基于水平集的医学图像分割改进算法

医学影像分割是图像分割中的难点,具有重要的应用价值。针对医学影像的特点和图像分割算法的性能差异,提出了一种水平集医学图像分割改进算法。首先通过曲线演化仿真,得出水平集算法核心-速度函数。其次选定速度函数实现对图像的粗略分割,将灰度值较大的区域设置成灰度值较小的值,通过仿真演化准确找到图像中目标区域。最后利用选定的速度函数通过初始算法,经过一定次数的迭代操作后实现了医学影像的准确分割。实验结果表明:该算法可以精确地找到肿瘤所在区域,具有较好的分割性能和一定的鲁棒性。

Improved Medical Image Segmentation Algorithm based on Level Set

Medical image segmentation,which is difficult to solve problem in the image segmentation,has important application in the medical field.For the features of medical image and the differences of the image segmentation algorithm,this paper prensents an improved medical image segmenration algorithm using the level set.Firstly this paper proposes the velocity function and the core of level set algorithm by the curve evolution emulation.Secondly,we use rough segmentation to image by choosing velocity function,changing the region of strong gray value into that of small gray value,and finding out the object region in image according to emulation evolution exactly.Finally,we fulfill exact segmentation in medical image through initial algorithm with the help of the chosen velocity function based on times of iterative operation.Experimental result shows that the proposed algorithm is able to identify the region of tumors exactly,as well as it is robust to image segmentation efficiently.