I am new to image processing and attempting to use ImageJ to answer one question - what is the thickness of the bacterial capsule?
The attached are TEM images of cross-sectioned cells. I need to estimate the size/thickness of the capsule (outermost region around the cells). Problem is I have 100s of such images.
I have tried to use the line tool first to set the scale and take several measurements of the capsule all around the cell. However, doing this for all the images will be tedious and will definitely add some bias.
Is there a more rigorous approach to doing this? Thanks in advance for your help.
Thank you for providing images, though for someone to help you it would probably be necessary to be more specific about what you are measuring. The black line to the edge of the fuzzy area? Outside of the black line?
Also, for any type of image processing, it would help to have original images, otherwise you will likely need to crop them to get rid of the text at the bottom.
If you can figure out where you want to start and end your measurement, I would guess the “easiest” method that doesn’t involve hand-drawing would be to create a mask of the area outside of the object. Once you have that area, and the inner circumference, you might be able to calculate the “average” distance out from the cells that the fuzzy area extends. It should be roughly the same as a dilation (1 pixel at a time) until the area matches or exceeds the masked area.
Masking the areas of interest might require a pixel classifier, for example using the Weka plugin. Masking the dark line should only take a threshold, I would think.
I think this older thread can be a help to measure the thichkness:
See also a variant thread:
which als mentions R solutions (if you are comfortable with R).
Thank you for the response. I have added an example image to the dropbox folder.
But like you described, it is the distance between the black line and the edge of the fuzzy area.
Also, all these are the “original” images obtained from our microscopy facility. I will check with them to see if I can get a version without the text.
I will try your suggested solution and let you know how it goes. Thanks.
@Bio7 , thanks for you response. There’s a wealth of information in those threads that should get me the information I need. Thanks much!