Color Theory

My latest experiment has been with photography; in taking images and pixelating them until you get large blocks of color. 

I started doing this to try and deduce the color spectrum of each picture; to use each block as a swatch of color. My thinking was that by making bigger pixels, you forced the image to generalize the colors, and create interesting harmonies.

After some experimenting; I started to apply these color harmonies to simple graphics I made to see what came of it. This is in a sense my own variation of websites like ColourLovers, and Kuler; it's basically a way of seeing colors that exist in your photographs. I started to take these colors and apply them to a see if I could arrange interesting harmonies:

1. Orange and Purple Canyons

  2. Beige Skyline

3. Glowing Sunset 

4. Beach Blues

What I found is that these colors do create a nice harmony. Generally speaking; I have found that landscapes work best for this technique. They provide gradients that are naturally occurring, and have a certain visceral appeal. 

But why; why do these colors work?

 From a scientific standpoint, my theory is that landscape photographs allow colors to 'blend' in the atmosphere. Another possible theory is that these harmonies work not because of scientific analysis of the color spectrum,  but because of an aesthetic 'preconditioning'. Our brains may be hardwired to understand these arrangements as harmonious; as the landscape color palettes have been the background of our lives, and our ancestors lives.

To understand more, I used a color spectrum tool to see where these colors sit on a color wheel, and see if there was any logic I could deduce:

 1. Orange and Purple Canyons:
Whats interesting about this color spectrum is that originally I had thought that they were going to be a complimentary color set (as suggested from my title for the picture). Instead, they were a sort of broad spectrum analogous. When you see the colors I ended up using in the composition (to the right) one might think that these colors were chosen at random. It is not until you see the full pictures palette (to the left) that you see that these are in some way colors that are associated.

2. Beige Skyline:  

The general trend is that these colors fit in one quadrant of the color wheel, making them analogous colors. 

3. Glowing Sunset-

Initially I thought I had made an error, that there were 2 'outliers' to the analogous colors spectrum, but when I did a mapping of the whole picture I found that there were 2 distinct groups of color; and indeed one outlier (it was a grey i used for a small detail, accidentally brought over from the previous comp)

4. Beach Blues-

Again, the initial color selection did not fully demonstrate what was happening. After seeing the whole color spectrum of the original picture (right) you can see that the colors selected make a general line of complimentary colors, varying from muted tones to more saturated. 

In conclusion:

My color spectrum test has shown that colors of a relative spectrum in Analogous and complimentary colors provide a good basis for a color palette, but more interestingly that these colors lie on a line of saturated to desaturated tones. While I can not scientifically conclude my theory of colors feeling right based on our preconditioned experience seeing them, I do feel that this does have some basis. 

I have enjoyed this little experiment of mine, and I will continue to do color comps to see what else I find. In the meantime, here are some other thoughts on applications for this project beyond color comps:

Backdrop- I have put these as my desktop, cycling through the images is soothing. They aren't busy enough to distract you, but they do have a certain visceral appeal. The arrangements allow us to 'fill in the blank' a little, the way a book makes you imagine what is there in your minds eye. I could see these being used on a larger scale, say as an advertisement; or as a mood setting element in a public space (I like the idea of wrapping  public transport vehicles in them, or a tube station in London.

Graphic Explorations

I've always been interested in pattern making. From my early childhood, I'd spent a lot of my time looking at Mughal carpets, and how the patterns interlock. I've since enjoyed making patterns based on geometric principles:

 I like to think of these crystalline structures as having more than one application. Perhaps a more elegant QR code, or a visualization of data; all sorts of applications.

Recently I started playing around with typeface in a similar manner. I was inspired my an 'S' I saw in a documentary; it was from an old Newspaper clipping. I enjoyed the serif balance, and I started to play with the shape. 

This was the original clipping. 

After recreating the shape in Illustrator, I started to play with it's symmetry.

Making it equally balanced; I thought 'why not make it something other than an 'S'? I liked the idea of making it a variation of infinity; I often get sidetracked by these sorts of thoughts.

 I started cutting away the shape, to get new forms

 And adding them again to the original

 What I was trying to do was play/experiment within the natural rhythms of the shapes, and see what I could come up with.

 Potential new forms arise

 As well as classic revisions. This is the official "Section" Symbol

 What started off as a simple interaction of shapes starts to take on multiple new meanings as you start to play with the flows and rhythms of positive and negative space.

 I also started to use the font cuts in more familiar ways, as geometric based patterns. This pattern could be an icon, or even a physical item. I could see it being cut out of metal and used as a decorative bowl. 

 What I enjoy about this process is that it leads me to conjure up all sorts of new directions of thought. Thinking Laterally helps open up my mind to all sorts of creative potentials. 

By this point I was appreciating the font in a whole new way; thinking of how its rhythms had geometric principles behind them.

This grid was breakthrough for me. By applying a geometric shape to the arrangement, I was able to recombine it using geometric principles. 

 On top, this set was placed sequentially, but I was not satisfied with the orientation of the blocks. By looking at the triangles governing the orientation, I was able to see a different way of organizing them, and made a hexagonal arrangement.

 This as can lead to all sorts of denser patterns. I ended up making a 'wallpaper' of pattern before heading in a different direction.

I started to formulate more geometric principles; that this triangular shape could be used in sequence. 

I went back and started recycling the shapes again, using the concept of geometric grids

 After this pattern, I started to see the positive/negative space in a new way. This looks like a graphical interpretation of an orchid to me. Then I saw something else, a mustache. 

 From this 'Mustache' I started to play again, making two variations before finally creating a new shape, the red lips. I made the lips based on what 'felt' like it would fit in the visual language; but it got me thinking. Why can't my software see the rhythms of the shapes, and start to create a grid around it? You could use all sorts of mathematical principles (in a visual, non-threatening way) to help generate all sorts of variations.

This notion of geometric rhythms lead me to one last destination; Fractals. Back to a Mandelbrot- inspired set, I recombined the shapes in a way that felt like I was following the curvature of the shapes; but I would be curious to see if I could make this happen in actual coding. 

I hope you can see now how this sort of visual playtime conjures up all sorts of new thoughts for me. It's a meditation for me, as well as a tool for generating new content. I'm interested to hear your thoughts and comments. Thanks!