2
$\begingroup$

I am new to image classification and hope to set up a model which will classify large images (I am using R keras). Each image will represent a 10m by 10m square with pixels representing 1 cm. I need the fine resolution to capture the detail in the images. I can follow and run the keras examples for "regular" images with less pixels, like the MNIST data or images of cars or fruits. However, I don't have my real data yet so I am unsure what will happen if I attempt to use images with 1M pixels.

My question is:

  • Is it possible to use images this large?
  • If so, are there any special considerations I need to take into account or special processing that would be required?
  • Where is a good place to start?
|cite|improve this question
$\endgroup$
  • $\begingroup$ Questions that are only about software (e.g. error messages, code or packages, etc.) are generally off topic here. If you have a substantive machine learning or statistical question, please edit to clarify. $\endgroup$ – gung♦ Feb 27 at 21:05
  • 2
    $\begingroup$ I have removed a reference to R, to make it, what I believe, on-topic (although it is still a bit broad but @DeltaIV seems to know a way to answer it). I wanted to also add tags like data processing, computation performance, scaling/scalability, but they do not exist yet here. Questions about this are also not a lot seen here. I imagine that this question might fit well/better/also at datascience.stackexchange $\endgroup$ – Martijn Weterings Mar 2 at 9:11
  • $\begingroup$ 1M pixels is not that many, but your accuracy will probably be ok using much smaller images $\endgroup$ – qwr Mar 2 at 21:14
  • $\begingroup$ you need to be more specific about your classification task $\endgroup$ – qwr Mar 2 at 21:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.