REGISTRATION OF 3-D IMAGES BY GENETIC OPTIMIZATION
REGISTRATION OF 3-D IMAGES BY GENETIC OPTIMIZATION
复制标题
DOI：
10.1016/0167-8655(95)00051-h
复制
发表时间：
1995-08-01
影响因子：
5.1
通讯作者：
ROUX, C
中科院分区：
文献类型：
--
作者：
JACQ, JJ;ROUX, C
我们提出了一个框架，用于解决医学成像中的三维配准问题的基础上的一个典型的遗传算法（cGA）。3-D配准的问题在所提出的两种应用情况中被陈述为优化问题，即，体积到体积和表面到体积配准。cGA使用随机适应度函数，该函数对数据空间的随机选择的样本进行操作。在更高的层次上，提出了一种自适应搜索空间缩放技术，该技术通过连续激活cGA过程来操作。前一个功能确保了较低的搜索算法的复杂性和最终解决方案的良好的精度。然后考虑体积到体积和表面到体积配准。介绍了特定于应用程序的功能（实际优化空间，适应度或距离函数，GA参数）。结果有关两个注册问题，使用3-D计算机断层扫描数据的介绍和讨论。
We present a framework for solving the 3-D registration problem in medical imaging based on a canonical genetic algorithm (cGA). The issue of 3-D registration is stated as an optimization problem in both application cases presented, i.e., volume-to-volume and surface-to-volume registration. The cGA uses a stochastic fitness function which operates on randomly selected samples of the data space. At a higher level, an adaptive search space scaling technique is presented which operates by successive activations of the cGA procedure. The former features ensure a lower complexity of the search algorithm and a good accuracy of the final solution. Volume-to-volume and surface-to-volume registration are then considered. The features that are specific to the application (the actual optimization space, the fitness or distance function, the GA parameters) are introduced. Results concerning two registration problems using 3-D Computerized Tomography data are presented and discussed.