基于形态学多尺度算法的肺部CT图像边缘检测  

The lung CT image edge detection based on multi-scale morphological approach

作　　者：商艳丽[1] 夏志成[1] 

机构地区：[1]兰州交通大学电子与信息工程学院,甘肃兰州730070

出　　处：《自动化与仪器仪表》2008年第1期43-45,共3页Automation & Instrumentation

摘　　要：医学图像边缘检测是医学图像处理和分析的基础,传统边缘检测算子对噪声敏感,检测到的图像边缘效果较差。本文提出了一种基于形态学多尺度算法的肺部CT图像边缘检测方法。首先对形态学边缘检测算子进行改进,然后利用形态学多尺度算法检测各尺度下的图像边缘,最后采用非均匀权值方法合成最终边缘。实验结果表明:该方法在检测出肺部图像边缘的同时能够很好地抑制噪声,是一种有效的肺部CT图像边缘检测方法。The medical image edge detection is the foundation of medical image following processing. The traditional edge detection operators are sensitive to the noise. This paper proposed a multi-scale morphological approach for the lung CT image edge detection, Firstly, It improves the traditional morphological edge detection operator. Then it detects image edge under these scale by multi-scale, Lastly, it gets the final edge by non-even coefficients. The experimental results indicates that this method can not only obtain the lung edge but also suppress noise. It is an effective edge detection method.

关 键 词：医学图像 数学形态学 多尺度 边缘检测 结构元素 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]