Colourimetry

[ian]

One thing I've noticed breeding slimes is they vary in colour. Anyone who's got Daywalker and Mazie on the go might have noticed that Dw is a lot more orange. I've also got some paler variants on the go.

I'm wondering - varying levels of one pigment more than one? In the lab the done thing would be to smoosh a few slimes in solvent and put it in a spectrophotometer. I don't have one of those at home. I do have a hell of a lot of cameras though, they're not great - colour cameras exploit the way our eyes work and only measure red green and blue rather than actual wavelengths but I think I can work with that. 

This is what I've got -

The camera's in the lid of a cardboard box, it's an Eos-M with the kit lens and a white LED ring, by keeping all the camera setting the same I'm hoping I'll get repeatability. 

Imagej makes it easy to just read out the average colour of a selected area  using the white card and colour filters as sanity checks. 

I can't read out actual values - domestic cameras aren't really designed for that but I can look at the ratio of the R,G, B channels and at least get a colour signature for a slime. 

The results so far - Orange slimes have more red than yellow ones! Well it's a start - I need a lot more images to see if there are any more subtle differences. 

[Patrick Barry]

a month and a half later?

[ian]

I went off on a tangent an the algorithm is now being tested for grain crop analysis - didn't see that one coming 

[Patrick Barry]

I do hope it proves useful and successful.

And applicable to oats :)

[ian]

Good one! 

The Github link is here -

https://github.com/Scalebar/Cropmeasure

I am starting to results with the slime stuff - but I'm also aiming to publish which means being a little careful with what I put online - I'm not afraid of it being stolen but journals can be arses about "You've already published that ( meaning on twitter,Fb etc ) so you can't use that data"