What is blue and what is purple?

[Ann]

To me, the question of what is blue and what is purple is endlessly fascinating. Since the subject has just been discussed regarding an image of the eclipsed Moon, I thought I'd post a few pictures which, to me, show the difference between blue and purple.

Click to view full size image

Purple. Photo: Gerard Stafleu.

Click to view full size image

Blue. Photo: Gerard Stafleu.

Click to view full size image

Blue. Photo: Normf. See this link:http://www.redbubble.com/people/normf/a ... punctulata

Click to view full size image

Red, blue, green and purple. "Bluish plums" by Ben Schonzeit (b. 1942).

Ann

[Chris Peterson]

To me, the question of what is blue and what is purple is endlessly fascinating.

I am a wildflower photographer, trying to document all our local species. I have an online album with the images sorted by color. It's an ongoing challenge to properly categorize those colors. The biggest problem is blue and purple (although occasionally there are saturated pinks that leak into those colors, as well). The problem stems from multiple factors: the flower colors themselves vary somewhat depending on conditions and genetics, the camera and processing introduce color errors, and the eye itself has the poorest color discrimination in the blue-violet range.

Ann, you might be interested in scanning the flower images that I've sorted into blue or purple. Most are pretty good, but some definitely look out of place. I hesitate to move them, however, as the ones I see (visually) in the wild often seem better as I've labeled them, despite how the images appear. Some, like this penstemon, even have a color gradient that seems to span blue and purple.

It's a little bugaboo that grates against my mild OCD need to cubbyhole things <g>.

[Ann]

Very interesting, Chris, and very beautiful.

I agree with what you said, that some flowers that look blue to the eye will often look purple in photos. A lot of factors influence the apparent color of a flower in pictures.

This, however, is what I think about your blue and purple flowers:

Birdbill Dayflower: Incredibly, amazingly blue! Flowers rarely come this wonderful shade of blue.

Mountain Gentian, Narrow Leaved Penstemon, Lanceleaf Chiming Bells and Lodgepole Lupine: These are all very obviously blue flowers.

Tall Chiming Bells, False Forget-Me-Not and "Unknown": These are pale blue flowers, but they are definitely blue all the same.

American Speedwell: This one doesn't look really blue to me. On the other hand, it belongs to the Veronica family, and they are very often blue. The flower could be slightly bluer than it looks here, in which case it would "cross the line from purple to blue".

Low Penstemon: Like you said, Chris, this one shows both blue and purple highlights.

Pleated Gentian, Alpine Vetch: These look very purple to me, particularly the Alpine Vetch. Are you sure that the Pleated Gentian has opened its flowers? Gentians are often blue.

As for your purple flowers, the Early Purple Vetch shows both blue and purple highlights. Your purple "unknown" also looks like a mixture of purple and blue.

All in all, Chris, thanks for showing us your flowers! Very beautiful!

Ann

[rstevenson]

One of the problems with photographing blue flowers, something I've tried to do many times and with little success, is that a lot of what we respond to as an atractive blue in a flower seems to be triggered in our vision by energy that's close to the ultraviolet end of the spectrum. And film (I haven't tried much of this digitally) is even less likely to record that end of the energy spectrum than our eyes, so we end up with disappointing images.

Here's a digital example from my garden. These blasted things are almost glaringly bright blue in the sun, a wispy purple in the evening shade, and impossible to get right any time of the day.

Rob

[Chris Peterson]

One of the problems with photographing blue flowers, something I've tried to do many times and with little success, is that a lot of what we respond to as an atractive blue in a flower seems to be triggered in our vision by energy that's close to the ultraviolet end of the spectrum. And film (I haven't tried much of this digitally) is even less likely to record that end of the energy spectrum than our eyes, so we end up with disappointing images.

