When making a cnn you could use the classic mnist dataset containing grayscale images. I am considering transforming them to simple binary images instead, the questions is should i? It will be much smaller file sizes and easier to determine the text from background. Why is the mnist dataset grayscaled and not binarized? Will it have any postive or negative effects when applying the binarized images in a cnn?
MNIST is not an interesting data set. You use MNIST to learn how to do machine learning that you will you on interesting data sets. Thus...
If you just want to figure out how to do the Keras code for an image classification problem, feel free to use 0/1 encoding of white vs not-white spaces in the images. It might lower the computational demands so you can learn faster and not wait around while your computer runs.
This may be ineffective in work on interesting data sets. For example, if the pencil just nicks a pixel, that slightly dark space does not count as dark to nearly the extent that a truly black pixel counts, yet your feature extraction regards them as the same. This could confuse your model when you use more interesting data.