Multiphase level set method and its application in cell segmentation

Authors:
Yayun Zhou;Arjan Kuijper;Lin He
Affiliations:
Siemens Research, München, Germany;Austrian Academy of Sciences, Linz, Austria;University of California, Los Angeles
Venue:
SPPRA '08 Proceedings of the Fifth IASTED International Conference on Signal Processing, Pattern Recognition and Applications
Year:
2008

Citing 5
Cited 0

Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations

Journal of Computational Physics
Scale-Space and Edge Detection Using Anisotropic Diffusion

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Multiphase Level Set Framework for Image Segmentation Using the Mumford and Shah Model

International Journal of Computer Vision
Solving the Chan-Vese model by a multiphase level set algorithm based on the topological derivative

SSVM'07 Proceedings of the 1st international conference on Scale space and variational methods in computer vision
Active contours without edges

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this work, we will use the active contour model without edges from Chan and Vese. Based on this model and the Four-Colour theorem, a multiphase level set method was developed. It has been applied in image segmentation in recent years and become a popular research topic due to its excellent performance. However, for segmentation of clustered cells this mathod fails. We extended the model with a pre-processing step based on the Perona - Malik method, which improved results dramatically. The theory of the model is given and its numerical implementation briefly introduced. The application of this model in real cell images is discussed.