Image Segmentation by Image Foresting Transform with Boundary Polarity and Shape Constraints

  • Lucy Mansilla USP
  • Paulo Miranda USP

Resumo


Image segmentation, such as to extract an object from a background, is very useful for medical and biological image analysis. In this work, we propose new segmentation methods for interactive segmentation of multidimensional images, based on the Image Foresting Transform (IFT), by exploiting for the first time non-smooth connectivity functions (NSCF) with a strong theoretical background. The new algorithms provide global optimum solutions according to an energy function of graph cut, subject to high-level boundary constraints (polarity and shape). Our experimental results indicate substantial improvements in accuracy in relation to other state-of-the-art methods, using medical images by allowing the customization of the segmentation to a given target object.

Referências

Bejar, H. and Miranda, P. (2014). Oriented relative fuzzy connectedness: Theory, algorithms, and applications in image segmentation. In XXVII Conference on Graphics, Patterns and Images (SIBGRAPI), pages 304–311, Rio de Janeiro, Brazil. IEEE.

Braz, C. and Miranda, P. (2014). Image segmentation by image foresting transform with geodesic band constraints. In IEEE International Conference on Image Processing (ICIP), pages 4333 – 4337, Paris, France.

Falc˜ao, A., Stolfi, J., and Lotufo, R. (2004). The image foresting transform: Theory, algorithms, and applications. IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(1):19–29.

Gulshan, V., Rother, C., Criminisi, A., Blake, A., and Zisserman, A. (2010). Geodesic star convexity for interactive image segmentation. In Proceedings of Computer Vision and Pattern Recognition, pages 3129–3136.

Mansilla, L. (2014). Image foresting transform with non-smooth connectivity functions: Adaptive weights, boundary polarity, and shape constraints. Master’s thesis, Institute of Mathematics and Statistics, University of S˜ao Paulo, Brazil. http://www.teses.usp.br/teses/disponiveis/45/45134/tde-17032014-121734/pt-br.php.

Mansilla, L., Cappabianco, F., and Miranda, P. (2013a). Image segmentation by image foresting transform with non-smooth connectivity functions. In XXVI Conference on Graphics, Patterns and Images (SIBGRAPI), pages 147–154, Arequipa, Perú. IEEE.

Mansilla, L., Jackowski, M., and Miranda, P. (2013b). Image foresting transform with geodesic star convexity for interactive image segmentation. In IEEE International Conference on Image Processing (ICIP), pages 4054–4058, Melbourne, Australia.

Mansilla, L. and Miranda, P. (2013a). Image segmentation by oriented image foresting transform: Handling ties and colored images. In 18th International Conference on Digital Signal Processing (DSP), pages 1–6, Santorini, Greece. IEEE.

Mansilla, L. and Miranda, P. (2013b). Image segmentation by oriented image foresting transform with geodesic star convexity. In Computer Analysis of Images and Patterns (CAIP), volume 8047, pages 572–579, York, UK.

Miranda, P. and Mansilla, L. (2014). Oriented image foresting transform segmentation by seed competition. IEEE Transactions on Image Processing, 23(1):389–398.

Veksler, O. (2008). Star shape prior for graph-cut image segmentation. In European Conference on Computer Vision (ECCV), volume 5304, pages 454–467.
Publicado
20/07/2015
MANSILLA, Lucy; MIRANDA, Paulo. Image Segmentation by Image Foresting Transform with Boundary Polarity and Shape Constraints. In: CONCURSO DE TESES E DISSERTAÇÕES (CTD), 28. , 2015, Recife. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2015 . p. 61-66. ISSN 2763-8820. DOI: https://doi.org/10.5753/ctd.2015.10003.