Nima Vakili - The authors are with the Mathematics and Computer Science Departments,University of Tarbiat Modares,Tehran,Iran
Mansoor Rezghi - The authors are with the Mathematics and Computer Science Departments,University of Tarbiat Modares,Tehran,Iran
S.Mohammad Hosseini - The authors are with the Mathematics and Computer Science Departments,University of Tarbiat Modares,Tehran,Iran

In this paper, we present a novel level set method for image segmentation by utilizing the Bayesian rule and Mixture of Gaussian distribution as a conductor which its parameters are obtained from EM-algorithm. We use an Adaptive direction function to make the curve evolution robust against the curve’s initial position and a nonlinear adaptive velocity to speed up the process of curve evolution and also a probability-weighted edge and region indicator function to implement a robust segmentation for objects with weak boundaries. The proposed method has the following features:1) By using the Bayesian rule that contains the regional features of image, it automatically determines the curve to shrink or expand; 2) To avoid the leakage at weak boundaries it accompanies the curve evolution an appropriate speed.

Active contour, image segmentation, level set, Gaussian Mixture distribution, EM- algorithm