0

I want to use Stacked Autoencoders for colour image classification. The example given on matlab site for image classification of MNIST dataset is only for black and white images which has only one colour channel. But for colour images, it has 3 colour channels, RGB. So what pre processing should i do to the colour images since colour images are matrix in 3 dimensions, for the stacked autoencoders to work. I just need the concept, not specific detailed answer.

URL for matlab example - https://in.mathworks.com/help/nnet/examples/training-a-deep-neural-network-for-digit-classification.html

user11622
  • 39
  • 1
  • 3
  • For converting RGB to grayscale, you can use [rgb2gray](https://www.mathworks.com/help/matlab/ref/rgb2gray.html) function. – Rotem Mar 21 '17 at 09:29
  • But i require to do it in RGB channel because there is little colour difference between the image to be detected and background. – user11622 Mar 21 '17 at 09:51
  • According to Matlab documentation: [trainautoencoder](https://www.mathworks.com/help/nnet/ref/trainautoencoder.html), **"The image data can be RGB data, in which case, each cell contains an m-by-n-3 matrix."** – Rotem Mar 21 '17 at 10:14

0 Answers0