An adaptive weighting parameter selection for improved integrated active contour model

In this paper, we propose an adaptive weighting parameter selection method for improved integrated active contour model. By employing a weighting parameter, we first integrate the local gradient information into the Chan-Vese model. In order to regularize the level set function, we present a new restraint term to improve the integrated active contour model. Then we propose a novel method to dynamically select the weighting parameter, which can adaptively adjust the weight between the local gradient information and the global region information. The proposed method is flexible to improve the segmentation quality and experiment results demonstrate its effectiveness. (C) 2015 Elsevier GmbH. All rights reserved.