Functional Networks for Image Segmentation of Cutaneous Lesions with Rational Curves

• Akemi Gálvez ^{[1] [2]} ; Iztok Fister ^{[2] [3]} ; Iztok Fister Jr. ^[3] ; Andrés Iglesias ^{[1] [2]}
  1. [1] Toho University

     Toho University

     Japón

     
  2. [2] Universidad de Cantabria

     Universidad de Cantabria

     Santander, España

     
  3. [3] University of Maribor

     University of Maribor

     Eslovenia

     

  Mostrar afiliaciones +
• Localización: 15th International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO 2020): Burgos, Spain ; September 2020 / coord. por Álvaro Herrero Cosío , Carlos Cambra Baseca , Daniel Urda Muñoz , Javier Sedano Franco , Héctor Quintián Pardo , Emilio Santiago Corchado Rodríguez , 2021, ISBN 978-3-030-57802-2, págs. 780-789
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • This paper considers the problem of image segmentation for medical images, in particular, cutaneous lesions. Given a digital image of a skin lesion, our goal is to compute the border curve separating the lesion from the image background. This problem can be formulated as an optimization problem, where the border curve is computed through data fitting from a set of points lying on the lesion boundary. Some recent papers have applied artificial intelligence techniques to tackle this issue. However, they usually focus on the polynomial case, ignoring the more powerful (but also more difficult) case of rational curves. In this paper, we address this problem with rational Bézier curves by applying functional networks, a powerful extension of the classical neural networks. Experimental results on some benchmark medical images show that this method performs well and can be successfully applied to this problem.