A Context-aware Delayed Agglomeration Framework for Electron Microscopy Segmentation

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.

References in corpus (4)

• Natural Scales in Geographical Patterns
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• Machine learning of hierarchical clustering to segment 2D and 3D images
• Annotating Synapses in Large EM Datasets

Cited by in corpus (7)

• A Multi-Pass Approach to Large-Scale Connectomics
• Morphological Error Detection in 3D Segmentations
• Anisotropic EM Segmentation by 3D Affinity Learning and Agglomeration
• Toward Streaming Synapse Detection with Compositional ConvNets
• Artificial Intelligence-Assisted Workflow for Transmission Electron Microscopy: From Data Analysis Automation to Materials Knowledge Unveiling
• What Properties are Desirable from an Electron Microscopy Segmentation Algorithm
• SANTIAGO: Spine Association for Neuron Topology Improvement and Graph Optimization