ImageJ Tutorial 1 - Measure Leaf Disease Area & Lesion Counts

[@abigailwolmarans9887]

This tutorial saved my life! Thank you!!

[@mpumigama2587]

Very helpful tutorial. Thank you!🤗

[@UFHortLab]

You’re very welcome 😊

[@SHUBHAMSINGH-gh6ln]

Hi Thanks for this tutorial. Can you please explain how will I calculate the skinsett of potato using this software.. like potato is round so will this software work in that aspect?

[@appiahkubi1203]

Hello please I'm working on my project and I'm to calculate the area that has been consumed by insect herbivores on leaf samples using imagej ..but I'm struggling with it .. please can help..I have multiple leafs of one species on the sheet in one image

[@hengsotheareak]

Hello dear, when I make binary, leave become white and background is black and it will analyse the black background. How can i switch it? Thank you for your reply.

[@wizardlegend07]

Thanks for the tutorial, please how can I get the picture you use in the tutorial ?

[@gurjotsingh2380]

how do we measure the length without manual scale

[@h.p.weerasuriya7278]

Thanks for the video...when selected binary, the infected area is shown in white, and the rest of the leaf in black, how to make it vise versa? Hope a reply soon:)

[@lizfuentes694]

Hello, go to 'edit', then to invert

[@SHUBHAMSINGH-gh6ln]

Excellent tutorial Suppose you have a 1000 picture to work with, how will you manage your data in that case , like for each sample we need to save 3 photos? Can you please give some tips about data management

[@pacifictrufflegrowers]

Why are you saving each image as a TIFF instead of JPEG?

[@basilelsioufy6834]

thank you so much from where can i download the photo of blackberry leaves

[@Mkayzeee]

Short but informative video!!! Thanks...You answered all my questions.... But on the colour threshold, how did you know the range that works for ur samples?

[@UFHortLab]

Hello, Tumisang! We are very glad you enjoyed the video! The simple answer is you have to judge the best threshold by adjusting the threshold ranges and checking the effect the changes to the threshold have visually until you are satisfied. Threshold ranges may be the same for all the images in a set, or it may very between images/select images in a set. The best way to determine the threshold ranges which work best for your samples is to open your image and then to open Image > Adjust > Color Threshold.... From there, you can determine by visual confirmation what Hue, Saturation, and Brightness ranges will capture the desired pixels within your image. If thresholds are similar for all images within a set, once you have determined the ranges once, you can run a macro using Process > Batch > Macro... to select those ranges for the whole image set. You can set the macro on the ImageJ Fiji version by going to Plugins > Macros > Record. Select the desired ranges for your image and clicking "Select" at the bottom of the Threshold Color window and go to Process > Binary > Make Binary. Then copy the text in the Record window and paste it into the Process > Batch > Macro... window, select input and output folders, and click "Process." You will need to check the quality of the thresholding in the output folder, and if thresholding quality is variability between images, it may be necessary to go back to the original images and adjust ranges for individual images.

[@Mkayzeee]

@@UFHortLab Thank you soo much! I’ll comment again when I’ve tried the batch method because I have a thousand photos to work on

[@SHUBHAMSINGH-gh6ln]

@@Mkayzeee Hi I hope you are doing well I will be working with this software for calculating the skinsett of potato. Since you have worked a lot on it , can I connect with you? I will be joining Colorado State University for this fall intake. Have a great day

[@Mkayzeee]

@@SHUBHAMSINGH-gh6ln unfortunately I’m not joining that program.

[@SHUBHAMSINGH-gh6ln]

@@Mkayzeee ok Thanks for your response

[@MsDX1990]

Is it necessary to set scale even for measuring area of diseased area?

[@UFHortLab]

Sorry for this late reply! Set scale is necessary to determine the diseased area in cm2. But if you are only trying to determine the percent diseased area, no, set scale is not necessary.

[@MKay-ys1hz]

Why is my area measurement for the total leaf and the diseased part the same? What am I doing wrong?

