Adjust

The Adjust module provides GPU-accelerated color and tone adjustments for images.

All adjust operations modify the image in-place.

Brightness

Adjusts image brightness by adding a factor to all RGB channels.

Example - Brighten:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.brightness(img, 0.2)

save(img, 'output.jpg')

Right-click to download and compare in full size

Example - Darken:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.brightness(img, -0.3)

save(img, 'output.jpg')

Right-click to download and compare in full size

Parameters:

image (Image): Image to adjust (modified in-place)
factor (float): Brightness adjustment. Positive values brighten, negative values darken

Use for: Quick exposure adjustments, creating fade effects

Contrast

Adjusts image contrast relative to middle gray (0.5).

Example - Increase Contrast:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.contrast(img, 1.5)

save(img, 'output.jpg')

Right-click to download and compare in full size

Example - Decrease Contrast:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.contrast(img, 0.5)

save(img, 'output.jpg')

Right-click to download and compare in full size

Parameters:

image (Image): Image to adjust (modified in-place)
factor (float): Contrast multiplier
factor > 1.0: Increases contrast
factor = 1.0: No change
factor < 1.0: Decreases contrast

Use for: Making images more punchy or creating washed-out effects

Saturation

Adjusts color intensity while preserving luminance.

Example - Increase Saturation:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.saturation(img, 1.8)

save(img, 'output.jpg')

Right-click to download and compare in full size

Example - Desaturate to Grayscale:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.saturation(img, 0.0)

save(img, 'output.jpg')

Right-click to download and compare in full size

Parameters:

image (Image): Image to adjust (modified in-place)
factor (float): Saturation multiplier
factor = 0.0: Grayscale
factor = 1.0: Original colors
factor > 1.0: More vibrant colors

Use for: Making colors pop, creating black and white images, subtle color grading

Gamma

Adjusts gamma correction (non-linear brightness). Unlike brightness, gamma affects midtones more than highlights or shadows.

Example - Brighten Midtones:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.gamma(img, 1.5)

save(img, 'output.jpg')

Right-click to download and compare in full size

Example - Darken Midtones:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.gamma(img, 0.6)

save(img, 'output.jpg')

Right-click to download and compare in full size

Parameters:

image (Image): Image to adjust (modified in-place)
gamma (float): Gamma value (must be positive)
gamma > 1.0: Brightens midtones
gamma = 1.0: No change
gamma < 1.0: Darkens midtones

Use for: Lifting shadows without blowing out highlights, display calibration, color grading

Hue

Shifts all colors around the color wheel by rotating hue values in HSV color space.

Example - Warm Sunset Tone:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.hue(img, 15)

save(img, 'adjust_hue_warm.png')

Right-click to download and compare in full size

Example - Cool/Cold Tone:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.hue(img, -30)

save(img, 'adjust_hue_cool.png')

Right-click to download and compare in full size

Example - Complementary Colors:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

# Shift to opposite colors on the color wheel
Adjust.hue(img, 180)

save(img, 'adjust_hue_180.png')

Right-click to download and compare in full size

Parameters:

image (Image): Image to adjust (modified in-place)
degrees (float): Hue rotation in degrees
degrees = 0: No change
degrees = ±30: Subtle warm/cool shift (color correction)
degrees = ±60-90: Noticeable color change
degrees = 120: Red → Green, Green → Blue, Blue → Red
degrees = 180: Complementary colors (opposite on color wheel)
degrees = 360: Full rotation (same as 0)

Use for:

Color correction: Fix white balance issues (±10-30°)
Creative effects: Transform entire color palette
Branding: Change product/logo colors globally
Season simulation: Summer → Autumn (shift towards orange/red)
Artistic filters: Psychedelic, vintage, or surreal color schemes

Note: Hue shift preserves brightness and saturation. Black, white, and gray pixels remain unchanged since they have no hue.

Vibrance

Smart saturation that protects already-saturated colors, especially skin tones.

Example - Portrait Enhancement:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.vibrance(img, 0.5)

save(img, 'adjust_vibrance_portrait.png')

Right-click to download and compare in full size

Example - Landscape:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.vibrance(img, 0.8)

save(img, 'adjust_vibrance_landscape.png')

Right-click to download and compare in full size

Parameters:

image (Image): Image to adjust (modified in-place)
amount (float): Vibrance adjustment (-1.0 to 1.0)
amount = -0.5: Subtle desaturation
amount = 0.0: No change
amount = 0.4: Recommended for portraits
amount = 0.8: Strong boost for landscapes

Use for: Enhancing colors without oversaturating skin tones, product photography, natural-looking color boosts

Saturation vs Vibrance

Saturation boosts all colors equally: - Red skin becomes orange - Already vibrant colors can clip

Vibrance is selective: - Protects already-saturated colors (skin tones) - Boosts muted colors more - Natural-looking results

Example comparison:

from pyimagecuda import load, Adjust, save

# Saturation: Can oversaturate skin
img1 = load("portrait.jpg")
Adjust.saturation(img1, 1.5)
save(img1, 'saturated.jpg')

# Vibrance: Protects skin tones
img2 = load("portrait.jpg")
Adjust.vibrance(img2, 0.5)
save(img2, 'vibrant.jpg')

Opacity

Adjusts image opacity by multiplying the alpha channel.

Example - Semi-Transparent:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.opacity(img, 0.5)

save(img, 'output.png')

Right-click to download and compare in full size

Example - Subtle Fade:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

Adjust.opacity(img, 0.8)

save(img, 'output.png')

Right-click to download and compare in full size

Parameters:

image (Image): Image to adjust (modified in-place)
factor (float): Opacity multiplier
factor = 0.0: Fully transparent
factor = 1.0: No change (original opacity)
factor > 1.0: Increases opacity (can exceed 1.0 for pixels with alpha < 1.0)

Use for: Creating fade effects, watermarks, layering with transparency, ghost/overlay effects

Note: This operation multiplies the existing alpha channel. If the image has no alpha channel or alpha is already 1.0, values > 1.0 will have no visible effect.

Combining Adjustments

All adjustments can be chained since they modify in-place:

Example:

from pyimagecuda import load, Adjust, save

img = load("photo.jpg")

# Apply multiple adjustments
Adjust.brightness(img, 0.1)
Adjust.contrast(img, 1.2)
Adjust.saturation(img, 1.3)

save(img, 'output.jpg')

Brightness vs Gamma

Brightness adds a constant value: - Affects all pixels equally - Can wash out highlights - Good for simple exposure fixes

Gamma applies a curve: - Affects midtones more than extremes - Preserves highlight and shadow detail - Better for lifting underexposed images

Example comparison:

from pyimagecuda import load, Adjust, save

# Brightness approach
img1 = load("dark_photo.jpg")
Adjust.brightness(img1, 0.3)
save(img1, 'brightened.jpg')

# Gamma approach (usually better)
img2 = load("dark_photo.jpg")
Adjust.gamma(img2, 1.5)
save(img2, 'gamma_corrected.jpg')