Applying fuzzy growing snake to segment cell nuclei in color biopsy images

Authors:
Min Hu;Xijian Ping;Yihong Ding
Affiliations:
Information Engineering University, Zhengzhou, China;Information Engineering University, Zhengzhou, China;Information Engineering University, Zhengzhou, China
Venue:
CIS'04 Proceedings of the First international conference on Computational and Information Science
Year:
2004

Citing 5
Cited 0

On active contour models and balloons

CVGIP: Image Understanding
A fast algorithm for active contours and curvature estimation

CVGIP: Image Understanding
Unsupervised cell nucleus segmentation with active contours

Signal Processing - Special issue on deformable models and techniques for image and signal processing
An evolutionary tabu search for cell image segmentation

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Snakes, shapes, and gradient vector flow

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a novel cell nucleus segmentation method for color esophageal biopsy image. For each nucleus of cell image, based on color characteristics of cell nucleus, a threshold separating the nucleus can be detected automatically in each RGB color component. According to the thresholds, two fuzzy domains are established for each color component with bell-curve and S-curve membership functions. Then we propose a novel growing snake to extract cell nucleus boundary. Described in polar coordinates, the proposed snake is driven by the potential energy and the growing energy integrating the fuzzification information of tristimulus components. The proposed model has low computation cost and strong anti-noise ability. The experiments on a number of cell images show encouraging results.