An IEC-based support system for font design

This paper describes our first trial to build a prototype of support system for font design utilizing the technique of Interactive Evolutionary Computation. The target domain is Japanese Katakana constructed from very simple stroke elements. Some parameters for drawing elements are encoded on the genome of each candidate individual. Starting from the initial population of sixteen individuals with random genes, the user can breed his/her favourite fonts through the graphical user interface based on a framework of Simulated Breeding. Each candidate is shown in the sub-window with a sample word arbitrarily set up by the user. We certified that it effectively works to produce a type of expressive glyphs.