Re: Is there a way to measure depth?

Posted by Nathaniel Ryckman on Apr 14, 2011; 7:16pm
URL: http://imagej.273.s1.nabble.com/Is-there-a-way-to-measure-depth-tp3684978p3684981.html

You could give Process->Find edges a try

It's really hard to determine the nature of the problem without seeing an example. Basically, you have to figure out how you are able to see the differences before you can give the computer explicit commands to do the same thing.

Unfortunately, this often has the potential to turn seemingly simple tasks into very complicated ones.

jkaufman wrote

I am following ASTM D1002 standard for my lap shear test, so the approximate area of the joint is 1 in x .5 in.  There are two types of failure of the joint, adhesion failure where the adhesive splits off from the aluminum, and there is a distinct edge between the aluminum and the adhesive.  The problem that I have from these failures is when the adhesive is clear or close in color to the aluminum, and therefore I cannot use thresholding to separate the two parts. Cohesion failure is much more difficult to analyze, as this is a failure of the adhesive, and often there are only small and scattered parts of aluminum visible.

I may be able to solve my problem with adhesion failure by coloring the adhesive with a black sharpie, however cohesion failure is too complex to simply color over.