An efficient local and global model for image segmentation

Abstract

In this paper, a new region-based active contour model using a variational level set formulation is proposed for image segmentation. The model is based on curve evolution, local statistical function and level set method. The energy function for the proposed model consists of two components: global component and local component. By introducing the local term, the images with intensity inhomogeneities can be efficiently segmented. Moreover, a smoothness regularization is derived from a Gaussian filtering term. This allows avoiding re-initialization while ensuring the smoothness of the level set function. The addition of the global term makes the model more flexible to the location of initial contour. Experimental results show that our method is less sensitive to the location of initial contour and demonstrate the performance of our model.