Saturday, 15 February 2014

python - Numpy scale down array representing image -


I have an image that is stored in a narrow array ~ 180x130 pixels.

I want to scale it so much that the image is NxN but to keep it that aspect ratio is kept, it means black space around the small dimension.

Is there a function to do this?

You can try a short example:

  a = Np.ones ((4,2)) np.pad (a, ((0,0), (1,1)), 'continuous', constant_values ​​= (0,0)) array ([[0 , 1., 1., 0.], [0., 1., 1., 0.], [0., 1., 1., 0], [0., 1., 1, 0 ]]  

In your case, I think you should have another argument ((0,0), (2525)) your picture array On the Pad with zero in both of the 25 columns. (Of course if you represent the static values ​​for "blank space" for your image format, you need a pad.)


No comments:

Post a Comment