Segmentation of yeast cell’s bright-field image with an edge-tracing algorithm

• DOI: 10.1117/1.jbo.23.11.116503
• Publication Date: 2018-11-20T02:51:25Z
• Authors: Linbo Wang; Simin Li; Zhenglong Sun; Gang Wen; Fan Zheng; Chuanhai Fu; Hui Li
• Keywords: Tracing

Phenotype analysis of yeast cell requires high-throughput imaging and automatic abundant image data. At first, each needs to be segmented labeled in the bright-field images. However, ambiguous boundary images leads failure traditional segmentation algorithms. We propose a algorithm based on morphological characteristics cells. Seed points are first identified along contour then connected by an edge tracing approach. In this way, “ill-detected” noise removed so that edges cells can successfully extracted with sparsely distributed densely packed images, normal morphology also correctly labeled.