I was curious as to if anyone has used the RPi3 for real-time image processing? If so, was there any computational delay while doing it real-time?

I am planning to use an external GPU to train a CNN that can classify weeds and crops. I want to make sure the RPi3 can handle image processing features like segmentation and be able to classify the weed and crops in real-time using pixel-wise classification(might require a CNN).


It depends on several factors:

  • Required accuracy of the classifier
  • Size of the input images
  • Number of image classes
  • Real-time requirements (how much time can the classifier take to do the job)

Given that, I managed to run a very simple classifier based on TensorFlow and a custom MobileNet model retrained in order to recognize two classes of images: apples from kiwis, given images of 160x120 pixels; you can see a demo in this video to get an idea:

Apple or Kiwi?


It depends on the size of each image and the structure of the CNN itself. However, I think that the RPi isn't powerful enough to get decent results in reasonable time.

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.