基于自适应区域生长算法的肝脏分割

把肝脏器官从医学图像中提取出来，为肝脏的三维重建及最终的仿真手术模拟提供正确的数据。针对腹部CT图像中脏器组织多、图像纹理结构复杂、灰度差别小、边缘不明显等特点，提出一种改进的自适应区域生长分割算法。该算法基于两个局部参数：待生长点的局部平均灰度和局部平均梯度，对传统区域生长算法的生长准则进行了改进。实验结果表明，得到的肝脏分割结果比传统区域生长算法分割结果更精确，可以为后续的肝脏三维重建及仿真手术提供准确的数据。

Segmentation of liver based on adaptive region growing

Extracting the human abdomen organs from medical images can provide accurate data for the following 3D reconstruction of abdomen organs and the final virtual surgery.For the complex characteristics of medical image, an improved algorithm of adaptive region growing is put forward.Based on two local parameters：The local mean value of the intensity ftmetion and the local mean value of the norm of the intensity gradient,the homogeneity criterion of this algorithm is improved.The experimental results show that： The segmentation results of the liver are more accurate by using adaptive region growing than using classical region growing algorithm,and it can provide accurate data for 3D reconstruction and surgical simulation of liver.