Image Processing

@cross/image provides a comprehensive set of chainable image processing operations. All methods modify the image in-place and return this, allowing you to chain multiple operations together.

Processing Categories

Filters

Apply various filters to enhance or modify images:

• Blur - Fast box blur for smoothing
• Gaussian Blur - High-quality blur with configurable sigma
• Median Filter - Strong noise reduction
• Sharpen - Enhance image details

Manipulation

Transform and modify image geometry:

• Resize - Scale images with bilinear or nearest neighbor interpolation
• Crop - Extract rectangular regions
• Composite - Layer images with alpha blending
• Fill Rectangle - Draw solid or semi-transparent rectangles

Color Adjustments

Adjust colors and tones:

• Brightness - Lighten or darken
• Contrast - Increase or decrease contrast
• Saturation - Adjust color intensity
• Exposure - Photographic exposure adjustment
• Grayscale - Convert to grayscale
• Invert - Invert all colors
• Sepia - Apply sepia tone

Chaining Operations

All processing methods return this, allowing elegant method chaining:

import { Image } from "@cross/image";

const data = await Deno.readFile("photo.jpg");
const image = await Image.decode(data);

// Chain multiple operations
image
  .resize({ width: 1920, height: 1080 })
  .brightness(0.1)
  .contrast(0.2)
  .gaussianBlur(1)
  .sharpen(0.5)
  .saturation(-0.2);

const output = await image.encode("png");
await Deno.writeFile("processed.png", output);

Performance Considerations

• Order Matters: Apply resize operations early to process fewer pixels
• Filter Radius: Larger radii increase processing time significantly
• Method Selection: Nearest neighbor resize is faster but lower quality than bilinear

Optimal Processing Order

image
  .resize({ width: 800, height: 600 }) // Reduce pixels first
  .brightness(0.1) // Fast color adjustments
  .contrast(0.2)
  .gaussianBlur(2) // Expensive filters last
  .sharpen(0.5);