質感圖形的空間結構特徵與其混色效果之關係

Abstract

人類的色彩視覺系統可辨識的色彩種類可達兩百萬種（Gouras，1991），拜色彩視覺遵循三色論之賜，吾人只需慎選三種基本色，即可混合出大部分可辨識範圍之內的色彩。色彩工業之所以能複製出大量色彩種類，即依循三色論以及加法混色的原理，十九世紀的點描派繪畫以及現代印刷業與顯示器產業，皆是以有限色彩種類輔以密集排列的色點混和出豐富的色彩。本研究的主要目的在於探討色點的排列方式對於主觀混色效果的影響，我們推測同樣數量比例的色點以不同的分布方式呈現，有可能造成不同的混色效果，因此設計多種紅綠面積比例各半、圖樣不一的質感圖形，與另一緊鄰的均質色彩平面進行色彩比對（color matching）作業，受試者可調整均質平面的紅綠比例以及明暗，調至與相鄰質感的，以此定量出混色的主觀效果。實驗結果有兩個主要發現：（1）不同樣式的質感圖形會產生不同的色相與明暗結果，這現象在只可調整色相的實驗二當中為明顯；（2）所有受試者對所有實驗刺激的主觀紅綠比例判斷都介於1.5：1到2.2：1之間，顯示出紅綠交錯的質感圖形當中，紅色具有明顯的知覺優勢。我們的結果除了突顯色點排列方式對混色效果的重要性之外，亦顯示色彩除了機基本的色相、明度、彩度之外，還可能包括「醒目性」這個特徵。

人類的色彩視覺系統可辨識的色彩種類可達兩百萬種（Gouras，1991），拜色彩視覺遵循三色論之賜，吾人只需慎選三種基本色，即可混合出大部分可辨識範圍之內的色彩。色彩工業之所以能複製出大量色彩種類，即依循三色論以及加法混色的原理，十九世紀的點描派繪畫以及現代印刷業與顯示器產業，皆是以有限色彩種類輔以密集排列的色點混和出豐富的色彩。本研究的主要目的在於探討色點的排列方式對於主觀混色效果的影響，我們推測同樣數量比例的色點以不同的分布方式呈現，有可能造成不同的混色效果，因此設計多種紅綠面積比例各半、圖樣不一的質感圖形，與另一緊鄰的均質色彩平面進行色彩比對（color matching）作業，受試者可調整均質平面的紅綠比例以及明暗，調至與相鄰質感的，以此定量出混色的主觀效果。實驗結果有兩個主要發現：（1）不同樣式的質感圖形會產生不同的色相與明暗結果，這現象在只可調整色相的實驗二當中為明顯；（2）所有受試者對所有實驗刺激的主觀紅綠比例判斷都介於1.5：1到2.2：1之間，顯示出紅綠交錯的質感圖形當中，紅色具有明顯的知覺優勢。我們的結果除了突顯色點排列方式對混色效果的重要性之外，亦顯示色彩除了機基本的色相、明度、彩度之外，還可能包括「醒目性」這個特徵。 關鍵字：加法混色；點描派繪畫；色彩工業；色彩比對；質感；醒目性 ABSTRACT Human color vision can discriminate about 2 million different colors (Gouras, 1991). However, thanks to the trichomatic nature of our color vision, only three well-chosen colors are needed to match any given color. This very fact is the foundation of the modern color display industry, in which three primal colors are used to create a full color gamut. Historically, from the Pointillism painting in 19th century to modern color printing and display devices, the effect of additive color mixture has often been achieved by means of the appropriate placing of small color dots or pixels. Little attention has been paid to the possible effect by the spatial arrangement of the dots on the appearance of the resultant colors. Our study aims at probing into the correlation between spatial arrangement of color dots and the resultant color mixture perception. Textures made of equal amount of red and green dots were created as the test stimuli. Subjects’ perception of the mixture color was then measured with a color-matching paradigm. Subjects were instructed to adjust the intensity and the red/green ratio of a homogeneous color field until its color appearance matched to that of the textured field. Our results show that: (1) Different texture patterns seem to elicit different red/green ratio values, particularly when the intensity of the homogeneous field is locked at a fixed value. (2) All subject gave red/green ratio a value greater than 1.0 (ranging from 1.5:1 to 2.2:1) for all texture patterns, which indicates that red is perceptually more dominant than green in a red/green texture. (3) There is a close interdependent relationship between the settings of red/green ratio and the intensity. When a color patch is turned up in its intensity, its red/green balance needs to be turned towards red, and vice versa. Over all, our results show that the spatial structure of a texture does affect its color appearance. Furthermore, our data also imply that there might exist a basic “saliency” dimension in addition to brightness, hue, and saturation for coding colors.