Automated Pipeline for Color Analysis in ImageJ (Fiji)

In this video, I share a macro (available here: doi.org/10.5281/zenodo.5595203) that automates the extraction of summary statistics from channels of three colorspaces (RGB, Lab, and HSB). This macro functions to 1) open an image from a directory folder, 2) threshold the image to segment out the object of interest from the background then "clear" the background so it is completely black, 3) separate the image into the component channels for each color space, 4) extract summary statistics for each channel in each colorspace, 5) export these summary statistics to a .csv file.
I also provide an optional R script (available here: doi.org/10.5281/zenodo.5595203) to automate the reformatting of the .csv file with the summary statistics from the imageJ so each channel is transposed into distinct columns. The final .csv file produced by running this code should have each image as a row, with each summary statistic (mean, mode, min, max, etc.) for each channel in each colorspace presented in a separate column for that image (row).
Step-by-step directions, color analysis macro, and R script available for download here: doi.org/10.5281/zenodo.5595203
ImageJ/Fiji available for download here: imagej.net/software/fiji/
Rstudio available for download here: www.rstudio.com/products/rstu...

[@agleto]

Hi! Thanks for this illustrative video and the protocol! Everything works fine

[@Hydro3D]

Brilliant man! Thank you for your service.

[@user-mq4mg8zt1c]

Good job Chris, keep it up:)

[@jameshyatt1907]

This has helped my data analysis so much

[@norafifah2604]

thank u love u

[@savannahb6049]

I keep getting this error: Error in value_vars(value.var, names(data)) : value.var values [c("Area1", "Mean1", "StdDev1", "Mode1", "Min1", "Max1")] are not found in 'data'. Would you possibly be able to help with this? :) Thanks

[@nicolavaratharajah2230]

This is honestly incredible work! is there any way to obtain the raw data? E.g. if I didn't want the mean A value for CIELAB but instead all the A values of all the pixels? Thank you

[@alecpalmiotti2984]

thanks for this informative video. I'm a little confused as to why/how saturation and brightness values are above 100 in your HSB data set. are brightness and saturation not measured on a scale from 1-100? if not, what is the scale, so I can convert this number in to a percent? thanks again for your help!

[@MrSierrarx]

Very nice script, just one question. Is it possible to do this measurements in the different particles within the same picture? let me explain, I'm trying to measure the emission of seedlings in petri dishes and will be very useful if I can do it by plant

[@jing107]

Hi! I ran into some problems with the program mistaking the object for the background and I don't know if you would be able to help me. When I run the program at first it looks fine since it shows the background black but when looking at the measured images at the end it shows the other way around, with the object as black. I don't know if the final image is wrong or just the entire process. I tried both automatic and customized thresholding. Thanks

[@CFStrock]

Hi Jing- Definitely sounds like a thresholding issue/incorrect thresholding settings for your images. The summary stats from all the channels will reflect the pixels that are not blacked-out in your final images that are saved back to your working directory. I think the solution lies in finding the correct thresholding values for your specific image set.

[@jing107]

​@@CFStrock Hi. I don't know why but I couldn't find the correct thresholding values so the object doesn't go black. In the end, since I'm working with beakers which have an easy shape, I decided to crop them by hand. Thank you for answering though!

[@francisakanyibah8948]

hi, please how do I download the whole color macro_all component?

[@CFStrock]

available here: doi.org/10.5281/zenodo.5595203