Imagej newcomer needs help with his assignment

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Imagej newcomer needs help with his assignment

SMoeller
This post was updated on .
Hello,

first of all my english is not the best but I hope everybody undestands me.
As you can see in my title, I am new to Imagej and i need hel with my assignemt.

My task is to count "errors" (so called "twins") in magnesium microstructure pictures.
In this original picture u can see an area with those twins (in the top of the picture) and an area without (bottom)
Image with and without twins

I tried to solve this problem with the "Analyze particles" tool but when thresholding the picture its not possible for me to seperate the twins from the cornborders neither from the corns.
Example:
Threshold BW

The next idea i have is pretty hard to explain with my english but I will give it a try.
I tried to bring out the area between the twins with the "Substract Background" tool so that these areas have a specific gray value (for example between 150 and 200). After that I determined the average area between the twins (for example 200 pix). Now I took a look at every pixel in the image and its gray value if its between 150 and 200 I added it up. Lets say the result was 500.000 pixel with a gray value between 150 and 200.
Now i calculate 500.000 / 200 = 2500 areas that are surrounded by twins.
I even found a plugin to count all pixels with specific gray value but there are 2 problems.
First of all its not the exact number of twins. The other problem is there are still many pixel that are not surrounded by twins but its gray value is between 150 and 200.
Background subtracted

I really panic a bit because i dont see a way to solve my task. I hope you understand my problem and somebody can help me to solve it.


Greetz
SMoeller