Classification and Analysis of Color Impression Using Genetic Algorithms Considering Individual Characteristics

Abstract

It has often been pointed out that each color has a particular impression and that there are certain similarities and individual differences in the relationships between a color and its impression. We now present a few examples of individual studies that have focused on color impression. In this paper, in order to examine the individual characteristics of color impressions, we propose a method that combines GA, PCA, and CA. GA is used to select suitable parameters from color attributes -namely, RGB, HSI, and L*a*b*. PCA is used to extract characteristics data from the selected parameters. Furthermore, CA is used to classify the data set into groups. In order to demonstrate the effectiveness of this proposed method, computer simulations are performed and the results are analyzed using data gathered from 6 individuals. As a result, characteristics data that were obtained by using GA differed for each subjects. Moreover, classification accuracy, which is the ratio classified same impression data into same groups, was more than 80% for all subjects. It is considered that the proposed method could be one of the effective methods for analyzing individual characteristics of color impression.