In a previous investigation it was found that the perceptual quality of images of natural scenes depended upon their subjective colorfulness when the images were transformed by varying of chroma.1 The transformations were performed over the color point distributions in the CIELUV perceptually uniform color space where each color point represented one pixel of a corresponding image. The experiments with uni-dimensional scaling of colorfulness have shown that the colorfulness can be represented as a linear combination of average chroma and variability of chroma. However, such a representation held only within each scene: there was no regular rule to compare different scenes even with significantly different average chroma of the original images. The results of multidimensional scaling of colorfulness reveal the multidimensional structure of this attribute. Possible interpretations of subjective dimensions are discussed.
Elena Fedorovskaya, Huib de Ridder, Sergej Yendrikhovskij, Frans Blommaert, "Multidimensional Structure of Colorfulness: Chroma Variation in Color Images of Natural Scenes" in Proc. IS&T 3rd Color and Imaging Conf., 1995, pp 130 - 133, https://doi.org/10.2352/CIC.1995.3.1.art00034