepted Answer · 2017-10-10 08:41:33Z

1

The channel represent usually the different colours used to code the image. For instance, in a RGB image, there is three channels: red, green and blue.

I guess your 3D image is coded with 1 channel (if each frame is code in Black & White) or 3 channels (if each frame is coded in colours).

Examples from the doc:

input_shape=(128, 128, 3) # for RGB pictures
input_shape=(128, 128, 128, 1) #  for 128x128x128 volumes with a single channel

edited Oct 10, 2017 at 8:41

answered Oct 10, 2017 at 8:25

PatriceG

4,1215 gold badges31 silver badges43 bronze badges

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

Collectives™ on Stack Overflow

Numpy arrays as image, image channels

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related