wcag_vision library

wcag_vision — an offline, algorithmic WCAG accessibility engine.

Modules:

  • Contrast — relative luminance, WCAG 2.x contrast ratios, alpha compositing for translucent colours, and AA/AAA conformance evaluation.
  • CVD simulation — colour-vision-deficiency simulation (protanopia, deuteranopia, tritanopia) using the Machado, Oliveira & Fairchild (2009) physiologically-based model.
  • Colour extraction — dominant-colour extraction via k-means clustering with deterministic k-means++ seeding, designed for one-shot analysis of a single captured frame.

Classes

ContrastReport
The result of evaluating a foreground/background colour pair against the WCAG 2.x contrast requirements.
ExtractedColor
One dominant colour found by k-means extraction, together with how much of the sampled image it accounts for.

Enums

CvdType
A type of colour vision deficiency (CVD) that can be simulated.
WcagConformanceLevel
A WCAG 2.x conformance level for text contrast.
WcagTextSize
The text-size category a contrast requirement applies to.

Constants

defaultClusterCount → const int
Default number of clusters (extractDominantColors's k).
defaultConvergenceThreshold → const double
Default centroid-movement convergence threshold, expressed as Euclidean distance in the RGB unit cube.
defaultMaxIterations → const int
Default cap on Lloyd iterations. Dominant-colour extraction on well-separated palettes converges in well under 10 iterations; 25 leaves generous headroom for noisy photographs without risking a long stall.
defaultMaxSamples → const int
Default maximum number of pixels sampled from the input.
maxContrastRatio → const double
The highest possible WCAG contrast ratio: 21.0, i.e. pure black on pure white (or vice versa).
minContrastRatio → const double
The lowest possible WCAG contrast ratio: 1.0, when both colours share the same relative luminance.

Functions

compositeOver(Color foreground, Color background) → Color
Alpha-composites foreground over background using the Porter-Duff "source-over" operator, returning the resulting blended colour.
contrastRatio(Color a, Color b) → double
Computes the WCAG 2.x contrast ratio between two opaque colours.
contrastRatioOver(Color foreground, Color background) → double
Computes the WCAG contrast ratio of a possibly semi-transparent foreground against an opaque background.
evaluateContrast(Color foreground, Color background) → ContrastReport
Evaluates the WCAG contrast between a foreground and background colour, returning a ContrastReport.
extractDominantColors(List<Color> pixels, {int k = defaultClusterCount, int seed = 0, int maxIterations = defaultMaxIterations, double convergenceThreshold = defaultConvergenceThreshold, int maxSamples = defaultMaxSamples}) → List<ExtractedColor>
Extracts the dominant colours of an image using k-means clustering.
extractDominantColorsAsync(List<Color> pixels, {int k = defaultClusterCount, int seed = 0, int maxIterations = defaultMaxIterations, double convergenceThreshold = defaultConvergenceThreshold, int maxSamples = defaultMaxSamples}) → Future<List<ExtractedColor>>
Extracts dominant colours off the main isolate — the entry point UI code should call.
relativeLuminance(Color color) → double
Computes the WCAG 2.x relative luminance of an opaque color.
simulateCvd(Color color, CvdType type) → Color
Simulates how color is perceived by a viewer with the given colour vision deficiency type, returning the simulated colour.
wcagThreshold(WcagConformanceLevel level, WcagTextSize size) → double
Returns the minimum contrast ratio required for the given level and size, per WCAG 2.x Success Criteria 1.4.3 (AA) and 1.4.6 (AAA).