If your SEM pictures have good contrast (means gray levels of pores are different of those of matrix) than any software, including ImageJ will work.

I've tried couple of software. Indeed, ImageJ is the best one!

If your SEM pictures have good contrast (means gray levels of pores are different of those of matrix) than any software, including ImageJ will work.

Image J is good enough, provided you have decent quality micrograph.

Thank you everyone for your willingness to help me. My only concern with image J is adjusting the threshold. Should I adjust the threshold in a way that minimizes the pores in the background and only consider the pores in the front plane? Or I need to consider all the pores (even the ones that are in the perspective?!). The pores in the background are smaller than the actual pores due to the perspective effect? Any suggestion is greatly appreciated...

Hey Neda,

This is a good question. Your perspective will likely distort the measure length to some degree. Depending on your depth of field, it may be hard to visually determine exactly what plane the scale bar refers to. This will change depending on your working distance and aperture diameter. To minimize these effects, you may want to focus on only measuring pores that are closest to the front plane as you mentioned. To increase the accuracy of your measurements, you can also zoom in and manually measure each pore in Image J. I know this can be exhausting, but it will likely provide you with the highest quality measurement data. I would make a note of how you measured your pores in your methods section for others that may read your work. Hope my suggestions help. Cheers!

Thank you Robert. I feel more confident now :)

It is difficult to understand what is "perspective" in your images (you did not supply any). Selective measurements are always bad - reviewers will justly question you. May be you need new SEM images with no "perspective".

I've attached an image of what Neda is talking about, I'm working on the same project with her. Basically we are trying to calculate porosity using ImageJ by adjusting the threshold and isolating just the fibers in the foreground. As you can see in the image, there are fibers with faint contrast in the background that we believe that we should not isolate for our porosity measurements. Let me know your feedback.

Hey Kameel,

Based on your image, I would think ImageJ would have a hard time automatically calculating the pore size. I also think it would be best to measure the pores individually by hand. I understand that measuring porosity can be pretty tricky by SEM because of human measurement bias, but based on your sample, I believe it is probably the best way. I found a paper that discusses one research group calculated the average porosity of their fibrous mats using SEM - maybe it might help.

Article Electrospun Fibrous Mats with High Porosity as Potential Sca...

I also found a manuscript that calculated the average porosity using another method (liquid intrusion analysis). I'm not sure if you have that available, but that is another route that might avoid any measurement bias.

www.tissueeng.net/lab/papers/33010_ftp.pdf

Concerning perspective interfering with the measurements, I can't think of a way to remove perspective you inherently get from having 3D mats. Its likely that the pores you measure are not all on the same plane, so every measurement will have varying degrees of error associated with it. The only suggestion I have is if you must use SEM to obtain this measurement, I would try to measure a very large number of pores to get a more representative sampling of the entire mat (probably hundreds if possible).

Best of luck!