dithering

func ErrorDifusionDithering ¶

func ErrorDifusionDithering(img image.Image, algorithm string, level int) image.Image

ErrorDiffusionDithering applies an error diffusion dithering effect to an image. This technique distributes the quantization error of a pixel to neighboring pixels according to different distribution patterns defined by various algorithms.

Supported algorithms:

• floyd-steinberg: Classic algorithm with good balance of quality and performance
• false-floyd-steinberg: Simplified version with fewer error terms
• jarvis-judice-ninke: Higher quality with wider error distribution (12 neighboring pixels)
• stucki: Modified Jarvis algorithm with improved weights
• atkinson: Partial error distribution that preserves detail (only distributes 3/4 of error)
• sierra: Good quality with moderate computational cost
• two-row-seirra: Two-row variant of Sierra algorithm with reduced complexity
• sierra-lite: Simplified one-row Sierra variant for faster processing
• none: No error diffusion applied (simple quantization)

Parameters:

• img: The input image to be processed
• algorithm: The name of the dithering algorithm to use (case-sensitive)
• level: The number of quantization levels per channel (1-10), where: Lower values (1-3) produce more posterized results with high contrast. Higher values (8-10) produce more subtle dithering with greater color depth.

Returns:

• image.Image: A new RGBA image with the dithering effect applied

func OrderedDithering ¶

func OrderedDithering(img image.Image, level, size int) image.Image

OrderedDithering applies an ordered dithering effect to an image using a Bayer matrix pattern. Ordered dithering creates a retro visual effect by systematically applying threshold patterns to reduce color depth while maintaining visual structure.

Parameters:

• img: The input image to be processed
• level: The number of quantization levels (1 - 20); higher values result in more color bands and less pronounced dithering effect. Values outside this range will be clamped.
• size: The size of the dithering matrix (must be a power of 2, e.g., 2, 4, 8); larger matrices create more complex dithering patterns. Non-power-of-2 values will be adjusted to the next even number.

Returns:

• A new image.Image with the ordered dithering effect applied, in RGBA64 format

Note: The alpha channel is also dithered by default. The function automatically handles bounds checking and matrix size adjustments.