How can I read pixel per wavelength and measure it

I’m a very beginner with imageJ and FiJi and I want to analyse the image per reading pixel per some wavelength and measure it. also I want to visualize the pixel within the image. Which pluggins are necessary or helpful?

Hi @claudia, I’m afraid I don’t fully understand what you’re trying to do. As you’re speaking about wavelengths, this usually means your acquisition was done in a way that different wavelengths are detected separately, e.g. with the filters in a fluorescent microscope.

In that case, they should have been stored in separate channels by the acquisition software, and you can simply analyze them one by one.

Otherwise, could you please elaborate a bit on your scenario, maybe even providing example images?

Hi, thanks for your answer. what I want to do is:
I have a normal image of a melanom, rgb, and every wavelength have its color, I want to examine a few wavelengths or color of some wavelengths within the image. I want to know how many pixel are there in the image with this rgb/color/wavelength. And then I want to draw the pixelareas at these colors/rgb to visualize the result. Sorry for my bad english! Now I try to do this in html5/javascript. It’s not so easy…because I’m a beginner in html5/javascript too. But it must be possible.

Hi @claudia,

I’m actually not quite sure about your reference to HTML5, how is this involved in the process?

However, from the rest of your description it sounds like you want to extract specific wavelengths (“colors”) from a mixed mode acquisition like an RGB image from e.g. histology sections?

Depending on the colors used, this is sometimes not as easy as it might seem, but there is the very nice Colour Deconvolution plugin by @gabriel, which should do what you need. Please note that the plugin is shipped with Fiji by default, so unless you are using plain ImageJ you don’t even have to install it separately.

Again, if you need more detailed instructions I can only encourage you to provide a sample image!


1 Like

Hi, I do not think you can extract accurate wavelength information from an arbitrary rgb image (which really has information on 3 broad channels only). In an rgb image you cannot for example differentiate a real yellow wavelength from the sum of red and green wavelengths.
The colour separation from Ruifrok and Johnston mentioned above (colour deconvolution) is suitable for unmixing subtractive colours (like histological dyes used in bright field microscopy or inks printed on white paper) and I guess might not work well in case of photos of pigmented melanomas.