Transforming Images#

Image Processing with PIL (Pillow)#

Python Tutorial: Image Manipulation with Pillow (16 min)

Pillow is Python’s leading library for image processing, offering powerful tools for editing, analyzing, and generating images. It handles various formats (PNG, JPEG, GIF, etc.) and provides operations from basic resizing to complex filters.

Here’s a minimal example showing common operations:

# /// script
# requires-python = ">=3.11"
# dependencies = ["Pillow"]
# ///

from PIL import Image, ImageEnhance, ImageFilter


async def process_image(path: str) -> Image.Image:
    """Process an image with basic enhancements."""
    with Image.open(path) as img:
        # Convert to RGB to ensure compatibility
        img = img.convert("RGB")

        # Resize maintaining aspect ratio
        img.thumbnail((800, 800))

        # Apply enhancements
        img = ImageEnhance.Contrast(img).enhance(1.2)

        return img.filter(ImageFilter.SHARPEN)


if __name__ == "__main__":
    import asyncio

    img = asyncio.run(process_image("input.jpg"))
    img.save("output.jpg", quality=85)

Key features and techniques you’ll learn:

  • Image Loading and Saving: Handle various formats with automatic conversion
  • Basic Operations: Resize, rotate, crop, and flip images
  • Color Manipulation: Adjust brightness, contrast, and color balance
  • Filters and Effects: Apply blur, sharpen, and other visual effects
  • Drawing: Add text, shapes, and overlays to images
  • Batch Processing: Handle multiple images efficiently
  • Memory Management: Process large images without memory issues

Basic Image Operations#

Common operations for resizing, cropping, and rotating images:

from PIL import Image


async def transform_image(path: str, size: tuple[int, int], rotation: float = 0) -> Image.Image:
    """Transform image with basic operations."""
    with Image.open(path) as img:
        # Resize with anti-aliasing
        img = img.resize(size, Image.LANCZOS)

        # Rotate around center
        if rotation:
            img = img.rotate(rotation, expand=True)

        # Auto-crop empty edges
        img = img.crop(img.getbbox())

        return img

Color and Enhancement#

Adjust image appearance with built-in enhancement tools:

from PIL import ImageEnhance, ImageOps


async def enhance_image(
    img: Image.Image, brightness: float = 1.0, contrast: float = 1.0, saturation: float = 1.0
) -> Image.Image:
    """Apply color enhancements to image."""
    enhancers = [
        (ImageEnhance.Brightness, brightness),
        (ImageEnhance.Contrast, contrast),
        (ImageEnhance.Color, saturation),
    ]

    for Enhancer, factor in enhancers:
        if factor != 1.0:
            img = Enhancer(img).enhance(factor)

    return img

Filters and Effects#

Apply visual effects and filters to images:

from PIL import ImageFilter


def apply_effects(img: Image.Image) -> Image.Image:
    """Apply various filters and effects."""
    effects = {
        "blur": ImageFilter.GaussianBlur(radius=2),
        "sharpen": ImageFilter.SHARPEN,
        "edge": ImageFilter.FIND_EDGES,
        "emboss": ImageFilter.EMBOSS,
    }

    return {name: img.filter(effect) for name, effect in effects.items()}

Drawing and Text#

Add text, shapes, and overlays to images:

from PIL import Image, ImageDraw, ImageFont


async def add_watermark(img: Image.Image, text: str, font_size: int = 30) -> Image.Image:
    """Add text watermark to image."""
    draw = ImageDraw.Draw(img)
    font = ImageFont.truetype("arial.ttf", font_size)

    # Calculate text size and position
    text_bbox = draw.textbbox((0, 0), text, font=font)
    text_width = text_bbox[2] - text_bbox[0]
    text_height = text_bbox[3] - text_bbox[1]

