Title :
3D Tracking of Cellulose Fibres in Volume Images
Author_Institution :
Swedish Univ. of Agric. Sci., Uppsala
fDate :
Sept. 16 2007-Oct. 19 2007
Abstract :
Segmentation of individual fibres in volume images is important when analysing the three dimensional (3D) fibre structure in paper and cellulose based composite materials. This paper presents a novel method for 3D tracking of individual fibres which can be used as a pre-segmentation step to a full cell wall segmentation or be used to estimate the fibre orientation. The tracking starts in one seed in each fibre and automatically extracts the local fibre orientation and the fibre centre point in each step using 3D information. Good results are obtained for cellulose fibres that are partially collapsed, cracked or irregularly shaped. The proposed method can also be used in other applications where tracking of tubular structures is of interest.
Keywords :
Radon transforms; X-ray microscopy; computer vision; computerised tomography; image segmentation; 3D tracking; cellulose fibres; full cell wall segmentation; image segmentation; three dimensional fibre structure; tubular structures; Composite materials; Data mining; High-resolution imaging; Image analysis; Image resolution; Image segmentation; Optical fiber networks; Optical imaging; Optical materials; X-ray imaging; 3D tracking; Cellulose fibres; Image segmentation; Radon transform; X-ray microtomography;
Conference_Titel :
Image Processing, 2007. ICIP 2007. IEEE International Conference on
Conference_Location :
San Antonio, TX
Print_ISBN :
978-1-4244-1437-6
Electronic_ISBN :
1522-4880
DOI :
10.1109/ICIP.2007.4380016
