Limits...
Parameter-free binarization and skeletonization of fiber networks from confocal image stacks.

Krauss P, Metzner C, Lange J, Lang N, Fabry B - PLoS ONE (2012)

Bottom Line: The size and intensity pattern of the template is automatically adapted to the input data so that the method is scale invariant and generic.Furthermore, the template matching threshold is iteratively optimized to ensure that the final skeletonized network obeys a universal property of voxelized random line networks, namely, solid-phase voxels have most likely three solid-phase neighbors in a 3 x 3 x 3 neighborhood.This optimization criterion makes our method free of user-defined parameters and the output exceptionally robust against imaging noise.

View Article: PubMed Central - PubMed

Affiliation: Department of Physics, Biophysics Group, Friedrich-Alexander University, Erlangen, Germany.

ABSTRACT
We present a method to reconstruct a disordered network of thin biopolymers, such as collagen gels, from three-dimensional (3D) image stacks recorded with a confocal microscope. The method is based on a template matching algorithm that simultaneously performs a binarization and skeletonization of the network. The size and intensity pattern of the template is automatically adapted to the input data so that the method is scale invariant and generic. Furthermore, the template matching threshold is iteratively optimized to ensure that the final skeletonized network obeys a universal property of voxelized random line networks, namely, solid-phase voxels have most likely three solid-phase neighbors in a 3 x 3 x 3 neighborhood. This optimization criterion makes our method free of user-defined parameters and the output exceptionally robust against imaging noise.

Show MeSH
Automatic adaption of template size.After subtracting the mean intensity from each template pixel, the borders of the pattern can be identified by their negative values. Thus, the template can be adjusted to an optimal size. Due to this feature, the algorithm is scale invariant.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3351466&req=5

pone-0036575-g008: Automatic adaption of template size.After subtracting the mean intensity from each template pixel, the borders of the pattern can be identified by their negative values. Thus, the template can be adjusted to an optimal size. Due to this feature, the algorithm is scale invariant.

Mentions: The template for each scan direction is generated by weighted averaging over a large amount () of randomly chosen 2D sections throughout the 3D image stack (Fig. 7). The weighting coefficient of each 2D section is proportional to the intensity of its central pixel. This weighting mechanism assures that only sections that contain a bright fiber at the center contribute significantly to the average. The resulting template represents a typical cross section of a fiber with a bright core and darker borders (Fig. 6). Due to the subtraction of the global mean intensity from each vector component (see above), core pixels are positive, while border pixels are negative. This change of sign is used to automatically adjust the template size (Fig. 8). Due to the automatic size adaption, the algorithm becomes scale invariant with respect to the input data and independent from the optical resolution of the recorded line networks.


Parameter-free binarization and skeletonization of fiber networks from confocal image stacks.

Krauss P, Metzner C, Lange J, Lang N, Fabry B - PLoS ONE (2012)

Automatic adaption of template size.After subtracting the mean intensity from each template pixel, the borders of the pattern can be identified by their negative values. Thus, the template can be adjusted to an optimal size. Due to this feature, the algorithm is scale invariant.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3351466&req=5

pone-0036575-g008: Automatic adaption of template size.After subtracting the mean intensity from each template pixel, the borders of the pattern can be identified by their negative values. Thus, the template can be adjusted to an optimal size. Due to this feature, the algorithm is scale invariant.
Mentions: The template for each scan direction is generated by weighted averaging over a large amount () of randomly chosen 2D sections throughout the 3D image stack (Fig. 7). The weighting coefficient of each 2D section is proportional to the intensity of its central pixel. This weighting mechanism assures that only sections that contain a bright fiber at the center contribute significantly to the average. The resulting template represents a typical cross section of a fiber with a bright core and darker borders (Fig. 6). Due to the subtraction of the global mean intensity from each vector component (see above), core pixels are positive, while border pixels are negative. This change of sign is used to automatically adjust the template size (Fig. 8). Due to the automatic size adaption, the algorithm becomes scale invariant with respect to the input data and independent from the optical resolution of the recorded line networks.

Bottom Line: The size and intensity pattern of the template is automatically adapted to the input data so that the method is scale invariant and generic.Furthermore, the template matching threshold is iteratively optimized to ensure that the final skeletonized network obeys a universal property of voxelized random line networks, namely, solid-phase voxels have most likely three solid-phase neighbors in a 3 x 3 x 3 neighborhood.This optimization criterion makes our method free of user-defined parameters and the output exceptionally robust against imaging noise.

View Article: PubMed Central - PubMed

Affiliation: Department of Physics, Biophysics Group, Friedrich-Alexander University, Erlangen, Germany.

ABSTRACT
We present a method to reconstruct a disordered network of thin biopolymers, such as collagen gels, from three-dimensional (3D) image stacks recorded with a confocal microscope. The method is based on a template matching algorithm that simultaneously performs a binarization and skeletonization of the network. The size and intensity pattern of the template is automatically adapted to the input data so that the method is scale invariant and generic. Furthermore, the template matching threshold is iteratively optimized to ensure that the final skeletonized network obeys a universal property of voxelized random line networks, namely, solid-phase voxels have most likely three solid-phase neighbors in a 3 x 3 x 3 neighborhood. This optimization criterion makes our method free of user-defined parameters and the output exceptionally robust against imaging noise.

Show MeSH