Detail publikace

Pre-processing for Segmentation of Computer Tomography Images

Originální název

Pre-processing for Segmentation of Computer Tomography Images

Anglický název

Pre-processing for Segmentation of Computer Tomography Images

Jazyk

en

Originální abstrakt

A new method for pre-processing of computer tomography (CT) images is proposed in the paper. The method is based on down-sampling of histogram of an intensity CT image. This helps to overcome a disadvantage of vommon segmentation algorithms - low contrast and blurring of iteresting objects in CT images. Inaccuracy resulted from the down-sampling procedure is improved using a Markov random field model which takes into account geometrical constraints of the processed image.

Anglický abstrakt

A new method for pre-processing of computer tomography (CT) images is proposed in the paper. The method is based on down-sampling of histogram of an intensity CT image. This helps to overcome a disadvantage of vommon segmentation algorithms - low contrast and blurring of iteresting objects in CT images. Inaccuracy resulted from the down-sampling procedure is improved using a Markov random field model which takes into account geometrical constraints of the processed image.

BibTex


@inproceedings{BUT17278,
  author="Tomáš {Červinka} and Ivo {Provazník}",
  title="Pre-processing for Segmentation of Computer Tomography Images",
  annote="A new method for pre-processing of computer tomography (CT) images is proposed in the paper. The method is based on down-sampling of histogram of an intensity CT image. This helps to overcome a disadvantage of vommon segmentation algorithms - low contrast and blurring of iteresting objects in CT images. Inaccuracy resulted from the down-sampling procedure is improved using a Markov random field model which takes into account geometrical constraints of the processed image.",
  booktitle="RADIOELEKTRONIKA 2005",
  chapter="17278",
  year="2005",
  month="january",
  pages="167",
  type="conference paper"
}