Systems, Man, and Cybernetics

 

Home

• [P1206] A multiresolution flow based multiphase image segmentation
  Author(s): Celia A.Z. Barcelos, Emilio Zorzo Barcelos, Jose A. Cuminato
  
  

  Abstract
  In this work it is proposed a variational model for simultaneous smoothing and multiphase image segmentation. By assuming that the pixel intensities are independent samples from a mixture of Gaussians and by interpreting the phase fields as probabilities of pixels belonging to a certain phase,it is derived the model formulation by maximizing the mutual information between image features and phase fields. The proposed energy functional $J_{\epsilon}$ consists of three parts: the smoothing term for the reconstructed image, the regularization for the boundaries in hard segmentation a likelihood estimator based on the density function. The segmentation and image denoising are obtained simultaneously thought the flow equation obtained by minimizing the energy functional with respect to the coefficients and variance in the mixture of Gaussians. We present experimental results on segmenting synthetic and natural color images to show the effectiveness of the proposed model.

  
   

Full Text (PDF)