Perlin noise makes use of a partially random series of numbers which are computed into an image. Two- and three-dimensional images created this way are intended to mimic natural objects such as the sun, clouds, and marble, for example. The concept was created in the mid-1980s by Ken Perlin, a computer science expert and college professor still as of 2007. It provides relatively smooth random functions compared to the capabilities of typical programming languages. Control for small scale as well as large-sized elements are possible.
Graphics rendering programs make use of Perlin noise. At the programming level, the simulation noise is computed using mathematical formulas. These complex formulas are used to generate graphics in one, two, or three dimensions. Various parameters are numerically defined in an equation. The number representing the noise value, along with the sum of other values, results in a graphical line in the first dimension.
In two dimensions, a computer-generated visual effect uses numerical values less than an image’s resolution, particularly a gray-scale image. Perlin noise can also be visualized in three dimensions. Textures of objects on a computer screen can be analyzed beyond just one side and at any point on the surface. These points can be moved to produce a rotating image, and various functions can be computed to change the image texture. This helps in the imaging of rectangular images and translating them to spherical representations.
Perlin noise can be used in the creative process using the same methods. It is used in animation, as the same principles can be applied to animated characters so their motion appears smooth. Realistic looking clouds as well as terrain can also be created from both a ground perspective and from above. Color and texture can also be added, so Perlin noise is beneficial for creating detailed simulations and images that are either abstract or realistic.
Computer programs control the value noise, so the user does not need to understand the mathematical concepts involved. One program uses an algorithm for choosing an input point, picking a gradient vector for nearby points, and calculating additional gradients. Calculations using coordinates then derive the image’s scale, and patterns repeating into smaller variations can be created to simulate the nature of a fractal landscape. Changing the scale of such patterns means making use of a numerical scaling feature called octaves. Various computer programs help to render detailed images based on numerical calculations that would take too long for a person to perform manually.
