1

I am new to image processing. I have hand scan images and they are huge in size (~14000x12000 pixels). I need to reduce image size without losing much details (like patterns and ridges present on the fingers and palm). My aim is to use those for classification using neural network. Currently I am getting out of memory error, when ever i tried to run the codes. I used thumb prints and it ran well because those are smaller in size. Please suggest algorithms or methods in which this problem can be solved. Thank you in advance.

Rimjhim Agrawal
  • 69
  • 1
  • 10
  • 2
    Principal component analysis is probably your best option. – nick Nov 07 '18 at 10:04
  • 1
    See [What is the best image downscaling algorithm (quality-wise)?](https://stackoverflow.com/q/384991) and the [Image scaling on Wikipedia](https://en.wikipedia.org/wiki/Image_scaling) for a discussion about different existing algorithms. – jdehesa Nov 07 '18 at 10:09

0 Answers0