摘要
目的探讨基于Mean Shift方法的肝脏CT图像的自动分割算法,以实现肝脏的自动分割。方法首先对原始图像进行单次Mean Shift平滑,滤除噪声的影响以增强算法的鲁棒性,然后通过Mean Shift迭代自动选取初始种子点,最后采用基于区域生长的方法实现肝脏CT图像的自动分割。结果实验证明此方法是一个准确、快速和有效的肝脏自动分割方法。结论采用本文中提出的方法,可有效地实现肝脏的自动分割。
Objective To assess the method based on Mean Shift for automatic segmentation of liver regions in CT images. Methods Firstly,the Mean Shift smoothing method achieved with a single iteration was applied to remove the noise in the original image for robustness enhancement of the algorithm. Then the initial seed point was selected automatically based on Mean Shift iterations and finally the region growing approach was utilized to implement automatic liver segmentation of the image. Results Experiment results showed that the presented algorithm was an accurate,fast and effective method for automatic segmentation of liver regions in CT images. Conclusion The new algorithm presented in this paper can efficiently implement automatic segmentation of liver regions from CT images.
出处
《中国医学影像技术》
CSCD
北大核心
2010年第12期2367-2370,共4页
Chinese Journal of Medical Imaging Technology
基金
上海市教委选拔培养优秀青年教师科研专项基金(358536)
