Measurement of smallest distance within selection?

Hi together,

I ve been using ImageJ now for months and it has been quite reliable for my puposes. There are still some Feature I could not figure out though they seem very simple to me.

For example I have to find out “bottlenecks” in selected structures in an Image, i.e. the smallest distance within a selection. Imagine imagej being used to measure the mimimum and the maximum widths of streets. The wand tool gives a very simple way to select structures like streets. But how do get the most narrow distance? And is there a way to get imagej not only to show the value but also to mark theses places on the Image?

Thanks for your support.

Welcome to the forum, @soennen, and sorry this sat unanswered for so long. It would be helpful if you could post a sample image, and a manual segmentation illustrating what you want to achieve. Then someone here will have a better chance of offering suggestions on how to automate it.

If you haven’t tried it already, the Trainable Weka Segmentation plugin is a powerful machine learning approach which lets you segment en masse after some initial user guidance. See also the Segmentation page of the ImageJ wiki for an overview of segmentation methods.

If you convert your selection to a binary image, then you can just calculate the distance transform (Process > Binary > Distance Map):

1 Like