Biomorphic Imagery
Some digital examples
by Stuart Smith
After generating thousands of digital images with my app, I came to see that they were falling into a small number of categories. Many images were purely geometrical compositions of lines and shapes. Another group had colored swaths of parallel, sinuous lines flowing all over the image surface. A third group — the one I want to focus on here — contained “biomorphic” forms, that is, shapes reminiscent of plants, animals, micro-organisms, etc. I’ll present here a series of eight images from this group, with a little commentary about each. But first, a word or two about the titles I’ve given them.
When I set out to make digital images with my computer, I usually don’t have in mind exactly what kinds of images I want. Consequently, I’m not ready with a set of catchy titles to apply to them. For me, naming images comes after they’re completed. Only then do I study them to see what titles I could give them. I might try, for example, to come up with titles that are clever or witty, or I might choose titles that suggest to the viewer a way of looking at the images. Or, given the typical nature of my digital images, I might merely give them titles like “Abstract 1”, “Abstract 2”, etc. — labels for quick reference in a file or catalog of images.
This image is a modification of one produced by the program that was the precursor to my digital art app. Fig. 2 below is the original source image. The stylized “fish” shape there has been altered in several different ways and deployed in different orientations in Fig. 1.
After studying this image, I decided to call it “Dolphin Dance,” which is also the name of a jazz standard tune by pianist Herbie Hancock. This type of circularly symmetrical image was produced by the precursor to my art app. With that program I could specify the order of symmetry, among other image attributes. This image has 5-fold rotational symmetry.
This image was made by an early version of my current art app. This was the first version to both make images like “Dolphin Dance” and then transform them. In other words, images of this type were created by “end-to-end” automation: everything you see in the image was done by a single run of a single program.
This pair of images was produced during a run of an early version of my art app. When I started the run I didn’t know it was going to generate anything like these jellyfish-like objects. I was mainly interested to see what the app would do with the multi-point stars routinely produced by the algorithm used by the precursor to my art app and that was later incorporated into the app.
The mildly creepy effect here was created by applying the GIMP/G’Mic-Qt “barbouillage” filter to an image generated by the app. I occasionally allow myself one step of post-processing by an external program like GIMP or Krita. Typically this is done to sharpen an image or make the colors more vivid. Here I wanted the “paint daub” effect of the barbouillage filter.
Figures 6, 7, and 8 have a more floral or vegetative appearance than Figs. 1 and 4. They are all modifications of symmetrical images like Fig. 2 above. The app doesn’t have inputs for “fishiness” or “leafiness” or any other qualitative descriptor. You put in certain numeric values and then wait to see what the app does with them. Of course, experimentation and the experience gained over time eventually enable one to make accurate guesses at what the app will produce.
This image was slightly re-touched with a GIMP filter to bring out the colors of the “flowers.”
My app generated this image by applying a single transformation to an image like Fig. 2 above. The transformation simply reinterpreted the cartesian (x,y) coordinates of each pixel as (angle, radius) in polar coordinates and then moved the pixels to their new locations. A GIMP smoothing filter was applied at the end to give a more painterly appearance.
© Copyright Stuart Smith 2023. All rights reserved
