Limits...
A context-aware delayed agglomeration framework for electron microscopy segmentation.

Parag T, Chakraborty A, Plaza S, Scheffer L - PLoS ONE (2015)

Bottom Line: In particular, given an over-segmented image or volume, we propose a novel framework for accurately clustering regions of the same neuron.Unlike existing agglomerative methods, the proposed context-aware algorithm divides superpixels (over-segmented regions) of different biological entities into different subsets and agglomerates them separately.We report significant improvements attained by the proposed approach in segmentation accuracy over existing standard methods on 2D and 3D datasets.

View Article: PubMed Central - PubMed

Affiliation: Janelia Research Campus, HHMI, Ashburn, VA, USA.

ABSTRACT
Electron Microscopy (EM) image (or volume) segmentation has become significantly important in recent years as an instrument for connectomics. This paper proposes a novel agglomerative framework for EM segmentation. In particular, given an over-segmented image or volume, we propose a novel framework for accurately clustering regions of the same neuron. Unlike existing agglomerative methods, the proposed context-aware algorithm divides superpixels (over-segmented regions) of different biological entities into different subsets and agglomerates them separately. In addition, this paper describes a "delayed" scheme for agglomerative clustering that postpones some of the merge decisions, pertaining to newly formed bodies, in order to generate a more confident boundary prediction. We report significant improvements attained by the proposed approach in segmentation accuracy over existing standard methods on 2D and 3D datasets.

No MeSH data available.


Split-VI of cytoplasm segmentation of two FIBSEM volumes.Left column: test volume 1, right column: test volume 2. Each curve is the average of results in 5 trials. Each point represents either a stopping point for clustering or bias parameter.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4446358&req=5

pone.0125825.g007: Split-VI of cytoplasm segmentation of two FIBSEM volumes.Left column: test volume 1, right column: test volume 2. Each curve is the average of results in 5 trials. Each point represents either a stopping point for clustering or bias parameter.

Mentions: The split-VI plot in Fig 7 show that both variants of the proposed CADA algorithm generates significantly low under and over-segmentation errors than those of Global [7] method in clustering cytoplasm regions only (mitochondria not merged). In order to analyze why this happens, we save the initial confidences (predictor confidence at the beginning of agglomeration) of hc(e) on all e that


A context-aware delayed agglomeration framework for electron microscopy segmentation.

Parag T, Chakraborty A, Plaza S, Scheffer L - PLoS ONE (2015)

Split-VI of cytoplasm segmentation of two FIBSEM volumes.Left column: test volume 1, right column: test volume 2. Each curve is the average of results in 5 trials. Each point represents either a stopping point for clustering or bias parameter.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4446358&req=5

pone.0125825.g007: Split-VI of cytoplasm segmentation of two FIBSEM volumes.Left column: test volume 1, right column: test volume 2. Each curve is the average of results in 5 trials. Each point represents either a stopping point for clustering or bias parameter.
Mentions: The split-VI plot in Fig 7 show that both variants of the proposed CADA algorithm generates significantly low under and over-segmentation errors than those of Global [7] method in clustering cytoplasm regions only (mitochondria not merged). In order to analyze why this happens, we save the initial confidences (predictor confidence at the beginning of agglomeration) of hc(e) on all e that

Bottom Line: In particular, given an over-segmented image or volume, we propose a novel framework for accurately clustering regions of the same neuron.Unlike existing agglomerative methods, the proposed context-aware algorithm divides superpixels (over-segmented regions) of different biological entities into different subsets and agglomerates them separately.We report significant improvements attained by the proposed approach in segmentation accuracy over existing standard methods on 2D and 3D datasets.

View Article: PubMed Central - PubMed

Affiliation: Janelia Research Campus, HHMI, Ashburn, VA, USA.

ABSTRACT
Electron Microscopy (EM) image (or volume) segmentation has become significantly important in recent years as an instrument for connectomics. This paper proposes a novel agglomerative framework for EM segmentation. In particular, given an over-segmented image or volume, we propose a novel framework for accurately clustering regions of the same neuron. Unlike existing agglomerative methods, the proposed context-aware algorithm divides superpixels (over-segmented regions) of different biological entities into different subsets and agglomerates them separately. In addition, this paper describes a "delayed" scheme for agglomerative clustering that postpones some of the merge decisions, pertaining to newly formed bodies, in order to generate a more confident boundary prediction. We report significant improvements attained by the proposed approach in segmentation accuracy over existing standard methods on 2D and 3D datasets.

No MeSH data available.