Aug 18, 2025
Thresholding Images for Rhizovision Root Trait Analysis
- Adriana Gonzalez-Alvarez1,
- Grady Welsh2,
- Shersingh Joseph Tumber-Dávila2
- 1University of Puerto Rico;
- 2Dartmouth College
- Tumber-Dávila LabTech. support email: [email protected]
Protocol Citation: Adriana Gonzalez-Alvarez, Grady Welsh, Shersingh Joseph Tumber-Dávila 2025. Thresholding Images for Rhizovision Root Trait Analysis. protocols.io https://dx.doi.org/10.17504/protocols.io.n2bvje6zpgk5/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: July 31, 2025
Last Modified: August 18, 2025
Protocol Integer ID: 223787
Keywords: Rhizovision, Fine root, Image analysis, images for rhizovision root trait analysis member, rhizovision root trait analysis member, scanning root, archival scans of fragmented root, clean scans in fiji, fragmented root, analyzing archival scan, thresholding image, clean scan, scan, dávila lab, crop, imagej, image
Abstract
Members of the Tumber-Dávila Lab developed this protocol for preparing and analyzing archival scans of fragmented roots. The protocol provides directions to compile, crop, and clean scans in Fiji (ImageJ) and analyze scans. Refer to the Tumber-Dávila Lab Workspace for related protocols on scanning roots.
Troubleshooting
Before start
Removing the borders (using Fiji)
Open Fiji (Schindelin et al. 2012).
Import the images that need cropping: File-> Import-> Image Sequence. Once a white box pops up, choose the folder with the images and check the box for virtual stack (the scans are big, and Fiji doesn’t have enough space for them) -> OK.
Use the rectangle feature (top left corner) to delineate the area that needs to be cleared by dragging the rectangle to an appropriate position that removes the edges of the image.
Note
There are arrows on the bottom of the box so you can see how the rectangle fits the different images and you can adjust the rectangle accordingly.
Get its coordinates: Edit-> Selection-> Properties (check the box that lists the coordinates) -> OK. You will get x, y coordinates for the four corners of your rectangle.
Note
Be careful not to crop any roots in the clearing process.
Once you have the coordinates, go to Process-> Batch-> Macro. Choose your input and output folders. Write the following code in the box and hit process (Images will save to the output folder).
// Set background color to white
setBackgroundColor(255, 255, 255); #When you clear the borders, it will replace the space with white instead of black, which is the default color.
// Create rectangle selection
makeRectangle(x, y, width, height); # width = x2-x1, height= y2-y1
// Clear everything outside the rectangle
run("Clear Outside");
Go through all the images to make sure all the borders are clear and identify shadows/stains; some borders will still be visible in the images because not all the images have the borders in the exact same place.
To remove stains and remaining borders, individually clear the images: File-> Open (select the image you want). Use the rectangle feature to select the part you wish to clear: Edit-> Clear. This will clear what’s inside the box. Save the image: File-> Save as-> JPEG.
Once all the borders and stains are cleared, the images are ready for RhizoVision!
Thresholding of images in RhizoVision Explorer
Open RhizoVision Explorer (Seethepalli and York, 2020).
Image pre-processing:
Note
You should open a few images to try out different settings before committing to any particular settings. To do this go to File-> Open-> Image. Try a couple of different settings and hit the “Run analysis” button to see the output image and the features. Adjust as needed.
To test the settings, we selected 3 images of each species. To make sure that the settings work in all scenarios, test one images with a few small roots, one that is noisy (e.g., lots of shadows), and another with a lot of roots. You can also manually count the root tips to see if they match the output value.
Choose the broken roots analysis mode.
Choose to convert pixels to physical units and select the DPI of the images (We used 1200 DPI).
For the image thresholding level, find a number that removes noise/specs without removing small roots. In our case, the 140-thresholding level worked well. This thresholding level is not universal, and it might need adjustment depending on the DPI of the images being analyzed.
Choose to remove non-root objects smaller than 1 mm.
Choose to enable root smoothing: Thresholding Level 2.
Feature extraction:
Choose to enable root pruning to erase false root tips. In our case, Threshold Level 5 was enough.
Running the analysis:
Once you have your settings, go to File-> Batch Analysis and choose the input and output folders. Check the box to save processed images and click the start button to run the analysis.
Let RhizoVision work its magic.
Note
The duration of this process will vary based on the number of scans and may take a couple of hours.
Output
An example of the output file is shown below:
Note
The output file includes features like number of root tips, total root length, average diameter, perimeter, volume, etc.
Calculation of root traits:
Calculate Specific Root Length (SRL) by dividing Total Root Length (listed in the output file) by the root dry weight.
Calculate Root Tissue Density (RTD) by dividing root dry weight by Volume (listed in the output file).
Protocol references
Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., Tinevez, J.-Y., White, D. J., Hartenstein, V., Eliceiri, K., Tomancak, P., & Cardona, A. (2012). Fiji: An open-source platform for biological-image analysis. Nature Methods, 9(7), 676–682. https://doi.org/10.1038/nmeth.2019
Seethepalli, A., & York, L. M. (2020). RhizoVision Explorer—Interactive software for generalized root image analysis designed for everyone (Version 2.0.3) [Computer software]. Zenodo. http://doi.org/10.5281/zenodo.4095629