Object Localization improved GrabCut for Lung Parenchyma Segmentation

This paper proposes an object localization improved GrabCut algorithm for lung parenchyma segmentation. The structure of lung CT images is complicated, to effectively detect lung nodules, accurate extraction of lung parenchyma is an important part of lung nodule detection. The common step in the ext...

Full description

Saved in:
Bibliographic Details
Published in:Procedia computer science 2018, Vol.131, p.1311-1317
Main Authors: Zhang, Shengchao, Zhao, Yuelong, Bai, Pengcheng
Format: Article
Language:English
Subjects:
GrabCut
lung parenchyma segmentation
object localization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes an object localization improved GrabCut algorithm for lung parenchyma segmentation. The structure of lung CT images is complicated, to effectively detect lung nodules, accurate extraction of lung parenchyma is an important part of lung nodule detection. The common step in the extraction of lung parenchyma is to segment the image first and then to detect the ROI of the segmented images. This paper proposes an object localization improved GrabCut[14] algorithm for lung parenchyma segmentation that can automatically select the appropriate bounding box that relatives to lung parenchyma, then use GrabCut algorithm in the bounding box to accurately segment lung parenchyma of CT image and provides effective basis for lung nodule detection. It overcomes the disadvantages of traditional GrabCut algorithm selecting bounding box manually. The experimental results show that the proposed algorithm can effectively segment the lung parenchyma of different morphologies and is insensitive to noise.
ISSN:1877-0509
1877-0509
DOI:10.1016/j.procs.2018.04.330