A CAI (Computer Aided Innovation) tool called AEGIS (Accelerated Evolutionary Graphic Interface System) is being developed and applied to graphic design. The tool, which uses TRIZ 'guided' algorithms, is being tested with graphic design experts. The AEGIS processes initial inputs (initial design ideas) and apply TRIZ based algorithms to them and outputs 'innovative designs'. These designs may further be processed using manual techniques to achieve the final outputs desired by designers. This paper discusses version 6.6 of AEGIS. In this version, Genetic Algorithms (GA's) are being utilised. Genetic algorithms are one of the best ways to solve a problem for which little is known. They are a very general algorithm and so will work well in any search space. All you need to know is what you need the solution to be able to do well, and a genetic algorithm will often be able to create a high quality solution. Genetic algorithms use the principles of selection and evolution to produce several solutions to a given problem. This paper considers the various entities and structures used and developed to implement Genetic Algorithms through which TRIZ guided effects and mutations are applied here to graphic design elements. It attempts to code TRIZ Principles based mutations (function/method parameters) into genes. This paper discusses the detailed implementation of Genetic Algorithms and hence the translation and implementation of TRIZ Principles to Genetic Algorithm structure to obtain generations of phenotypes (Images). A complete input or output image in Graphic Design is generally considered to consist of four main components: the background, the logo (branding), the text on the packaging and the extra effect (extra logo or design). TRIZ 'algorithms' are applied to these components individually and then these components are recombined. This is achieved by implementing these components as layers. GA's are inspired by nature and evolution. Each organism is considered to be made of cells which in turn consist of chromosomes; chromosomes are made of genes (which are functional blocks of DNA). Each gene controls a particular property of an aspect or behaviour or part of the organism. The different possibilities which a property can have are called 'Alleles'. Genes have a particular place in a chromosome (this is called 'locus'). This paper discusses the detailed implementation of Genetic Algorithms and hence the translation and implementation of TRIZ Principles to Genetic Algorithm structure to obtain generations of phenotypes (Images). The first GA enabled version of AEGIS discussed here has three chromosomes[l]. Outputs are shown applied to fonts for the chromosomes implemented. A brief comparison of the time taken to produce designs using AEGIS compared to professional packages points to AEGIS producing large time savings. The outputs from the software i.e., designs, are promising, and have to some extent motivated professional graphic designers to accept that this tool has benefits in aiding them to produce more innovative designs or at least speed up this process. Some 'design' examples using GA's are shown in this paper. In the future a further 14 chromosomes are being implemented in AEGIS.
展开▼
