Image sequence segmentation combining global labeling and local relabeling and its application to materials science images

Abstract

ABSTRACT Accurately segmenting a series of 2D serial-sectioned images for multiple, contiguous 3D structures has importantapplications in medical image processing, video sequence analysis, and materials science image segmentation.While 2D structure topology is largely consistent across c onsecutive serial sections, it may vary locally becausea 3D structure of interest may not span the entire 2D sequence. In this paper, we develop a new approach toaddressthis challengingproblem by consideringboth the gl obal consistency and possible local inconsistency of the2D structural topology. In this approach, we repeatedly propagate a 2D segmentation from one slice to another,and we formulate each step of this propagation as an optimal labeling problem that can be ec iently solvedusing the graph-cut algorithm. Speci“cally, we divide the optimal labeling into two steps: a global labeling thatenforces topology consistency, and a lo cal labeling that identi“es possible topology inconsistency. We justify theeectiveness of the proposed approach by using it to segmen t a sequence of serial-sectionmicroscopic images of analloy widely used in material sciences and compare its performance against several existing image segmentationmethods.Keywords: Segmentation, Materials, Propagation, Topology Constraints, Local and Global