Robust Evolution Method of Active Contour Models and Application in Segmentation of Image Sequence

Guoqi Liu,Haifeng Li

doi:10.1155/2018/3493070

Abstract

Active contour models are widely used in image segmentation. In order to obtain ideal object boundary, researchers utilize various information to define new models for image segmentation. However, the models could not meet all scenes of image. In this paper, we propose a block evolution method to improve the robustness of contour evolution. A block matrix is consisted of contours of former iterations and contours of shape prior, and a nuclear norm of the matrix is a measure of the similarity of these shapes. The constraint of the nuclear norm minimization is imposed on the evolution of active contour models, which could avoid large deformation of the adjacent curves and keep the shape conformability of contour in the evolution. The shape prior can be integrated into the block evolution method, which is effective in dealing with missing features of images and noise. The proposed method can be applied to image sequence segmentation. Experiments demonstrate that the proposed method improves the robust performance of active contour models and can increase the flexibility of applications in image sequence segmentation.

Highlights

Object extraction and image segmentation [1] are an important and fundamental topics in computer vision and image processing
Snakes or active contour models (ACM) [2] which have shown their great performances are the key methods for image segmentation
The active contour is usually represented by landmarks in parametric active contours and an energy functional was originally introduced by Kass et al [2], while contours in implicit approaches are represented by level set [3, 5], which offers great flexibility for the curve topology

Summary

Introduction

Object extraction and image segmentation [1] are an important and fundamental topics in computer vision and image processing. Snakes or active contour models (ACM) [2] which have shown their great performances are the key methods for image segmentation. A contour is evolved by minimizing some certain energies to match the object boundary while preserving the smoothness of the contour. There are broadly two types of active contour models according to the representation of the curve, that is, parametric active contours [2] and implicit active contours [3,4,5]. The active contour is usually represented by landmarks in parametric active contours and an energy functional was originally introduced by Kass et al [2], while contours in implicit approaches are represented by level set [3, 5], which offers great flexibility for the curve topology. The numerical computations of evolving level set function can be elegantly performed by using the mature numerical algorithm of partial differential equations (PDE) [6]

Methods

Results

Conclusion