基于熵和局部邻域信息的高斯约束CV模型

Gaussian Regularizing CV Model Based on Entropy and Local Neighborhood Information

基于曲线演化的Chan-Vese(CV)模型常常不能准确分割非均匀性且结构复杂的医学图像.针对此缺点,文中提出了一种基于熵和局部邻域信息的高斯约束CV模型,作者利用熵构造内部和外部区域能量的权值系数,加强了对曲线演化的控制;同时将曲线上各点的局部邻域信息引入到曲线演化过程中,提高了分割的准确性,并降低了区域内灰度不均匀等因素对曲线演化的影响;高斯约束保证了曲线演化过程中的稳定性、光滑性,同时不需要曲线周长约束项和重复初始化.利用Circular Hough变换对左心室壁内、外膜进行初始定位,避免了人工设置初始轮廓,减少了曲线向目标轮廓演化时间和初始轮廓位置敏感性对分割结果的影响.作者对心脏MR图像的左心室内、外膜进行了分割.结果表明文中方法能够快速准确地分割左心室壁内、外膜,抗干扰能力强,鲁棒性好.

Due to the heterogeneous and complex constructions in medical images, the Chan-Vese （CV） model cannot achieve satisfying segmentation. We propose a novel Gaussian regularizing CV model using entropy and local neighborhood information. In the cost function of this model, the interior and exterior energies are weighted by the entropy, which improves the robust of the evolving curve. The local information of the curve is considered rather than global image statis- tics, which reduces the impact of the heterogeneous grays inside of regions and improves the seg- mentation results. The Gaussian kernel is utilized to regularize the level set function, which not only keeps the level set function smooth and stable, but also removes the traditional Euclidean length term and re-initialization. To reduce the sensitivity to the initialization, the Circular Hough transformation is used to obtain the initialization automatically. The encouraging results on the cardiac images indicate that the present method has the advantage of high accuracy and strong robustness to segment the endocardium and epicardium of the left ventricle.