How to calculate width of worms in the image?

One way is to find the centroid of object and measure minimal distance between boundary point and centroid.
Another way measure minimal distance through centroid of object.Is there any way to implement in the ImageJ?
Image

What about divide area by length? Skeletonize to get length.

-Esteban

2 Likes

@amaze

Indeed: the question is dealt with here:

Another possibility:fFor actual length

run("Ridge Detection", "line_width=10 high_contrast=230 low_contrast=87 darkline displayresults add_to_manager method_for_overlap_resolution=NONE sigma=3.39 lower_threshold=0 upper_threshold=0.85 minimum_line_length=0 maximum=100000");
roiManager("Show All without labels");

roiManager("Show All with labels");
roiManager("Show All without labels");
roiManager("Select", 23);
roiManager("Measure");
roiManager("Select", 26);
roiManager("Measure");
roiManager("Select", 29);
roiManager("Measure");

C124|329|328.887|
C127|383|382.243|
C130|340|339.399|

Results.csv (219 Bytes)

The Ridge Detection Plugin can also give you breadth.(width)
Check the correct boxes.

1 Like

Thanks @Mathew. I appreciate your help.

Thanks @estebanf. I will try it.

Hi @amaze,

another possibility would be to use the Local Thickness plugin (→Analyze→Local Thickness) in Fiji

Since your provided image is already binary for the analysis you cound add after your segmentation the following analysis:

run("Set Measurements...", "area mean standard modal min center perimeter fit shape feret's integrated median skewness kurtosis display redirect=None decimal=3");
run("Invert");
run("Analyze Particles...", "exclude clear add");
run("Local Thickness (masked, calibrated, silent)");
roiManager("Measure");

I inverted your image since it seems that local thickness expects white objects.

In the results table you can then actually use different outputs.
Mean = average thickness/width
Median = median thickness/width
Max / min = smallest and biggest width
Mode = the most prominent (most often present) width

This might give you potentially even a more distinct read-out

3 Likes

Thanks @biovoxxel. I use your tips.