Detail publikace

4D dynamic contrast enhanced CT data registration using optical flow

Originální název

4D dynamic contrast enhanced CT data registration using optical flow

Anglický název

4D dynamic contrast enhanced CT data registration using optical flow

Jazyk

en

Originální abstrakt

This paper deals with 4D (3D + time) registration method based on optical flow. The main goal is to obtain the velocity field, which is used to correction of displacement in the thoracic data. The one of the most widely used optical flow method published by Horn and Schunck, is utilized in this paper. The proposed method is based on H-S scheme, which is extended by pyramidal and multi-level approach. This allows estimation a larger displacements and this algorithm is thus more robust. Moreover the optical flow method was extended for three-dimension images that develops in time. The performance of proposed algorithm was applied on POPI model, the public model of lung data, which was deformed by synthetic velocity field.

Anglický abstrakt

This paper deals with 4D (3D + time) registration method based on optical flow. The main goal is to obtain the velocity field, which is used to correction of displacement in the thoracic data. The one of the most widely used optical flow method published by Horn and Schunck, is utilized in this paper. The proposed method is based on H-S scheme, which is extended by pyramidal and multi-level approach. This allows estimation a larger displacements and this algorithm is thus more robust. Moreover the optical flow method was extended for three-dimension images that develops in time. The performance of proposed algorithm was applied on POPI model, the public model of lung data, which was deformed by synthetic velocity field.

BibTex


@inproceedings{BUT107632,
  author="Roman {Jakubíček} and Jiří {Jan}",
  title="4D dynamic contrast enhanced CT data registration using optical flow",
  annote="This paper deals with 4D (3D + time) registration method based on optical flow. The main goal is to obtain the velocity field, which is used to correction of displacement in the thoracic data. The one of the most widely used optical flow method published by Horn and Schunck, is utilized in this paper. The proposed method is based on H-S scheme, which is extended by pyramidal and multi-level approach. This allows estimation a larger displacements and this algorithm is thus more robust. Moreover the optical flow method was extended for three-dimension images that develops in time. The performance of proposed algorithm was applied on POPI model, the public model of lung data, which was deformed by synthetic velocity field.",
  address="LITERA",
  booktitle="Proceedings of the 20th Conference STUDENT EEICT 2014 Volume 3",
  chapter="107632",
  howpublished="print",
  institution="LITERA",
  year="2014",
  month="april",
  pages="222--226",
  publisher="LITERA",
  type="conference paper"
}