快速行进法(fast marching method,FMM)已被证明在图像分割方面具有优势,在此基础上提出了一个混合分割的算法。这个方法加入了图像分割后处理步骤,成功解决了活体肝脏CT系列图像自动分割问题。首先是通过滤波去噪等处理得到速率系数图像,然后根据CT图像相邻层间的相似特点计算FMM所需的参数进行图像分割,最后使用开运算修正肝脏边缘。整个序列分割过程只需用户定义一个种子点,减少了人工干预,从而提高了效率和准确性。
It has been proved that Fast marching method(FMM) has advantage in image segmentation.Based on this method,a hybrid segmentation method is proposed.Added with an image segmentation post-processing,the segmentation of liver slices has been solved successfully.In this method,first,a speed image is produced after pretreatment such as filtering and noise reduction and etc.According to the characteristics of serial CT liver images,FMM parameters are attained from contiguous slice outcome to segment the CT image....