Abstract: This article comes up with a medical image segmentation algorithm based on partial threshold method.In X-ray medical image，the target has a wide dynamic range and the overlapping interval of the target and the background gray is large.This article first enhances the featured texture of the target，and then adopts the minimized partition including the target zone，and at last partial threshold method will be incorporated with density function to segment the image.The experiment reveals that this method can effectively separate the bone image of medical pictures.

Key words: image segmentation, partial threshold method, minimized partition, density function

摘要： 提出了一种局部阀值法的医学图像分割算法。针对X线医学图像中存在目标动态范围广，目标与背景灰度值重叠区间大等特点，在对目标特有纹理进行增强的基础上，采用了包含目标区域最小化的分区，最后利用局部阀值法结合密度函数分割图像。实验表明，该方法能有效分割出医学图像中的骨质影像。

关键词: 图像分割, 局部阀值法, 最小化分区, 密度函数