As a recent research hot issue, obtaining the accurate 3 D organ models of Visible Human Project (VHP) has many significances. Therefore, how to extract the organ regions of interest (ROI) in the large-scale color slice image data set has become an urgent issue to be solved. In this paper, we propose a method framework based on OneCut algorithm and adjacent image geometric features to continuously extract the main organ regions is proposed. This framework mainly contains two parts: firstly, the OneCut algorithm is used to segment the ROI of target organ in the current image; secondly, the foreground image (obtained ROI) is corroded into several seed points and the background image (other region except for ROI) is refined into a skeleton. Then the obtained seed points and skeleton can be transmitted and mapped onto the next image as the input of OneCut algorithm. Thereby, the serialized slice images can be processed continuously without manual delineating. The experimental results show that the extracted VHP organs are satisfactory. This method framework may provide well technic foundation for other related application.

作为最近的研究热点，获得可见人类计划（VHP）的精确3D器官模型具有许多意义。因此，如何在大规模彩色切片图像数据集中提取感兴趣的器官区域（ROI）已成为亟待解决的问题。本文提出了一种基于OneCut算法和相邻图像几何特征连续提取主要器官区域的方法框架。该框架主要包括两部分：第一，使用OneCut算法对当前图像中目标器官的ROI进行分割。其次，将前景图像（获得的ROI）腐蚀成几个种子点，将背景图像（除ROI以外的其他区域）精致化为骨架。然后，可以将获得的种子点和骨骼传输并映射到下一张图像，作为OneCut算法的输入。由此，无需人工划定就可以连续地处理序列化切片图像。实验结果表明，提取的VHP器官令人满意。该方法框架可以为其他相关应用提供良好的技术基础。