本論文旨在探討如何蒐集具有中華文化特質的色彩資料庫,藉此建立一漢風色彩資訊系統,以供設計師在其創作過程中,對色彩的選擇有一更寬廣的工作環境。我們利用我國歷史文物、民間工藝、建築、服飾等融入生活中常見的中國風色彩,以國人直覺的方式就地採集,利用攝影機、掃瞄器等設備予以取樣,再藉由SOM類神經網路功能把漢風色彩群組化。群組化的結果將於電腦上繪出3D的特徵映對圖,便予觀察漢風色彩的分佈狀況,且經色彩專家加以整理,才案定為漢風色彩。同時,我們也運用電腦的RGB值與CIE XYZ、Lab的轉換技術,對應出漢風色碼及其他如Munsell、NCS等色系資料。最後,採用概念設計模式的方法建立一多媒體資料庫來儲存所搜集到的漢風色彩資料,並以資建立一人性化的漢風色彩搜尋環境。
The objectives of this paper is to develop a Sino-color dictionary system by using computer technology. We investigate Sino-colors by using Chinese intuition from different subjects, such as Chinese architecture, clothing, opera, custom, painting, etc. These Sino-color samples are stored into computers by using a scanner or digital camera. The SOM neural network model is used to classify these colors into different clusters. According to these clusters' information, a 3D Sino-color feature map will, then, be drawn on a computer. Finally, color experts will define the representative Sino-colors through the distribution appearing in the resulting feature map. The notation translation between RGB and other color systems, such as Sino-color notation, CIELab and Munsell, is required since most computer systems represent color using RGB model. After a set of Sino-colors being determinated, we develop a multi-medium database to store these knowledge, along with other richful image information. The system includes a friendly user interface which allows users to retrieve the information by using color chips, color notations, color names, color images, and utilization cases. We hope the system can assist designers in creating their products and improving their products quality.