摘要
为了克服噪声、弱边界和灰度不均匀现象对图像分割的影响,提出空间信息熵驱动的活动轮廓图像分割模型。该模型通过空间信息熵刻画图像灰度的变化,利用图像的全局统计信息来克服灰度不均匀性。并构造一种新的基于空间信息熵的符号压力函数来促使轮廓曲线向边缘靠近,最终停留在图像的边缘,完成对目标的分割。此外,该模型采用二值水平集方法求解,避免了繁杂的计算过程。实验结果表明,该模型可以克服复杂背景对图像分割的影响,实现快速准确的分割。
In order to reduce the impacts of noise, weaken boundary and intensity inhomogeneity, a new active contour model driven by spatial entropy applied on medical image segmentation is proposed. The spatial entropy of image is abstracted to describe the intensity variation. As a result, the global statistic intensity information is used to overcome the intensity inhomogeneity. In addition, the signed pressure force function based on the spatial entropy is structured, which drives the contour curve close to the boundary of the target and achieves the segmentation of the object. In addition, this model is implemented by the binary level set method, which avoids complex computation. The experimental results show that this method can overcome the influence of complex background on the segmentation and realize fast and accurate segmentation.
作者
马翔
楚莹莹
陈允杰
MA Xiang;CHU Ying-ying;CHEN Yun-jie(School of Mathematics,Engineering University of CAPF,Xian 710086,China;School of Mathematics and Statistics,Nanjing University of Information Science and Technology,Nanjing 210044,China)
出处
《控制工程》
CSCD
北大核心
2018年第11期2010-2016,共7页
Control Engineering of China
关键词
灰度不均匀
图像分割
空间信息熵
符号压力函数
二值水平集方法
Intensity inhomogeneity
image segmentation
spatial entropy
signed pressure force function
binary level set method
