I know about software, that estimates the reverse transform:

http://www.noah.org/wiki/Wavelength_to_RGB_in_Python

Possibly you can derive an inverted algorithm from that ... but that's for sure not really accurate

I don't think this problem has a single solution.  For instance, if a pixel is 'brown' (a specific mixture of R G and B values), there is no single wavelength of light which looks 'brown' - the colour of the pixel derives from a mixture of wavelengths.  I think this may be true of every RGB colour to a greater or lesser degree.

I may be wrong, but my thought is that there are several different ways to make the same 'brown' colour by mixing different wavelengths of light, so it is not possible to convert 'brown' into a single unique mixture of wavelengths.

Hello!

As far as I know, it's not possible to convert RGB values to one specific wavelength. So, I recommend you convert these RGB values to another color system: Lab / Chroma (C) and Hue angle (H). This color system is helpful to compare color changes in leaves or fruit (maturation).

Thanks  a lot for all your contribution. I found a very interesting paper about how to modify the IDS camera to be able to capture the wavelength from RGB. Please see the attachment.