Research And Application Of Flower Pattern Design Based On Genetic Algorithm

With the high-speed development of computer-aided design technology, people not only use computer as a visual tool for the product design, but begin to apply machine intelligence to the product conceptual design which is the most important and primary process . This technique could be able to help designers to give their inspiration and imagination to create product with more originality and improvement. The method of creative design based on object evolutionary is a CAD technology with the application of evolutionary computation that belongs to the field of artificial intelligence. To futher satisfy the client, the creative design can help improving the appearance, qulity, performance and so on. As the theory of evolutionary computation grows up, this method has been one of most important technique in the CAD field. Thus, creative design based on genetic algorithm which belongs to the evolutionary computation has been a popular research spot.Genetic Algorithms (GA) is a adaptive heuristic search algorithm. The basic concept of GA is designed to simulate processes in natural system necessary for evolution, specifically those that follow the principles first laid down by Charles Darwin of survival of the fittest. With research for many years, it has been a mature optimization and search algorithm applied to solve many engineering problems such as function optimization, automation control, manufacturing scheduling and so on. As a method of evolutionary computation, the application in the field of computer aided creative design had been proved successful and the product it creats is amazing especially in the architecture design. In the future, this method will have more and more extensive application in the relevant fields.Pattern design is the one of most important topics in the sphere of art design. Because of its wide and significant applications, flower pattern design which is a subset of pattern design is the highly attractable topics through all. With the characteristics of high adaptability and variety, flower pattern can be found in almost all of art decoration. Through the analysis of the process of flower pattern design, this paper apply genetic algorithm to the creation of the flower pattern design, with which the designers could gained more inspiration and enhance their working efficiency. Meanwhile, the application of curve fitting to the design process can make better use of the materials from the resources. The main work is as follows:1. Apply the genetic algorithm to the flower pattern design based on the part curves.In the view of variety of part curves and symmetry of the entire pattern, a method of creation of the flower pattern based on the part curves is proposed. With the coding of the initial control points of part curves, a number of pattern design works can be produced using the genetic operation. And these generated patterns have different part and similitude full shape. Correspond coding scheme, generation of initial population, parameters, genetic operations and constraints are constructed to improve and fit the algorithm. The instances show that this method can effectively help designers to create more various designings.2. Improve the Genetic algorithm-based fitting Bézier curves and apply it to the flower pattern interactive design process.An improved adaptive genetic algorithm (AGA) is applied in the least squares curve fitting. The crossover probability and mutation probability are adjusted adaptively through the progress so that the convergence speed in the progress of curve fitting is increased. Besides, in the process of interactive design, the initial population can be constrained refer to the data point using the curve tools. Thus the fitting speed can be further enhanced. With this method apply to the flower pattern design, people could make best of use of the design resources.3. Design and implement a flower pattern design system based on genetic algorithm.The system is designed and developed on the platform of WindowsXP with VC++ 6.0 and utilized the GDI+ two-dimension graphics class library. The main modules of the system are realized. Finally, the analysis of instances shows that the design results are quite satisfying...