一种基于 ROEWA 算子改进的 CV 模型用于乳腺超 声图像分割

基于区域的几何活动轮廓(Chan-Vese,CV)模型是乳腺超声图像中常用的一种分割算法.但传统的CV模型不能满足乳腺超声图像分割精度高、速度快的要求.因此,文章提出了一种基于指数加权平均比率(Ratio of Exponential Weighted Averages,ROEWA)算子改进的CV模型,用于乳腺超声图像中病灶区域的分割.首先,计算乳腺超声图像的ROEWA算子.其次,基于图像的ROEWA算子构建边缘指示函数,用于代替CV模型中的Dirac项.最后,去除平滑项,从而提高曲线演化的速度.实验结果表明,文章提出的算法不仅能提高分割的精度,而且能显著提高分割的速度.

Segmentation of Breast Ultrasound Image with an Improved CV Model Based on ROEWA Operator

The Chan-Vese (CV) model has been widely investigated for breast ultrasound (BUS) image segmentation. However, the traditional CV model can not meet the requirement of high precision and speed for BUS segmentation. To address this issue, an improved CV model based on the ratio of exponentially weighted averages (ROEWA) operator was proposed in this paper. Firstly, the ROEWA of the BUS image was calculated. Then, a ROEWA-based edge indicator function was built to replace the Dirac term of traditional CV model. Finally, the smoothing term was removed to improve the speed of curve evolution. Experimental results show that the advantages of the proposed model in terms of computational efficiency and accuracy.