| 一种改进的变分水平集的图像分割算法 |
| |
| 引用本文： | 张桂梅,周飞飞,储 珺. 一种改进的变分水平集的图像分割算法[J]. 图学学报, 2015, 36(5): 740. DOI: 10.11996/JG.j.2095-302X.2015050740 |
| |
| 作者姓名： | 张桂梅 周飞飞 储 珺 |
| |
| 摘 要： | 为了解决灰度不均匀现象对医学图像的干扰问题，提出了基于局部极性信息的活动轮廓模型。通过引入局部图像信息，该模型能有效地分割灰度不均匀图像。在规则化项中增加的能量惩罚项，使得水平集函数在演化过程中保持为近似的符号距离函数。该算法将图像分割问题归结为曲线能量泛函的最小化，首先建立包含局部灰度信息（极性信息）和改进的符号距离函数的曲线演化能量泛函；然后采用变分水平集方法求解能量函数的最小值，得到最终的分割结果。真实医学图像和人工合成图像的实验结果表明，此方法对灰度不均匀的医学图像有较高的分割精确度，在图像分割速度上有较大提高。由于利用了局部灰度信息，可以有效地分割灰度不均匀的医学图像，而改进后的变分水平集可以完全避免重新初始化，使得图像分割效率大大提高了。
|
| 关 键 词： | 变分水平集 图像分割 符号距离函数 极性信息 |
An Improved Variational Level Set Used in Image Segmentation Algorithm |
| |
| Zhang Guimei,Zhou Feifei,Chu Jun. An Improved Variational Level Set Used in Image Segmentation Algorithm[J]. Journal of Graphics, 2015, 36(5): 740. DOI: 10.11996/JG.j.2095-302X.2015050740 |
| |
| Authors: | Zhang Guimei Zhou Feifei Chu Jun |
| |
| Abstract: | To solve the problem caused by intensity inhomogeneity in medical images, this paperproposed an active contour model based on the local polarity information. By incorporating the localinformation of image, the proposed model can efficiently segment the image which intensity isnonuniform. Through introducing a penalizing energy term into the regularization, the level setfunction can be approximated as a signed distance function all the time in the process. This algorithmregarded the image segmentation as the task to minimize the curve of energy functional. Firstly, toestablish the evolution curve of energy functional including the local gray level information (polarityinformation) and the improved signed distance function. Then to solve the minimization of energyfunction by using the variational level set method to be able to get the final segmentation result. Theexperiment results of real medical images and artificial synthesis images have shown that our methodcan get a higher precision of segmentation for the medical images with uneven grayscale. In addition,the image processing speed has been improved greatly. Because of the utilization of local grayinformation, we can segment the medical image with uneven grayscale effectively, the improvedvariational level set can completely avoid the re-initialization, and greatly improve the efficiency ofimage segmentation. |
| |
| Keywords: | variational level set image segmentation symbolic distance function polarity information |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《图学学报》浏览原始摘要信息 |
|
点击此处可从《图学学报》下载免费的PDF全文 |
|