    # Position text at bottom-right
    x = img.width - text_width - 10
    y = img.height - text_height - 10

    # Add text with shadow
    draw.text((x + 2, y + 2), text, font=font, fill="black")
    draw.text((x, y), text, font=font, fill="white")

    return img

Memory-Efficient Processing#

Handle large images without loading them entirely into memory:

from PIL import Image
import os


async def process_large_images(input_dir: str, output_dir: str, max_size: tuple[int, int]) -> None:
    """Process multiple large images efficiently."""
    os.makedirs(output_dir, exist_ok=True)

    for filename in os.listdir(input_dir):
        if not filename.lower().endswith((".png", ".jpg", ".jpeg")):
            continue

        input_path = os.path.join(input_dir, filename)
        output_path = os.path.join(output_dir, filename)

        with Image.open(input_path) as img:
            # Process in chunks using thumbnail
            img.thumbnail(max_size)
            img.save(output_path, optimize=True)

Practice with these resources:

Watch these tutorials for hands-on demonstrations:

Image Processing Tutorial for beginners with Python PIL in 30 mins

Image Processing with ImageMagick#

ImageMagick is a powerful command-line tool for image manipulation, offering features beyond what’s possible with Python libraries. It’s particularly useful for:

  • Batch processing large image collections
  • Complex image transformations
  • High-quality format conversion
  • Creating image thumbnails
  • Adding text and watermarks

Basic Operations:

# Format conversion
convert input.png output.jpg

# Resize image (maintains aspect ratio)
convert input.jpg -resize 800x600 output.jpg

# Compress image quality
convert input.jpg -quality 85 output.jpg

# Rotate image
convert input.jpg -rotate 90 output.jpg

Common Data Science Tasks:

# Create thumbnails for dataset preview
convert input.jpg -thumbnail 200x200^ -gravity center -extent 200x200 thumb.jpg

# Normalize image for ML training
convert input.jpg -normalize -strip output.jpg

# Extract dominant colors
convert input.jpg -colors 5 -unique-colors txt:

# Generate image statistics
identify -verbose input.jpg | grep -E "Mean|Standard|Kurtosis"

Batch Processing:

# Convert all images in a directory
mogrify -format jpg *.png

# Resize multiple images
mogrify -resize 800x600 -path output/ *.jpg

# Add watermark to images
for f in *.jpg; do
    convert "$f" -gravity southeast -draw "text 10,10 'Copyright'" "watermarked/$f"
done

Advanced Features:

# Apply image effects
convert input.jpg -blur 0x3 blurred.jpg
convert input.jpg -sharpen 0x3 sharp.jpg
convert input.jpg -edge 1 edges.jpg

# Create image montage
montage *.jpg -geometry 200x200+2+2 montage.jpg

# Extract image channels
convert input.jpg -separate channels_%d.jpg

# Composite images
composite overlay.png -gravity center base.jpg output.jpg

Watch this ImageMagick tutorial (16 min):

ImageMagick Introduction (16 min)

Tools:

Tips:

  1. Use -strip to remove metadata and reduce file size
  2. Monitor memory usage with -limit memory 1GB
  3. Use -define for format-specific options
  4. Process in parallel with -parallel
  5. Use -monitor to track progress

Error Handling:

# Check image validity
identify -regard-warnings input.jpg

# Get detailed error information
convert input.jpg output.jpg 2>&1 | grep -i "error"

# Set resource limits
convert -limit memory 1GB -limit map 2GB input.jpg output.jpg

For Python integration:

# /// script
# requires-python = ">=3.9"
# dependencies = ["Wand"]
# ///

from wand.image import Image


async def process_image(path: str) -> None:
    """Process image with ImageMagick via Wand."""
    with Image(filename=path) as img:
        # Basic operations
        img.resize(800, 600)
        img.normalize()

        # Apply effects
        img.sharpen(radius=0, sigma=3)

        # Save with compression
        img.save(filename="output.jpg")

Note: Always install ImageMagick before using the Wand Python binding.

Backward Dbt Extracting Audio and Transcripts Forward