ACCURATE AND ROBUST IMAGE REGISTRATION BASED ON RADIAL BASIS NEURAL NETWORKS

Sarnel H., ŞENOL Y., Sagirlibas D.

23rd International Symposium on Computer and Information Sciences, İstanbul, Türkiye, 27 - 29 Ekim 2008, ss.314-318 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/iscis.2008.4717914
  • Basıldığı Şehir: İstanbul
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.314-318
  • Anahtar Kelimeler: Image registration, affine transformation, radial basis function neural network
  • Dokuz Eylül Üniversitesi Adresli: Evet

Özet

Neural network-based image registration using global image features is relatively a new research subject and the schemes devised so far use a feedforward neural network to find the geometrical transformation parameters. In this work, we propose to use a radial basis function neural network instead of feedforward neural network to overcome lengthy pre-registration training stage. This modification has been tested on a typical neural network-based registration method using discrete cosine transformation features in the presence of noise. The proposed scheme does not only speed up the training stage enormously, but also increases the accuracy and robustness against additive white noise owing to the better generalization ability of the radial basis function neural networks.