Diatoms classification using frequency domain techniques

• Carlos Grima ^[1] ; Fernando Tadeo ^[1] ; Mª Teresa Álvarez ^[1] ; Jose Luis Arribas ^[2]
  1. [1] Universidad de Valladolid

     Universidad de Valladolid

     Valladolid, España

     
  2. [2] Empresa Depuradora de Aguas Residuales, Ávila
• Localización: XXIV Jornadas de Automática: León, 10, 11 y 12 de septiembre de 2003, Universidad de León, E.I. Industrial e Informática, 2003, ISBN 84-931846-7-5
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • This paper presents an approach to the automaticclassification of diatoms based on image-processingtechniques. The main contribution is the development ofa pattern evaluated from a frequential analysis ofimages taken through microscope, post-processed toextract the relevant characteristics that distinguish thedifferent species. The final classification is carried outusing a competitive neural network, with promisingresults