Lots of people find that, when learning to paint, it’s perfectly possible to follow a technique for sketching the underlying shapes correctly. What is much more difficult is matching the colours in a real scene with mixtures of the paints available in a given palette.
When you play with any image processing tool such as The Gimp (or Photoshop if you have that kind of cash), the ‘eyedropper’ tool can be used to demonstrate that the actual colour extracted from a region of a digital scene is very different from how it looks when located beside its neighbouring regions.
Painting therefore requires a) determination of what the local hue in a part of a scene should be, when extracted from the perceptual influence of the surrounding colours, and b) mixing the correct proportions of basic palette colours to achieve this.
Today’s invention attempts to overcome both these difficulties. First, a digital camera captures an image of the scene which is to form the subject of a painting. The image is blurred slightly and colour quantised and each region, larger than a few pixels, is digitally labeled with its average RGB coding.
These figures are then translated into the nearest equivalent CMYK values at each position in the image. A mechanical dispense device contains replaceable paint tubes in each of these four colours (cyan, magenta yellow and black). When operated, (by eg clicking on the screen of the laptop on which the software is running) the device squeezes out paint in the required proportions. These can then be mixed and applied to the corresponding location in the painting.
After using this system for while, it should be possible to start to mentally link the local colour in the scene more directly with the paint mix required (I’ve just discovered another possible application).
