I'm struggling to wrap my head around how I can use the output of a correlation matrix to determine if my image is present.

If I am trying to match characters, how can I compare the outputs of these correlation matrix's to determine the best match?

Lets say I have the MNIST database:

if I compare one of the test digits to my labeled training set using correlate2d, how do I know which matrix is the best choice to label my test digit?

for i,j in enumerate(training_digits):

  target_digit = test_digits[1] 
  template = training_digits[i]
  corr = signal.correlate2d(target_digit, template, boundary='symm', mode='full')

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.