Tezos

Mineral Composition #74

Description

Since its conception, this project has been modified over the past seven months by adding new elements and techniques to improve its overall aesthetic and impact, as well as my own skills as a generative artist. Mineral Composition is an artwork inspired by the natural beauty and complexity of minerals: crystalline and ordered arranges of ionic elements. As an admirer of nature, I have been particularly drawn to the way that just a simple atomic rearrangement in the presence of certain metals can result in a merge of vibrant colors, brightness, and contrast. My deep appreciation for this subject grew even stronger during my master's degree in Materials Science, which led me to search for a way to incorporate this knowledge into generative art. The idea to use k-means clustering to play with mineral colors came to me during a casual chat with a very clever friend (Estienne). They suggested trying this machine learning algorithm, which involves grouping similar colors into clusters and selecting representative colors from each cluster. I was immediately intrigued by the idea and decided to give it a try. I gathered a selection of 20 mineral images and used them as samples for the clustering system. The results were nothing short of stunning - the algorithm was able to identify and extract a wide range of colors from the minerals and would have been difficult to replicate using traditional methods. I knew that this approach to color selection would be a game-changer for my artwork, and I was eager to put it into action. Using the colors extracted through k-means clustering, I was able to create a truly one-of-a-kind color palette for each minted generative token one that perfectly captured the essence of minerals in all their vibrant and varied hues. And the best thing is even with repeated picked minerals, the palette won't never be the same. Additionally, the FF-based drawing can vary in three models and, when combined with various texture elements such as irregular grids, geometrical patterns, and stained canvases, provides the viewer with a pleasing experience by allowing them to examine every single detail across the entire canvas. "The main intention was to create a piece that would have a fine art feel and be suitable for printing at high resolutions to be displayed as part of an environment's decoration. To achieve this, I took great care in crafting every detail of this work, ensuring that it could be reproduced and enjoyed in all its detail and intricacy by collectors. Overall, Mineral Composition is an art piece that combines multiple basic and medium techniques of creative coding and machine learning. It is a project that captures the essence of minerals in a completely innovative way and will be a welcome addition to any art collection. _______________________________ Variety of editions is guaranteed by several features: * Backgrounds with paper colors or with a picked color from the mineral palette. * 20 real minerals as color samples to create infinite palettes by k-means clustering. * Controlled variety of scale; * Controlled variety of particle number; * Animation time; * Symmetrical mode (mirror mode) possibility (15%); * Three drawing models for FF filaments (opening and closing screws or spring screws. _______________________________ Interactive instructions: After finishing the animation, press: * [L] to check your mineral and palette info; * [K] to save your .png file in screen size (smaller); * [n = 2 to 5] to save your .png file in A[n] ISO216 printable sizes; Depending on the RAM and CPU usage, it takes from a few seconds to a minute to render the image. Advice: the system needs enough free RAM for rendering A2 sized images. It was tested on Chrome, Brave, Firefox, Opera, Edge and Safari, but please make sure it works on your device before. _______________________________ Prints can be ordered. Shipping costs to discuss. DM me on twitter for more information! _______________________________ Creator: Kukuti Made with p5.js January 2023 License: CC NY-SA 4.0