[@Finn-hk8qn]

How do you know what values of hue, saturation and brightness should be chosen in color threshold? I tried to use the numbers you show but it doesnt work with the leaf I photographed. I tried to adjust the values but no matter the combination, the whole lamina won't color

[@UFHortLab]

Hello, Finn! Sometimes finding good hue, saturation, and brightness values takes adjusting those values back and forth to find the best values. If it is impossible to capture the entire lamina, it may be that lighting (e.g., not enough light at the beginning of the day), background color (if similar to leaf color, causes problems), or image quality (e.g., low resolution) need to be improved. Also, if there are lesions on the leaf, they may be a different color than the leaf, making it difficult to capture the area of the entire leaf. One option in that case is to capture the healthy leaf area, capture the diseased leaf area, and combine the two values.

[@Finn-hk8qn]

@@UFHortLab Thank you for your reply. I've started taking photos using flashlight and that'd solved the problem. However the measurement itself doesn't seem to be working properly, I guess the issue is scaling. When taking photos of the same leaf and measuring area in both the value between them differs up to 2 cm even. Each leaf is photographed on white sheet of paper with a ruler beside it. Im setting the scale for every photo because I don't have a tripod or stand of any kind so I assume a centimeter in one picture may represent a different number of pixels than it does in the other. Do you have any advice regarding this problem?

[@UFHortLab]

@@Finn-hk8qn A couple things which could help would be the make sure all photos are taken centered above and fully perpendicular to the leaf. Also, if the leaf is big enough that it can move a lot, taping the leaf to the sheet of paper so it is totally flat could help! We do that sometimes here too with big leaves.

[@UFHortLab]

Also, it is good practice to place the ruler as close to the leaf as possible! That will help make sure the scale of the ruler and the scale of the leaf are the same.

[@mahrukhbukhari4791]

very informative video. how can wevopen file as preview???

[@UFHortLab]

Hello, Mahrukh! I am delighted you enjoyed the video! Can I ask a couple questions so I can answer your question fully? 1) Which file would you like to open? 2) Would you like to open that file just to view its contents, open the file in the Preview program, or what specifically can I help you with there? If the goal is to see the file contents prior to editing, as long as you have a copy of the image at the stage you want saved onto your computer, you can drag and drop the image icon onto the ImageJ graphical user interface and the ImageJ program will open it. You can also open the same image multiple times in ImageJ and only edit one opened copy. In order to view it in the Preview application, you can right click the image icon on your computer, go to "open with" and select Preview.

[@soriyarin718]

that software can't be installed on my Mac

[@UFHortLab]

Hello, Soriya! What version of the Mac processing system are you using? (To find what version of the processing system you are using, click on apple icon in the upper left, and click "About This Mac" from the dropdown menu. The first two lines of the popup should say the common name and the code. For example, mine is macOS Catalina, version 10.15.4.) It is possible if the version is very old, it may not be compatible with the ImageJ/Fiji software, which is constantly being updated. Also, I know my Mac sometimes had troubles when the storage space is almost full! Do you have plenty of space on your computer?

[@lokbahadurpun986]

I don't know, why am I getting inaccurate area measurement results although I followed the same procedure. I get leaf area value far more than the actual range

[@UFHortLab]

Hello, Lok! Sorry you are being plagued by an area too big! I feel like that is one of the most common things which can occur since there are a couple ways ImageJ will take an area different than the black pixel area: Sometimes after converting an image to binary if you do something which knocks it out of binary then measuring the area takes the whole area of the image. The best thing to do in that case is to make the image binary (Process > Binary > Make Binary) right before measuring (Control+M). If that still doesn't fix the problem, it may be your Control+M is measuring the wrong measurements. To correct that, go to "Analyze > Set Measurements" and to make sure to select "Area" and "Limit to threshold." Control+M will measure the area of the entire image or the area of your current selection if "Limit to threshold" is not selected.

[@mafaldatendeiro298]

@@UFHortLab Thanks for the explanation I was having the exact same problem, and it worked with your first suggestion!

[@0YazZmaN0]

6:00

[@CamilaHelenMendoncaRodrigues]

thank you for saving my life hahaha

[@flame2903]

90% nosense 10% useful information. U need to top working at LAB, instead get more education.

[@mbkhan01]

Boring procedure

[@abdelheqzouaimia5256]

Hi, thank you for this helpful tutorial, could you give me your e-mail, all of this didn't work for my sample and i need your help, best regards

[@UFHortLab]

Hello, Abdelheq! I am delighted you like the video! If you need help, our contacts can be found on the corresponding UF EDIS article+UF directory!: edis.ifas.ufl.edu/publication/HS1382