3) The following regression equation was obtained by the use of multiple regression analysis for (F1) visual appeal, (F2) practicality, (F3) preference, in the case of brand yukata, (F1) visual appeal = 0.506 + 0.273 ×(F2) practicality + 0.230 ×(F3) preference. In the case of (F3) preference, it was found that they rated brand yukata significantly higher than classic yukata, classic kimono and modern kimono. In the case of (F2) practicality, it was found that they rated classic kimono significantly higher than brand yukata, classic kimono, and modern kimono. 2) As a result of comparing the mean factor score of classic kimono, modern kimono, classic yukata, and brand yukata, in the case of (F1) visual appeal, it was found that high school girls rated modern kimono and brand yukata significantly higher than classic kimono and classic yukata. The results were obtained by factor analysis are and the multiple regression analysis as follows: 1) The factor structure of high school girl's opinion on image for kimono and yukata are interpreted as (F1) visual appeal, (F2) practicality and (F3) preference. In this paper, high school girl's opinion on image for kimono (modern kimono and classic kimono) and yukata (brand yukata and classic yukata) were studied. KeywordsInteractive Genetic Algorithm–Fitness modeling–Customer preference–Color Restrictions are coded by proposed mutation and crossover operators. The number of users evaluations is achieved with proposed combination and speedup in IGA evolution is achieved when color Experimental results show that 80% reduction in Takes the case of the color design of a building for illustration purposes.
Two functions are proposed for coding color restrictions. Search space for the IGA is very large, proposed system reduces it creating genotypes with colors restrictions developed withĪ new graph coloring technique. Neural Networks to model the fitness function and find the optimally colored image according to user’s preference. This paper proposes the combination of Interactive Genetic Algorithm (IGA) with Artificial Experiments are performed in order to confirm the usefulness of the proposed system.Ĭustomers preferences are difficult to predict automatically because they are subjective and relative to human likeness patterns The modification procedures are repeated until a use is satisfied with presented fashion design and coordinate. In the modification part, based on user's evaluation of presented design and coordinate, the target impression value is updated on impressions spaces and parameter values of modified fashion design and coordinate are calculated. Based on the results of the impression estimation part, the parameter values of shirts and outerwear, and the color scheme are determined, and the initial candidates of fashion design and coordinate are presented to a user. In the impression estimation part, impression of an inputted adjective, i.e., an impression value, is estimated using impressions spaces of clothes shape and color schemes. This system is composed of three main parts: An impression estimation part, a shirts and outerwear design and coordinate part, and a modification part based on user's evaluation.
Fashion coordinate consists of a shirt form, an outerwear form, and their color schemes. This paper proposes a fashion design and coordinate generation system that takes account of impression of clothing and user's KANSEI. The evaluation experiments are performed for the evaluation of the method, and the results show that the coordinate method is applicable. The last one is to propose the generation method of initial coordinates candidates. In order to obtain knowledge on the relation between combination and impressions, four types of combinations are considered based on the impressions space, and the evaluation experiments are performed for evaluation of the combinations. The second one is the analysis of impressions of the combinations of outerwear and a shirt. The evaluation experiments are performed for the evaluation of clothes samples selected based on the impressions space. Through the pre-experiments, two impression factors are extracted and an impressions space consisting of two factors axes is constructed. The first one is the analysis of impressions of clothes in order to get knowledge of fashion coordinates. For the construction of the coordinates system, three items are discussed.
This paper proposes the fashion coordinates generation system reflecting impressions expressed by an image word.
0 kommentar(er)
