3d non-separable moment invariants and their use in neural networks

HIGHLIGHTS

• What: The authors show that non-separable moments may outperform the separable ones in terms of recognition power and robustness thanks to a better distribution of their zero surfaces over the image space. The authors show the robustness to resampling errors improved more than twice and the recognition rate increased by 2-10 % comparing to most common descriptors. In the last section the authors show how these invariants can be used in state-of-the-art neural networks for image recognition. The authors generalize their idea into 3D and the authors demonstrate their usage as standalone descriptors . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.