Film, of course, is highly limiting. But most digital cameras these days actually have quite decent blue response. The trick with flowers is avoiding saturation on any one of the channels (you can see extreme saturation in the image you attached, where the petals look white). Saturation in a single channel is not so obvious, but radically alters the hue or saturation. What is very helpful is a camera that can display the full color histogram for an image, since this makes channel saturation obvious. For flowers shot in sunlight, I normally adjust the exposure compensation down by at least one stop (which most cameras allow you to do). This effectively underexposes the image, but avoids saturation. The colors can then be adjusted in Photoshop or the like afterwards to approximate the actual appearance. Shooting in RAW mode, where you have more than 8 bits per color channel, is quite useful when shooting underexposures.

If you're serious about getting good pictures, avoiding direct sunlight is also important. I often use a diffuser umbrella, which lets lots of filtered light through, but avoids the huge dynamic range of highlight and shadow.

[geckzilla]

Chris, have you been taking photos in RAW format? Photoshop just has an incredible RAW import tool that lets you adjust the photo to something closer to what you see with your eyes. I mean, sure, human memory isn't very good for replicating color, but it's got to be better than letting the camera decide for itself.

edit: Yes, I did read the whole thread after writing this post...

[Chris Peterson]

Chris, have you been taking photos in RAW format? Photoshop just has an incredible RAW import tool that lets you adjust the photo to something closer to what you see with your eyes. I mean, sure, human memory isn't very good for replicating color, but it's got to be better than letting the camera decide for itself.

edit: Yes, I did read the whole thread after writing this post...

I have tried several RAW processing tools, and do like best the one in Photoshop (CS5), so that's all I use these days.

For flowers that can hold up to it, I sometimes bring them back and have them on the desk while I'm processing images. My memory for color is pretty good, but you can't beat an actual example.

However, for most wildflowers, the natural variation that occurs between specimens is the biggest variable, not my imaging method or processing skills.

[geckzilla]

Well, I guess you need a tag cloud to sort your flowers instead of strict categories. Chris Flower Gallery 2.0 future upgrade?

[Chris Peterson]

Well, I guess you need a tag cloud to sort your flowers instead of strict categories. Chris Flower Gallery 2.0 future upgrade?

I could imagine some kind of color picker tool, where the reader selects a color on a scale, and then the system displays flowers based on that selection. Might be a fun project...

[alter-ego]

I am a wildflower photographer, trying to document all our local species. I have an online album with the images sorted by color.
....
It's a little bugaboo that grates against my mild OCD need to cubbyhole things <g>.

Chris,
Very nice - Thanks for sharing.

[Ann]

About thirty years ago, I was obsessed with a flower whose color seems to straddle the border between blue and purple. When this "border-color" just made it into the realm of blue, I thought it was absolutely fantastic. I tried to photograph the color and get it just so, but I must say I failed. Here, however, is a picture of this flower, whose Latin name is Hepatica Nobilis, by someone who knew how to do it:

Click to view full size image

I have been unable to find out who took the photo. But to me, the color of this flower seems to glow from within, and I find it incredibly lovely.

Ann

[Ann]

It made me nervous that I couldn't make the Hepatica Nobilis images show up, so I'll make another attempt with another blue denizen of the Scandinavian biosphere, the blue-shell mussel. These, however, are oyster shells, I think:

Click to view full size image

Photo: Biopix/JC Schou.

More ordinary, very common blue-shell mussels look like this:

Click to view full size image

Photo: Biopix/N Sloth.

It's almost impossible to find blue or even bluish rocks in Scandinavia. If you walk on a sandy beach, you will find pebbles and grains of sand that are light gray, dark gray, black, almost white, yellowish and pink. And then you will find these blue mussel shells. I really wonder what it is about them that makes them blue.

To me, the realization that some things were really blue "up there" was one of the motivations for me to take an interest in astronomy. I hasten to add, however, that it was not the prime reason for my interest, and it was not the original reason, either.

Ann

[rstevenson]

Click to play embedded YouTube video.

Rob

[geckzilla]

Oh, good, now I finally know what Röyksopp's version of that song is saying. I admit it, I had no idea that was the original and not Röyksopp's. This is how things get lost and mixed up.

[Star*Hopper]

Being a watercolorist, I tend to think strongly in terms of hue. Aside from photography's ability (especially in the digital age) to accurately record and reproduce them, I don't have a bit of realized difficulty determining which is which. I have more problem with where purple becomes violet.