Operations Research Transactions ›› 2014, Vol. 18 ›› Issue (3): 79-87.

• Original Articles • Previous Articles     Next Articles

 Solving Chan-Vese model for image segmentation via BB algorithm

PENG Yaxin1, CHEN Sasa1, SHEN Chaomin2, YING Shihui1,*   

  1. 1. Department of Mathematics, College of Science, Shanghai University, Shanghai 200444, China, 2. Department of Computer Science, East China Normal University, Shanghai 200241, China
  • Online:2014-09-15 Published:2014-09-15

Abstract: This paper proposed a new approach for image segmentation-----BB algorithm. The advantage of this algorithm was using the current and last points' information to determine the step-size at each step. Firstly, the paper transformed the CV model into optimization problem through a variational level set method. Secondly, BB algorithm was introduced to solve the optimization problem. Then, the paper analyzed the convergence of BB algorithm, which provided a theoretical basis for the application of the algorithm in the CV model. At last, the proposed algorithm was compared with the conventional steepest descent method and conjugate descent method on several real data. The results validated that the proposed BB algorithm was faster with comparable accuracy.

Key words: image segmentation, CV model, BB algorithm

CLC Number: