Colour Parametrization in a Multiparametric Image Interface

Multiple views of the same area are used in many applications such as remote sensing and medical diagnostics. To compare the same locus in different views, one can either: analyze views side-by-side, layer a plot with several dimensions using parameters of colour, or use highlighting and linking techniques. This latter approach is discussed and implemented in Quail (a QUantitative Analysis in common Lisp). Two colour systems, RGB (red, green, blue) and HSV (hue, saturation, lightness), are linked to human colour perception: the former to cone cell types, and the latter to light properties. RGB is used in colour monitors; HSV is intuitive to humans. This implementation converts HSV to RGB. Colour choice, in terms of HSV, is used here to represent image values and in highlighting image values. The choice of what colour parameter to use as the dimension is crucial; lightness and saturation are ordinal, and hue is categorical. This interface gives three choices to represent image values with: by lightness, by hue, or by a combination of the two. This latter choice has the advantage in that it is ordinal and provides a wider range of colours. 2 1