表面距离
我真的很困惑，我在网上找不到任何关于表面距离的有用信息，它是什么以及如何计算它。
I'm really stuck and i can't find any thing helpful in the net about surface distance, what is it and how can i calculate it.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
我猜这里的问题涉及地球表面两点之间的距离。如果两点靠得很近，您可以将地球视为平坦并使用毕达哥拉斯定理：
http:// /www.purplemath.com/modules/distform.htm
如果它们相距较远，您需要考虑地球是球形的事实。在这种情况下，沿着地球表面的两点之间的距离是“大圆”距离：
http://www.movable-type.co.uk/scripts/latlong.html
这两种方法都不包括当地拓扑（山丘）、障碍物（如河流）或限制车辆的影响路线网络（公路、铁路），所有这些都使计算变得复杂。
如果您尝试实现这些算法，请确保您了解海里和法定英里之间的区别，否则您的解决方案将是错误的。
I'm guessing the problem here involves distance between two points on the surface of the earth. If the two points are close together you can treat the earth as flat and use the Pythagorean Theorem:
http://www.purplemath.com/modules/distform.htm
If they're further apart you need to consider the fact that the earth is spherical. The distance between two points in this case along the surface of the earth is the "great circle" distance:
http://www.movable-type.co.uk/scripts/latlong.html
Neither of these methods include the influence of local topology (hills), obstacles (like rivers), or vehicles restricted to a network of routes (roads, railways), all of which complicate the calculations.
Make sure you understand the difference between Nautical Miles and Statute Miles if you try to implement these algorithms, or your solution will be wrong.
基于表面距离的指标（默认为间距）
注意：这些度量是对称，这意味着从 A 到 B 的距离与从 B 到 A 的距离相同
。分割体积 (A) 的每个轮廓体素，与参考体积 (B) 的最近轮廓体素的欧几里得距离被计算并存储为
list1。还对参考体积 (B) 的轮廓体素执行此计算，存储为list2。list1和list2合并得到list3。豪斯多夫距离是list3的最大值。豪斯多夫距离 95% 百分位是list3的 95% 百分位。平均表面距离是list3的平均值。中值表面距离是list3的中值。Std surface distance是list3的标准偏差。参考文献：
seg-metrics是一个简单的包，用于计算 <医疗图像分割（后缀为.mhd、.mha、.nii、.nii.gz的图像或.nrrd），并将它们写入 csv 文件。Surface Distance based metrics (with spacing as default)
Note: These metrics are symmetric, which means the distance from A to B is the same as the distance from B to A.
For each contour voxel of the segmented volume (A), the Euclidean distance from the closest contour voxel of the reference volume (B) is computed and stored as
list1. This computation is also performed for the contour voxels of the reference volume (B), stored aslist2.list1andlist2are merged to getlist3.Hausdorff distanceis the maximum value oflist3.Hausdorff distance 95% percentileis the 95% percentile oflist3.Mean (Average) surface distanceis the mean value oflist3.Median surface distanceis the median value oflist3.Std surface distanceis the standard deviation oflist3.References:
seg-metricsis a simple package to compute different metrics for Medical image segmentation(images with suffix.mhd,.mha,.nii,.nii.gzor.nrrd), and write them to csv file.