Sunglasses Styling Optimization System Based on User Interactions
Considering sunglasses’ design features like large capacity, short period, quick modification, being difficult to accurately capture the user demand and so on, this paper achieved the optimal design of sunglasses form, and developed a prototype system based on interactive genetic algorithm, which realized the optimization mechanism from three aspects, as lens form coding, visualized population construction and users’ interactive evaluation model design. The algorithm program is developed based on three-dimensional design platform Solid works, and running as macro. The software extracts parameters from the user defined model and automatically generates new designs with the parameters, and then displays them for user’s evaluation which drives the optimization process to go circularly.
KeywordsSunglasses Optimization User interaction Interactive genetic algorithms
The paper is sponsored by Chinese National Natural Science Fund (No. 60975048), Zhejiang Natural Science Fund (No. Y1111111), Zhejiang Public Welfare Technology Research Project (No. 2010C31065) and Wenzhou Science & Technology Program (No. G20090098, G20100195). Many thanks to government’s generosity to our research work.
- 6.Gong DW, Hao GS (2007) Interactive genetic algorithms theory and applications, vol 77. National Defense Press, Beijing, pp 8–10Google Scholar
- 7.Biles JA, Anderson PG, Loggi LW (1996) Neural network fitness functions for a musical IGA. Int Symp Intell Ind Autom Soft Comput 90:39–44Google Scholar
- 8.Wang SF, Wang XF, Xue J (2005) An improved interactive genetic algorithm incorporating relevant feedback. In: Proceedings of 2005 international conference on machine learning and cybernetics, vol 9. Guangzhou, pp 2996–3001Google Scholar
- 9.Wang LH (2007) A comparison of three fitness prediction strategies for interactive genetic algorithms. J Inf Sci Eng 23(2):605–616Google Scholar
- 10.Sugimoto F, Yoneyama M (2002) Hybrid fitness assignment strategy in IGA: a method to compose fitness. In: Proceedings of the 2002 IEEE workshop on multimedia signal processing. ST Thomas, Virgin Islands, pp 284–287 Google Scholar
- 11.Rasheed K (2000) Informed operators: speeding up genetic-algorithm-based design optimization using reduced models. Proc Genet Evol Comput Conf 55:628–635Google Scholar
- 12.Abboud K, Schoenauer M (2002) Surrogate deterministic mutation. Artif Evol 99(1):103–115Google Scholar
- 14.Lee JH, Cho SB (2002) Analysis of direct manipulation in interactive evolutionary computation on fitness landscape. In: Proceedings of the 2002 congress on evolutionary computation, vol 87. Honolulu, pp 460–465Google Scholar