T. Michaux

We don’t have enough information about this author to calculate their statistics. If you think this is an error let us know.
Learn More
A new model which permits visual patterns to be invariant to affine transforms (translations, rotations, and dimensions) is presented. A training multilayer fully connected network of ADALINE neurons is proposed as a preprocessing step for invariant image extraction. A second neural network has been trained by the popular backpropagation algorithm for(More)
  • 1