Detail publikace

Image Reconstuction in Electrical Impedance Tomography with Total Variation and Gauss-Newton Algorithm

KOUAKOUO NOMVUSSI, S. MIKULKA, J.

Originální název

Image Reconstuction in Electrical Impedance Tomography with Total Variation and Gauss-Newton Algorithm

Typ

článek ve sborníku mimo WoS a Scopus

Jazyk

angličtina

Originální abstrakt

In this paper, the Total Variation and Gauss-Newton algorithms are used to reconstruct the conductivity maps in Electrical Impedance Tomography. The Total Variation gives better results than the Gauss-Newton algorithm. The quality of the reconstructed conductivity maps is clear and almost identical to the image of the forward problem when using the Total Variation method on the whole dataset. In contrast, the Gauss-Newton algorithm provides a smooth image of the object without a perfect reconstruction of its edges. The correlation coefficient of the Total Variation is higher than that of the Gauss-Newton method. The drawback of these two algorithms is their inability to reconstruct the image from noisy data.

Klíčová slova

Total Variation; Gauss Newton; EIT; EIDORS

Autoři

KOUAKOUO NOMVUSSI, S.; MIKULKA, J.

Vydáno

25. 4. 2023

ISBN

978-80-214-6153-6

Kniha

Proceedings I of the 29th Student EEICT 2023

Strany od

1

Strany do

5

Strany počet

5

BibTex

@inproceedings{BUT183567,
  author="Serge Ayme {Kouakouo Nomvussi} and Jan {Mikulka}",
  title="Image Reconstuction in Electrical Impedance Tomography with Total Variation and Gauss-Newton Algorithm",
  booktitle="Proceedings I of the 29th Student EEICT 2023",
  year="2023",
  pages="1--5",
  isbn="978-80-214-6153-6"
}