A way to avoid any mistake is to always add a ruler in the photo (e.g. I always put insect wings on a millimeter paper to take the picture with microscope). As soon as you get e reference in the picture, using the program tpsDIG (http://life.bio.sunysb.edu/morph/soft-dataacq.html) easily allows you to set scale and calculate measurements in mm.

Hello there, thanks for the answer!

I'm using TPSdig as well to have the mm reference using their digital ruler for pixels. I just want to avoid using a ruler in every wing because I know I can just use a conversion number when I have my zoom fixed and the correlation between pixel and mm. So I can use it in every wing.

I suggest to use a morphometry software package for your measurements like ImageJ or ImageProPlus. You can calibrate the images based on your ruler in this software and make direct measurements of your specimens like length, area etc.

If you have your camera at the exact same distance from your objects (the wings) for all pictures, with exactly the same zoom setting and with the object in the same place in the picture (i.e. in the center to avoid distortion), you could use the same conversion factor in all pictures. Of course, you should test this out on several pictures including a ruler to see to that the method is accurate. Note that with this method there is no possibility to adjust focus by changing the distance between the lens and the object. As soon as you do that you need to include a ruler to get the new conversion factor.

I always used Image J. Regards Stefano

As Dimitrios Koureas mentioned, I suggest using specific software for morphometry. I’ve been using Motic Image Plus 2.0 with good results. It’s a friendly software with a lot of options for measuring samples and you can easily calibrate your equipments. It is distributed with Motic digital microscopy items but I guess it can be found in the web.

I agree the rest of people, you should calibrate the image using a ruler and a software like ImageJ or SigmaScan, and then make direct measurements in mm.

Hi Heverton!

Since you use TPSDig anyway, I do not see any problem. You will get a mm-scale for your measurements as soon as you have set the pixel/mm-scale for your first image. You will get a conversion number (scale) for this and all the subsequent images without doing anything. This number will just change if you have to re-scale another image because your resolution or zoom has changed. If not, you just have to do this once and you will get all your linear measurements in mm as far as you do meaurements instead of landmarks. If you do the latter, the procedure is the same anyway.

Cheers

Juergen

Hello Heverton,

I also use tpsDIG to get results in mm.

However to check some results of my measurements I usually make back-calculations from pixels to mm manually.

In your example: 5 mm = 500 pixels, 1 mm = 100 pixels, 1 pixel=0,01 mm

Just multiply the result of your measurement in pixels to 0,01 and you will have your result in millimeters.

It's no matter which zoom you use to make such back-calculations. But if you'd like to use the same index for measurements of many pictures, they must be taken with the same zoom and this index should be calculated for one of these pictures.

If you really are going to use only one magnification, you can figure out the conversion from pixels to mm: then your conversion will be pixels x mm/pixel = mm. But it's easy to calibrate all the magnifications so you have more choice. It's harder if your microscope does not have stops at particular mags, but if it does, just calculate mm/pixel at each of the magnifications and tape a little chart to the scope. If you do this, though, don't forget to keep track of which mag you used; I usually put the mag into the name of the file. And I agree with Joacim, you want to measure more than once to get an idea of the uncertainty. However, if you are doing morphometrics and absolute size is not of interest, you can just do everything in pixels because the differences in size will mostly be factored out, whatever the units, when you do procrustes superimposition.

Wow, lots of answers! Thank you so much, people!!!

I will try it this week and post the results here.

I took all the photos using exactly the same zoom and image conditions to avoid mistakes in the analysis. Also, I'm going to use the 18 landmarks procedure to distinguish my populations and also to calculate other parameters like how reliable my marks are...

All these procedures I already know how to do it, I was only having this trouble converting the pixels to mm...As Dr. Dany mentioned, I can do all the analysis in pixels, but I want the correspondent mm number because it makes more sense when showing the data.

can any one else tell me how we convert pixels height into the cm height ?

I also calibrate the window for each specimen within the software package. I'm using ImageJ KIWI for leaves, which are similar to measurements on insect wings (most of my landmarks are along prominent venation and along the margin). I just make sure I place my ruler in the same portion of my herbarium card before I place it on the flatbed scanner. Best wishes!

Is this helpful? http://www.unitconversion.org/typography/pixels-x-to-millimeters-conversion.html

I always used ImageJ, but the scaling information gets saved with the image file only when saved after scaling, in TIFF format.

Depending on the quality of lens, field of views, positioning of objects in the image, and accuracy requirements, you might like to consider lens distortion calibration for the images/cameras.

https://www.microscopyu.com/tutorials/distortion

Hi! Heverton,

I believe you got help at the time you asked this question. Can you help me, please? I have taken my measurement using different magnifications which were recorded. How can I convert pixels into mm bearing in mind the different magnifications I used? Note, for each of the magnifications, the graticule calibration corresponds to a given pixel.