0

我最近一直在使用 ITK,但我并不是很熟练,所以如果我的问题看起来很幼稚,我深表歉意!这是问题:我已经在我的机器(Win 7 x64)上成功构建了几个 ITK 示例,然后尝试运行一个示例来测试结果,因为我对分水岭分割感兴趣我尝试了它(WS3D),但是除了必须定义的输入和输出图像文件外,还需要定义一个名为:LabelImageFile 的附加参数,就像这样:

WS3D 输入图像文件标签图像文件输出图像文件

不幸的是,我不知道什么是 LabelImageFile,我怎样才能获得特定图像的标签图像?如果有人能帮助我,我将非常感激,非常感谢,Shawn

4

1 回答 1

2

The watershed segmentation algorithm looks for a function f that is a continuous height function defined over an image domain. Such that a catchment basin is defined as the set of points whose paths of steepest descent terminate at the same local minimum of f. This catchment basin is your watershed result. The label image that the filter is expecting is the function f or the height function. You can generate one such image by applying gradient magnitude to the original image.

于 2012-08-25T09:05:23.997 回答