Daffodils

Well, over a half metre of snow fell in the space of two days last week, and we’ve finally (more or less) cleaned up the roads.  Of course my snowblower picked that moment to break, so I was stuck with the shovel. I took some good photos of the snow dump in 3D, and was thinking of posting them… but I’m sick of snow, and don’t want to think about it for a few more days. Instead, I’m thinking ahead about spring.

Daffodils(don’t ask me which species) are one of the first flowers to pop up at my parents’ home; usually they are in bloom by the end of April.  When they bloomed in 2013, I had only been taking infrared photos for a few months, and had just completed building my first short pass filter. It was a perfect opportunity to try out my filter. Unfortunately, my full spectrum camera at the time, a modified Panasonic Lumix DMC-SZ7, wasn’t capable of looking deep into the ultraviolet range, as I figured out later. Live and learn.

By the beginning of May, I had a beautiful flower to test out my camera filters. In the visible spectrum(using an unmodified Canon T3 Rebel), the daffodil looked like this:

This is the only photo in the shade- after this photo was taken, the sun came out. I switched to my full spectrum camera. With no filter on it restricting the light entering the sensor, the daffodil looked like this:

The flower itself isn’t very different in full spectrum. The stem and the other plants, however, have shifted colour significantly- the infrared light bleeding in is interpreted by the camera sensors as red light.

Switching to an infrared filter(a Zomei 760 nm filter), the infrared light reflected by the vegetation is much more obvious:

As expected, the vegetation is much lighter. Interestingly, despite the lightness in the infrared range, the daffodil’s flower is still brighter than the surrounding plants.

Finally, it was time to try my short pass filter out. The filter was designed to allow light in the, violet, and ultraviolet range, with a bit of blue light as well.

I had the same problem as I do now with the short pass filters I use: getting a proper white balance is (almost) impossible. My Panasonic Lumix allowed me to set a white balance. However, it was a primarily a point-and-shoot camera, with few manual controls, so it was even harder to properly set the photo up than it is with my DSLR now. Most of the photo is violet, with overexposed sections being yellow.

That said, I found the photo very exciting. The yellow daffodil, which appeared to my eyes as one colour, showed a significant colour variation with the short pass filter. The outer petals were much lighter than the inner ‘cone’ of petals(apparently called the cup or corona), and the stigma in the centre is brighter than either. It was my first attempt at Ultraviolet photography, and was a success…

…Or, at least that was what I thought at the time. As I mentioned, my Lumix wasn’t the best at ultraviolet photos, as the exposure time maxed out at one second. As I gained more experience with full spectrum photography, I started to compare my earlier photos in more detail. This is when I started to look at the colour channels in my photographs separately. Looking back at my earlier photos, I broke the visible spectrum photo(the first photo in this post) into the blue, green and red channels.

The daffodil has the same features in the blue channel as it does with the short pass filter. I prefer the filter, as I find the colour interesting, but that’s irrelevant. The channels are, almost by definition, monochromatic. It seems strange to me that this variation existed. The daffodil looks like an even shade of yellow to the naked eye; the variation around the blue wavelength isn’t visible to me.

What I think is the case is that there are the different amounts of blue in the flower. However, the flower is mostly yellow, and the small amount of blue in the flower, even where it is most noticeable, is lost among the large amounts of yellow. The blue channel is very dark in this photo- not much light is reflected from this wavelength. I was able to confirm this hypothesis by looking at the flower with the short pass filter in front of my eye- removing the longer wavelength colours made the variation more apparent. Interestingly, the veins of the flower are also slightly visible in the petals. The short pass filter might display them as well, but due to the long exposure, the flower is slightly blurry.

There isn’t much to say about the green channel- in this range, the flower is an even colour, with very little variation.

In the red channel, the colour is similar as well. However, The corona of the flower is slightly brighter than the outer petals. This only seems to be the case in the red range of the spectrum- in near-infrared, the corona and petals look to be the same shade again.

If I get a chance this spring, I’ll probably try to see how a daffodil looks in true ultraviolet. I’ll probably post the results after that. Until then, I’ll just dream about green grass and vibrant flowers while I continue to shovel the snow.