Jump to content
UltravioletPhotography

Recommended Posts

One of my projects this year is to try some full-colour IR work, by taking 3 shots with different band-pass filters and then using those as colour channels as follows:

 

Blue Channel: CWL = 750nm approx Filter = R72 + Midwest Optical BP735

Green Channel: CWL = 850nm aprox Filter = Midwest Optical BN 850

Red Channel: CWL = 1000nm approx. Filter = MidWest Optical LP1000

 

I'm waiting for the sun to return to try this out on landscapes - my initial trials on things like flowers and other stuff indoors has not found much by way of colour across this part of the spectrum unless I really wind up the saturation.

 

But there is one area where I have found colour - the sky comes out a lovely blue. This is illustrated nicely in the following photograph, taken in December - saturation has not been increased. there is some colour banding in the fence shadow at bottom-right and on the tree just to the left of the main tree - this is caused by the shadows moving between the three colour separation exposures. I also took some visible shots, which looked quite similar apart from the colour of the grass!

 

Camera was a full-spectrum modified Sony A6000, with a Canon EF 28mm / f2.8 lens.

 

post-245-0-09740700-1584984096.jpg

Link to comment

Just like you did in UV...

 

That blue sky is really cool, that's what made me think about the possibility of seeing stars in daytime in infrared. Basically only your "IR-blue" channel has some signal from the sky.

Link to comment
Andy should give it a try in SWIR (outdoors). I would really like to see a human using, say, 1350 nm, 1450 nm and 1550 nm. Maybe we have blue skin and yellow hair. People and water are the most interesting things to photograph in SWIR.
Link to comment
Are these bandpass or longpass filters? If the latter, it might be appropriate to perform subtractions to get the bandpass signals for the first two channels.
Link to comment

The blue and green channel filters are bandpass.

 

The red channel filter is a longpass, but is in effect bandpass because the sensitivity of a CMOS sensor ends at around 1120 nm. The graph below shows the transmission patterns taking into account typical CMOS sensitivity - the peaks for the three curves have been equalised.

 

post-245-0-05099900-1585044552.jpg

Link to comment

I like it Bernard. I will also have to give that a try.

I have a Lp700nm fiilter that I will cut with a leakyish BW486 filter. This should give me 680/690 to 760/780.

A got a BN850, for the 800nm band like you.

Also have a Lp1000, that os really a 950nm filter. Similar to yours.

So just my 700nm bandpass will cut earlier.

 

If I have a chance may try it an a subject in the backyard that I really like in these odd times.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...