colormap

A colormap, or color map, is a function that maps scalar values to colors for visualization. It converts a data range into a sequence of colors that can be displayed as images, heatmaps, or textures.

Colormaps are typically categorized as sequential, divergent, or qualitative. Sequential palettes progress through a single hue from light to dark to encode increasing magnitude. Divergent palettes use two contrasting hues diverging from a neutral mid-point, emphasizing deviations around a reference value. Qualitative palettes assign distinct colors to discrete categories without implying ordering. Perceptual criteria are increasingly emphasized: perceptually uniform colormaps aim for constant perceived change per unit of data, and many designers prefer palettes built in perceptual color spaces such as CIELAB or CAM02-UCS and then converted to RGB.

Implementation usually involves a lookup table or a color interpolation scheme. A colormap can be defined as a list of anchor colors; intermediate values are produced by interpolating between anchors. Colormaps are frequently used in heatmaps, volume rendering, MRI and CT visualization, meteorology, and geographic information systems. In software, colormaps are often referred to as lookup tables (LUTs) that map scalar intensities to RGB triplets.

Accessibility considerations include avoiding color contrasts that are indistinguishable to colorblind viewers; choosing palettes with monotonic luminance; and providing alternative representations such as grayscale or annotated scales.

Examples of well-known colormaps include viridis, plasma, and inferno for perceptual uniformity, and diverging palettes like 'coolwarm' or 'RdBu'; in software libraries such as matplotlib and MATLAB these palettes are readily available.